Ultralytics
guides/parking-management/ #11093
giscus[bot]
bot
Replies: 35 comments 66 replies
-
|
Hi , when i run the first script i get this error: File "C:\Users\explo\miniconda3\Lib\site-packages\ultralytics\solutions\parking_management.py", line 71, in upload_image How can i solve? |
Beta Was this translation helpful? Give feedback.
-
|
Hello! I suggest to fix the uplad_image function like this: And i tested the code with this video But the model is not recognizing the cars (cup, microwave ecc instead) Do you have some videos template to test the code? |
Beta Was this translation helpful? Give feedback.
-
|
@eXplosive57 hello! Thanks for sharing your updated Regarding the issue of the model not recognizing cars correctly and identifying other objects like cups or microwaves instead, it sounds like it might be a case of model misclassification, which can be due to a variety of factors:
For testing purposes, you can use standard traffic or parking lot videos which are often used in object detection benchmarks, or you can capture your own dataset from typical parking areas. Here is an example of initializing and using the parking management model: import cv2
from ultralytics.solutions.parking_management import ParkingManagement
# Ensure you have the correct model path and polygon JSON
management = ParkingManagement(model_path="yolov8n.pt")
polygon_json_path = "path/to/your/bounding_boxes.json"
cap = cv2.VideoCapture("path/to/your/video.mp4")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].and results[0].boxes:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
cap.release()
cv2.destroyAllWindows()This setup should help you streamline the detection process. If the issue persists, try fine-tuning the detection model or adjusting the confidence and IOU thresholds. Let me know how it goes or if you need further assistance! |
Beta Was this translation helpful? Give feedback.
-
|
Thank you, @eXplosive57, for providing the prompt solution! I've been able to replicate the issue and will address it promptly. Regarding the model, training it on the visdrone dataset could enhance its ability to detect objects in drone or aerial views. Thanks VisDrone Dataset Docs: https://docs.ultralytics.com/datasets/detect/visdrone/ |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for your answers! @RizwanMunawar @pderrenger |
Beta Was this translation helpful? Give feedback.
-
|
@eXplosive57 This issue has been resolved! You can utilize the parking management system by installing the latest ultralytics PYPI package using the command mentioned below! Thanks! |
Beta Was this translation helpful? Give feedback.
-
|
Hello, I use the iOS system, when using the following code to meet the data standards, the image can not be loaded, pngd,jpg,jpeg can not be loaded, help me to answer, thank you |
Beta Was this translation helpful? Give feedback.
-
|
@Gxiaodiao hello, Thank you for reaching out with your issue regarding image loading in the iOS system using the To assist you better, could you please provide the following details:
This information will help us understand the problem more clearly and provide a more accurate solution. Meanwhile, ensure that the image paths are correct and accessible from your application. Looking forward to your response! |
Beta Was this translation helpful? Give feedback.
-
|
Can it support camera from the side or angle instead from top view? |
Beta Was this translation helpful? Give feedback.
-
|
Yes, Ultralytics YOLOv8 can support camera angles from the side or at an angle, not just from a top view. The model's robust detection capabilities allow it to effectively identify vehicles in various orientations and angles, which can be particularly useful in dynamic parking environments. Adjusting the detection thresholds and fine-tuning the model with angle-specific training data can further enhance accuracy for side or angled views. If you have any more questions or need further assistance, feel free to ask! 🚗👍 |
Beta Was this translation helpful? Give feedback.
-
|
Hi Paula,
Thanks for your reply, at this moment I'm using a licensed subscription
platform to detect the vehicle presence at the parking bays.
I'd expect YOLOv8 could offer an open source solution which could reduce
the license cost.
Please provide an Example Workflow using YOLOv8 based on the use of a solar
powered camera using ftp function to send parking bay images to a ftp
server.
Do reply to me soon, thank you.
Adi Y Rosl
…On Fri, Jun 7, 2024 at 9:13 PM Paula Derrenger ***@***.***> wrote:
Yes, Ultralytics YOLOv8 can support camera angles from the side or at an
angle, not just from a top view. The model's robust detection capabilities
allow it to effectively identify vehicles in various orientations and
angles, which can be particularly useful in dynamic parking environments.
Adjusting the detection thresholds and fine-tuning the model with
angle-specific training data can further enhance accuracy for side or
angled views. If you have any more questions or need further assistance,
feel free to ask! 🚗👍
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQXHTL6NTQV7HOZOAWLZGGWXFAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMBSGIZTO>
.
You are receiving this because you commented.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Hi Adi, Thank you for reaching out! It's great to hear that you're considering YOLOv8 for your parking management needs. YOLOv8 can indeed provide a cost-effective, open-source solution for vehicle detection in parking bays. Below, I'll walk you through an example workflow to set up a parking management system using a solar-powered camera that sends images to an FTP server. Example Workflow for Parking Management with YOLOv81. Setting Up the FTP ServerEnsure your solar-powered camera is configured to send images to your FTP server. You'll need the FTP server details (hostname, username, password, and the directory where images are stored). 2. Downloading Images from FTP ServerUse Python's import ftplib
import os
ftp = ftplib.FTP('ftp.yourserver.com')
ftp.login('username', 'password')
ftp.cwd('/path/to/images')
filenames = ftp.nlst() # List of filenames in the directory
for filename in filenames:
local_filename = os.path.join('/local/path/to/save/images', filename)
with open(local_filename, 'wb') as file:
ftp.retrbinary('RETR ' + filename, file.write)
ftp.quit()3. Selecting Parking PointsUse the from ultralytics import solutions
solutions.ParkingPtsSelection()This will allow you to create polygons around parking spaces and save the configuration as a JSON file. 4. Parking Management CodeHere's a complete example of how to set up the parking management system using YOLOv8: import cv2
from ultralytics import solutions
# Path to the JSON file created with the point selection tool
polygon_json_path = "bounding_boxes.json"
# Initialize video capture from the local directory where images are saved
cap = cv2.VideoCapture("/local/path/to/save/images/video.mp4")
assert cap.isOpened(), "Error reading video file"
w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writer to save the output
video_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management object
management = solutions.ParkingManagement(model_path="yolov8n.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()
video_writer.release()
cv2.destroyAllWindows()Optional Arguments for CustomizationYou can customize the appearance and behavior of the parking management system using optional arguments: management = solutions.ParkingManagement(
model_path="yolov8n.pt",
txt_color=(0, 0, 0),
bg_color=(255, 255, 255),
occupied_region_color=(0, 255, 0),
available_region_color=(0, 0, 255),
margin=10
)This setup should help you get started with a cost-effective and efficient parking management system using YOLOv8. If you have any further questions or need additional assistance, feel free to ask! 🚗👍 |
Beta Was this translation helpful? Give feedback.
-
|
Hi,
I've deployed the Yolov8 to test it and I've got the error message when
verifying the installation as follows.
yolo task=detect mode=predict model=yolov8n.pt source='
https://ultralytics.com/images/zidane.jpg'
Traceback (most recent call last):
File "/yolov8env/bin/yolo", line 5, in <module>
from ultralytics.cfg import entrypoint
File "/yolov8env/lib/python3.12/site-packages/ultralytics/__init__.py",
line 10, in <module>
from ultralytics.data.explorer.explorer import Explorer
File
"/yolov8env/lib/python3.12/site-packages/ultralytics/data/__init__.py",
line 3, in <module>
from .base import BaseDataset
File "/yolov8env/lib/python3.12/site-packages/ultralytics/data/base.py",
line 12, in <module>
import cv2
File "/yolov8env/lib/python3.12/site-packages/cv2/__init__.py", line 181,
in <module>
bootstrap()
File "/yolov8env/lib/python3.12/site-packages/cv2/__init__.py", line 153,
in bootstrap
native_module = importlib.import_module("cv2")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/importlib/__init__.py", line 90, in
import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ImportError: libGL.so.1: cannot open shared object file: No such file or
directory
I did try to search and install libGL.so.1 but cannot find it,
Do reply to me soon, thanks.
…On Sat, Jun 8, 2024 at 4:26 AM Glenn Jocher ***@***.***> wrote:
Hi Adi,
Thank you for reaching out! It's great to hear that you're considering
YOLOv8 for your parking management needs. YOLOv8 can indeed provide a
cost-effective, open-source solution for vehicle detection in parking bays.
Below, I'll walk you through an example workflow to set up a parking
management system using a solar-powered camera that sends images to an FTP
server.
Example Workflow for Parking Management with YOLOv8 1. *Setting Up the
FTP Server*
Ensure your solar-powered camera is configured to send images to your FTP
server. You'll need the FTP server details (hostname, username, password,
and the directory where images are stored).
2. *Downloading Images from FTP Server*
Use Python's ftplib library to download images from the FTP server.
import ftplibimport os
ftp = ftplib.FTP('ftp.yourserver.com')ftp.login('username', 'password')ftp.cwd('/path/to/images')
filenames = ftp.nlst() # List of filenames in the directory
for filename in filenames:
local_filename = os.path.join('/local/path/to/save/images', filename)
with open(local_filename, 'wb') as file:
ftp.retrbinary('RETR ' + filename, file.write)
ftp.quit()
3. *Selecting Parking Points*
Use the ParkingPtsSelection tool to define parking areas.
from ultralytics import solutions
solutions.ParkingPtsSelection()
This will allow you to create polygons around parking spaces and save the
configuration as a JSON file.
4. *Parking Management Code*
Here's a complete example of how to set up the parking management system
using YOLOv8:
import cv2from ultralytics import solutions
# Path to the JSON file created with the point selection toolpolygon_json_path = "bounding_boxes.json"
# Initialize video capture from the local directory where images are savedcap = cv2.VideoCapture("/local/path/to/save/images/video.mp4")assert cap.isOpened(), "Error reading video file"w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writer to save the outputvideo_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()video_writer.release()cv2.destroyAllWindows()
Optional Arguments for Customization
You can customize the appearance and behavior of the parking management
system using optional arguments:
management = solutions.ParkingManagement(
model_path="yolov8n.pt",
txt_color=(0, 0, 0),
bg_color=(255, 255, 255),
occupied_region_color=(0, 255, 0),
available_region_color=(0, 0, 255),
margin=10
)
This setup should help you get started with a cost-effective and efficient
parking management system using YOLOv8. If you have any further questions
or need additional assistance, feel free to ask! 🚗👍
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQWTFLZQL5J2E6MIANDZGIJQXAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMBXGE2DO>
.
You are receiving this because you commented.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Hi there, Thank you for reaching out! It looks like you're encountering an issue with the To resolve this issue, you can install the missing library using the following command, depending on your operating system: For Ubuntu/Debian: sudo apt-get update
sudo apt-get install -y libgl1-mesa-glxFor CentOS/RHEL: sudo yum install -y mesa-libGLFor macOS, you can use Homebrew: brew install libglAfter installing the necessary library, try running your YOLOv8 command again: yolo task=detect mode=predict model=yolov8n.pt source='https://ultralytics.com/images/zidane.jpg'This should resolve the |
Beta Was this translation helpful? Give feedback.
-
|
I got this message.
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Package libgl1-mesa-glx is not available, but is referred to by another
package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source
E: Package 'libgl1-mesa-glx' has no installation candidate
I'm using Ubuntu 24.04 LTS
…On Sat, Jun 8, 2024 at 7:43 PM Glenn Jocher ***@***.***> wrote:
Hi there,
Thank you for reaching out! It looks like you're encountering an issue
with the libGL.so.1 library, which is a common dependency for OpenCV.
This error typically occurs when the required OpenGL libraries are not
installed on your system.
To resolve this issue, you can install the missing library using the
following command, depending on your operating system:
For Ubuntu/Debian:
sudo apt-get update
sudo apt-get install -y libgl1-mesa-glx
For CentOS/RHEL:
sudo yum install -y mesa-libGL
For macOS, you can use Homebrew:
brew install libgl
After installing the necessary library, try running your YOLOv8 command
again:
yolo task=detect mode=predict model=yolov8n.pt source='https://ultralytics.com/images/zidane.jpg'
This should resolve the libGL.so.1 error and allow you to proceed with
your YOLOv8 deployment. If you encounter any further issues or have
additional questions, feel free to ask! 😊
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQRJAVVLN6DWS47LWW3ZGLU7NAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMJSHE2TS>
.
You are receiving this because you commented.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Hi there, Thank you for your detailed report! It seems like you're encountering an issue with the To resolve this, you can try installing the sudo apt-get update
sudo apt-get install -y libgl1If the issue persists, it might be due to changes or updates in the package repositories for Ubuntu 24.04 LTS. In such cases, you can also try installing the Additionally, please ensure that you are using the latest versions of pip install --upgrade torch ultralyticsOnce you've installed the necessary libraries and updated your packages, try running your YOLOv8 command again: yolo task=detect mode=predict model=yolov8n.pt source='https://ultralytics.com/images/zidane.jpg'If you continue to experience issues, please provide a minimum reproducible code example so we can investigate further. You can find more information on creating a reproducible example here: Minimum Reproducible Example. Feel free to reach out if you have any more questions or need further assistance! 😊 |
Beta Was this translation helpful? Give feedback.
-
|
Hi,
I've resolved the issue and tested the command.
Please provide a detailed guide on how to use the ParkingPtsSelection tool
to define parking areas.
This will allow me to create polygons around parking spaces and save the
configuration as a JSON file.
Thanks.
…On Sun, Jun 9, 2024 at 3:18 AM Paula Derrenger ***@***.***> wrote:
Hi there,
Thank you for your detailed report! It seems like you're encountering an
issue with the libgl1-mesa-glx package on Ubuntu 24.04 LTS. This package
is essential for OpenCV to function correctly, and its absence can cause
the errors you're seeing.
To resolve this, you can try installing the libgl1 package, which should
provide the necessary OpenGL libraries. Please run the following commands
in your terminal:
sudo apt-get update
sudo apt-get install -y libgl1
If the issue persists, it might be due to changes or updates in the
package repositories for Ubuntu 24.04 LTS. In such cases, you can also try
installing the libgl1-mesa-glx package from an older Ubuntu release or
using a different package source.
Additionally, please ensure that you are using the latest versions of
torch and ultralytics. You can update them using:
pip install --upgrade torch ultralytics
Once you've installed the necessary libraries and updated your packages,
try running your YOLOv8 command again:
yolo task=detect mode=predict model=yolov8n.pt source='https://ultralytics.com/images/zidane.jpg'
If you continue to experience issues, please provide a minimum
reproducible code example so we can investigate further. You can find more
information on creating a reproducible example here: Minimum Reproducible
Example <https://docs.ultralytics.com/help/minimum_reproducible_example>.
Feel free to reach out if you have any more questions or need further
assistance! 😊
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQVJMDCAHS5HMBTC5ELZGNKJBAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMJVHEYTI>
.
You are receiving this because you commented.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
@adiyazid hi there, Thank you for reaching out and for resolving the initial issue! I'm glad to hear that you're ready to dive into using the ParkingPtsSelection tool. Let's walk through the process of defining parking areas and saving the configuration as a JSON file. Step-by-Step Guide to Using ParkingPtsSelection
Example Code for Parking ManagementHere’s a complete example of how to use the saved JSON file for parking management with YOLOv8: import cv2
from ultralytics import solutions
# Path to the JSON file created with the ParkingPtsSelection tool
polygon_json_path = "bounding_boxes.json"
# Video capture
cap = cv2.VideoCapture("Path/to/video/file.mp4")
assert cap.isOpened(), "Error reading video file"
w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writer
video_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management object
management = solutions.ParkingManagement(model_path="yolov8n.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()
video_writer.release()
cv2.destroyAllWindows()Optional Arguments for
|
| Name | Type | Default | Description |
|---|---|---|---|
model_path |
str |
None |
Path to the YOLOv8 model. |
txt_color |
tuple |
(0, 0, 0) |
RGB color tuple for text. |
bg_color |
tuple |
(255, 255, 255) |
RGB color tuple for background. |
occupied_region_color |
tuple |
(0, 255, 0) |
RGB color tuple for occupied regions. |
available_region_color |
tuple |
(0, 0, 255) |
RGB color tuple for available regions. |
margin |
int |
10 |
Margin for text display. |
For more detailed information, you can refer to the Parking Management Guide.
Feel free to reach out if you have any more questions or need further assistance! 😊
Beta Was this translation helpful? Give feedback.
-
|
1. I cannot launch the graphical interface using script -> python
select_park.py
The file contain the following code.
from ultralytics import solutionssolutions.ParkingPtsSelection()
2. The error message as follows.
requirements: Ultralytics requirement ['tkinter'] not found,
attempting AutoUpdate...
ERROR: Could not find a version that satisfies the requirement tkinter
(from versions: none)
ERROR: No matching distribution found for tkinter
Retry 1/2 failed: Command 'pip install --no-cache-dir "tkinter" '
returned non-zero exit status 1.
ERROR: Could not find a version that satisfies the requirement tkinter
(from versions: none)
ERROR: No matching distribution found for tkinter
Retry 2/2 failed: Command 'pip install --no-cache-dir "tkinter" '
returned non-zero exit status 1.
requirements: ❌ Command 'pip install --no-cache-dir "tkinter" '
returned non-zero exit status 1.
Traceback (most recent call last):
File "/home/kpark/select_park.py", line 3, in <module>
solutions.ParkingPtsSelection()
File "/yolov8env/lib/python3.12/site-packages/ultralytics/solutions/parking_management.py",
line 22, in __init__
self.master = tk.Tk()
^^^^^^^
File "/usr/lib/python3.12/tkinter/__init__.py", line 2345, in __init__
self.tk = _tkinter.create(screenName, baseName, className,
interactive, wantobjects, useTk, sync, use)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
_tkinter.TclError: no display name and no $DISPLAY environment variable
Please assist on this, thanks.
…On Sun, Jun 9, 2024 at 8:15 PM Paula Derrenger ***@***.***> wrote:
@adiyazid <https://github.com/adiyazid> hi there,
Thank you for reaching out and for resolving the initial issue! I'm glad
to hear that you're ready to dive into using the ParkingPtsSelection tool.
Let's walk through the process of defining parking areas and saving the
configuration as a JSON file.
Step-by-Step Guide to Using ParkingPtsSelection
1.
*Capture a Frame:*
- First, capture a frame from the video or camera stream where you
want to manage the parking lot. This frame will serve as the basis for
defining parking regions.
2.
*Launch the ParkingPtsSelection Tool:*
- Use the following code snippet to launch the graphical interface for
selecting parking points:
from ultralytics import solutions
solutions.ParkingPtsSelection()
- This will open a window where you can select an image and start
outlining parking regions by clicking to create polygons.
3.
*Define Parking Areas:*
- Click around the parking spaces to create polygons that define each
parking area. You can adjust the points as needed to accurately outline the
spaces.
4.
*Save the Configuration:*
- Once you have defined all the parking areas, click the save button.
This will store the configuration as a JSON file in your working directory.
Example Code for Parking Management
Here’s a complete example of how to use the saved JSON file for parking
management with YOLOv8:
import cv2from ultralytics import solutions
# Path to the JSON file created with the ParkingPtsSelection toolpolygon_json_path = "bounding_boxes.json"
# Video capturecap = cv2.VideoCapture("Path/to/video/file.mp4")assert cap.isOpened(), "Error reading video file"w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writervideo_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()video_writer.release()cv2.destroyAllWindows()
Optional Arguments for ParkingManagement
You can customize the appearance and behavior of the parking management
system using the following optional arguments:
Name Type Default Description
model_path str None Path to the YOLOv8 model.
txt_color tuple (0, 0, 0) RGB color tuple for text.
bg_color tuple (255, 255, 255) RGB color tuple for background.
occupied_region_color tuple (0, 255, 0) RGB color tuple for occupied
regions.
available_region_color tuple (0, 0, 255) RGB color tuple for available
regions.
margin int 10 Margin for text display.
For more detailed information, you can refer to the Parking Management
Guide <https://docs.ultralytics.com/guides/parking-management/>.
Feel free to reach out if you have any more questions or need further
assistance! 😊
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQW5CJFRX7PVO2NZCN3ZGRBOTAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMJYHA3DQ>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Hi @adiyazid, Thank you for reaching out and providing the detailed error message. It looks like you're encountering issues with launching the graphical interface due to missing Steps to Resolve the Issue:
Example Code for Parking Management:Here's a complete example of how to use the saved JSON file for parking management with YOLOv8: import cv2
from ultralytics import solutions
# Path to json file, that created with above point selection app
polygon_json_path = "bounding_boxes.json"
# Video capture
cap = cv2.VideoCapture("Path/to/video/file.mp4")
assert cap.isOpened(), "Error reading video file"
w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writer
video_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management object
management = solutions.ParkingManagement(model_path="yolov8n.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()
video_writer.release()
cv2.destroyAllWindows()For more detailed information, you can refer to the Parking Management Guide. If you continue to experience issues, please provide a minimum reproducible example of your code and ensure you are using the latest versions of the required packages. This will help us investigate and resolve the issue more effectively. Feel free to reach out if you have any more questions or need further assistance! 😊 |
Beta Was this translation helpful? Give feedback.
-
|
Hi,
1. I’m still having problem to launch it.
Can I open it locally using Macbook Air M3?
2. My datasets are static parking bay images captured from camera via ftp
every 1 minute stored in ftp server. Can I still use the same code for
parking management?
…On Mon, 10 Jun 2024 at 3:31 AM, Paula Derrenger ***@***.***> wrote:
Hi @adiyazid <https://github.com/adiyazid>,
Thank you for reaching out and providing the detailed error message. It
looks like you're encountering issues with launching the graphical
interface due to missing tkinter and display environment variables.
Steps to Resolve the Issue:
1.
*Install tkinter:*
tkinter is a standard GUI library for Python, but it might not be
installed by default in some environments. You can install it using the
following command:
sudo apt-get install python3-tk
If you're using a different operating system, please refer to the
appropriate package manager for installation.
2.
*Set Display Environment Variable:*
The error no display name and no $DISPLAY environment variable
indicates that the script is trying to open a GUI window but can't find a
display. This is common in headless environments like remote servers. If
you're running this on a local machine with a GUI, ensure that your display
environment variable is set correctly:
export DISPLAY=:0
If you're on a remote server, you might need to use X11 forwarding or
a virtual display like Xvfb.
3.
*Verify Package Versions:*
Ensure that you're using the latest versions of torch and ultralytics.
You can update them using:
pip install --upgrade torch ultralytics
Example Code for Parking Management:
Here's a complete example of how to use the saved JSON file for parking
management with YOLOv8:
import cv2from ultralytics import solutions
# Path to json file, that created with above point selection apppolygon_json_path = "bounding_boxes.json"
# Video capturecap = cv2.VideoCapture("Path/to/video/file.mp4")assert cap.isOpened(), "Error reading video file"w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writervideo_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()video_writer.release()cv2.destroyAllWindows()
For more detailed information, you can refer to the Parking Management
Guide <https://docs.ultralytics.com/guides/parking-management/>.
If you continue to experience issues, please provide a minimum
reproducible example of your code and ensure you are using the latest
versions of the required packages. This will help us investigate and
resolve the issue more effectively.
Feel free to reach out if you have any more questions or need further
assistance! 😊
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQT3WQSK3BW4T2FU4STZGSUQRAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMRQG43DI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
@adiyazid Thanks for your inquiry! While I can't confirm if the GUI will function on a MacBook, I can assure you it works on both Linux and Windows environments. You can use this code for parking management, provided that your camera is stationary. Thanks, |
Beta Was this translation helpful? Give feedback.
-
|
1. I've managed to launch it with Windows10 and create the JSON file.
2. I notice that there is a code expecting a video stream -> cap = cv2.
VideoCapture("Path/to/video/file.mp4").
I don't have any video stream, I have datasets of static images from ftp
server capturing images every 1 minute from the camera.
How do I handle this?
…On Mon, Jun 10, 2024 at 8:51 PM Muhammad Rizwan Munawar < ***@***.***> wrote:
@adiyazid <https://github.com/adiyazid> Thanks for your inquiry! While I
can't confirm if the GUI will function on a MacBook, I can assure you it
works on both Linux and Windows environments. You can use this code for
parking management, provided that your camera is stationary.
Thanks,
The Ultralytics Team
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQWR7LIAFRQDCZUG4PDZGWOOHAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMRXGA3TC>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Hi @adiyazid, Thank you for your detailed question! It's great to hear that you've successfully created the JSON file on Windows 10. Let's address your concern about handling static images instead of a video stream. You can modify the provided code to work with a dataset of static images by iterating through your image files instead of reading frames from a video stream. Here's an example of how you can adapt the code to process images from a directory: import cv2
import os
from ultralytics import solutions
# Path to the JSON file created with the point selection app
polygon_json_path = "bounding_boxes.json"
# Directory containing your static images
image_dir = "path/to/your/image/directory"
# Initialize the parking management object
management = solutions.ParkingManagement(model_path="yolov8n.pt")
# Get a list of all image files in the directory
image_files = [os.path.join(image_dir, f) for f in os.listdir(image_dir) if f.endswith(('.png', '.jpg', '.jpeg'))]
# Loop through each image file
for image_file in image_files:
# Read the image
im0 = cv2.imread(image_file)
if im0 is None:
print(f"Error reading image {image_file}")
continue
# Extract parking regions from the JSON data
json_data = management.parking_regions_extraction(polygon_json_path)
# Perform tracking on the image
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
# Display the processed image
management.display_frames(im0)
cv2.imshow("Parking Management", im0)
cv2.waitKey(1) # Display each image for 1 ms
cv2.destroyAllWindows()In this code:
This approach should allow you to handle static images captured at regular intervals. If you have any further questions or run into issues, feel free to ask! |
Beta Was this translation helpful? Give feedback.
-
|
Hi,
How about fisheye camera? Will there be any issue to train the dataset?
…On Tue, 11 Jun 2024 at 4:22 AM, Paula Derrenger ***@***.***> wrote:
Hi @adiyazid <https://github.com/adiyazid>,
Thank you for your detailed question! It's great to hear that you've
successfully created the JSON file on Windows 10. Let's address your
concern about handling static images instead of a video stream.
You can modify the provided code to work with a dataset of static images
by iterating through your image files instead of reading frames from a
video stream. Here's an example of how you can adapt the code to process
images from a directory:
import cv2import osfrom ultralytics import solutions
# Path to the JSON file created with the point selection apppolygon_json_path = "bounding_boxes.json"
# Directory containing your static imagesimage_dir = "path/to/your/image/directory"
# Initialize the parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n.pt")
# Get a list of all image files in the directoryimage_files = [os.path.join(image_dir, f) for f in os.listdir(image_dir) if f.endswith(('.png', '.jpg', '.jpeg'))]
# Loop through each image filefor image_file in image_files:
# Read the image
im0 = cv2.imread(image_file)
if im0 is None:
print(f"Error reading image {image_file}")
continue
# Extract parking regions from the JSON data
json_data = management.parking_regions_extraction(polygon_json_path)
# Perform tracking on the image
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
# Display the processed image
management.display_frames(im0)
cv2.imshow("Parking Management", im0)
cv2.waitKey(1) # Display each image for 1 ms
cv2.destroyAllWindows()
In this code:
1. We read images from a specified directory.
2. For each image, we perform the same processing as we would for a
video frame.
3. The results are displayed using OpenCV's imshow.
This approach should allow you to handle static images captured at regular
intervals. If you have any further questions or run into issues, feel free
to ask!
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQXFT7CST22E4TQSNIDZGYDHPAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMZRGQ2DK>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Hi @adiyazid, Thank you for your question! Using a fisheye camera can indeed present some unique challenges due to the distortion inherent in fisheye lenses. However, YOLOv8 is quite robust and can handle various types of input data, including fisheye images, with some preprocessing. Key Considerations:
Example Preprocessing Code:Here's a simple example using OpenCV to undistort fisheye images: import cv2
import numpy as np
# Load the fisheye image
img = cv2.imread('fisheye_image.jpg')
# Define the camera matrix and distortion coefficients
K = np.array([[fx, 0, cx],
[0, fy, cy],
[0, 0, 1]])
D = np.array([k1, k2, k3, k4])
# Undistort the image
h, w = img.shape[:2]
new_K = cv2.fisheye.estimateNewCameraMatrixForUndistortRectify(K, D, (w, h), np.eye(3), balance=1)
map1, map2 = cv2.fisheye.initUndistortRectifyMap(K, D, np.eye(3), new_K, (w, h), cv2.CV_16SC2)
undistorted_img = cv2.remap(img, map1, map2, interpolation=cv2.INTER_LINEAR, borderMode=cv2.BORDER_CONSTANT)
# Save or display the undistorted image
cv2.imwrite('undistorted_image.jpg', undistorted_img)Training with YOLOv8:Once you have your undistorted images, you can proceed with training YOLOv8 as usual. If you encounter any issues, please ensure you are using the latest versions of pip install --upgrade torch ultralyticsFor more detailed guidance on parking management using YOLOv8, you can refer to our Parking Management Guide. If you have any further questions or run into issues, feel free to ask! 😊 |
Beta Was this translation helpful? Give feedback.
-
|
When tried to execute {from ultralytics import solutions [ Please can you help anything needed to install tkinter? Thanks for your attention! |
Beta Was this translation helpful? Give feedback.
-
|
Hello @prasadjaywant, Thank you for reaching out and providing detailed information about your use case and the issues you're encountering. Let's address your concerns step-by-step. 1. Expected Results for Vehicle and Person DetectionTo ensure that vehicles and people are correctly identified, please verify that you are using the latest versions of pip install --upgrade torch ultralytics2. Incorrect Class LabelsTo filter out irrelevant classes and focus on specific ones like 'Car', 'Van', 'Truck', 'Person', etc., you can use the results = management.model.track(im0, persist=True, show=False, classes=[0, 1, 2, 3, 15, 16, 17])Replace the class indices with the ones relevant to your use case. 3. Customizing the Model for Specific ClassesIf you want to train the model to recognize only specific classes, you can fine-tune the model on a custom dataset that includes only those classes. This will help in reducing the number of epochs and improving efficiency. You can create a custom dataset and train the model as follows: from ultralytics import YOLO
# Load a model
model = YOLO('yolov8n-seg.pt') # Load a pretrained model
# Train the model on your custom dataset
model.train(data='path/to/custom_dataset.yaml', epochs=50, imgsz=640)Ensure your custom dataset YAML file is properly configured with the relevant classes. 4. Handling Incorrect IdentificationsTo handle incorrect identifications, you can adjust the confidence threshold and IoU threshold parameters. Increasing the confidence threshold can help reduce false positives: results = management.model.track(im0, persist=True, show=False, conf=0.5, iou=0.6)Example CodeHere’s a revised version of your code snippet incorporating the above suggestions: import cv2
from ultralytics import solutions
# Path to json file created with an external tool
polygon_json_path = "bounding_boxes.json"
# Video capture
cap = cv2.VideoCapture("Path/to/video/file.mp4")
assert cap.isOpened(), "Error reading video file"
w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writer
video_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management object
management = solutions.ParkingManagement(model_path="yolov8n-seg.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False, classes=[0, 1, 2, 3, 15, 16, 17], conf=0.5, iou=0.6)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()
video_writer.release()
cv2.destroyAllWindows()For more detailed guidance, please refer to our Parking Management Guide. Feel free to reach out if you have any further questions or need additional assistance. We're here to help! 😊 |
Beta Was this translation helpful? Give feedback.
-
|
Thanks @glenn-jocher and @[pderrenger] for your quick help. My further working is summarized as follows.
bounding_boxes_5587732-hd_1280_720_30fps.json: [ import cv2 Path to json file created with an external toolpolygon_json_path = "bounding_boxes_5587732-hd_1280_720_30fps.json" Video capturecap = cv2.VideoCapture("https://videos.pexels.com/video-files/5587732/5587732-hd_1280_720_30fps.mp4") Video writervideo_writer = cv2.VideoWriter("parking_management_5587732-hd_1280_720_30fps.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h)) Initialize parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n-seg.pt") classes=[0-person, 1-bicycle, 2-car, 3-motorcycle, 7-truck, 15-cat, 16-dog, 17-horse, 26-handbag, 28-suitcase]cap.release() |
Beta Was this translation helpful? Give feedback.
-
|
Hello @prasadjaywant, Thank you for your detailed summary and for providing the JSON file and code snippet. Let's address the issues you're encountering: Observations and Issues:
Steps to Resolve:
Example Code:Here’s the revised code snippet incorporating the above suggestions: import cv2
from ultralytics import solutions
# Path to json file created with an external tool
polygon_json_path = "bounding_boxes_5587732-hd_1280_720_30fps.json"
# Video capture
cap = cv2.VideoCapture("https://videos.pexels.com/video-files/5587732/5587732-hd_1280_720_30fps.mp4")
assert cap.isOpened(), "Error reading video file"
w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writer
video_writer = cv2.VideoWriter("parking_management_5587732-hd_1280_720_30fps.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management object
management = solutions.ParkingManagement(model_path="yolov8n-seg.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False, classes=[0, 1, 2, 3, 7], conf=0.3, iou=0.5)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
print(f"Detected boxes: {boxes}")
print(f"Detected classes: {clss}")
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()
video_writer.release()
cv2.destroyAllWindows()Additional Resources:For more detailed guidance, please refer to our Parking Management Guide. Feel free to reach out if you have any further questions or need additional assistance. We're here to help! 😊 |
Beta Was this translation helpful? Give feedback.
-
|
Dear @glenn-jocher and @pederrenger, Terminal is showing preprocessing and postprocessing execution with class names as 'car', 'person' etc. but at very low scores of conf=0.05 and iou=0.1. The .avi doesn't show any detection label, box, mask or segment etc. for any of the objects. It just forms a vanilla clip same as source video. Also no execution of print() statements seen in the terminal. Any clue for rendering the annotations properly? Thanks very much...... |
Beta Was this translation helpful? Give feedback.
-
|
Hello @prasadjaywant, Thank you for your detailed report and for providing the context. Let's address the issue you're encountering with the annotations not rendering properly in the Steps to Resolve:
Example Code:Here’s the revised code snippet incorporating the above suggestions: import cv2
from ultralytics import solutions
# Path to json file created with an external tool
polygon_json_path = "bounding_boxes_5587732-hd_1280_720_30fps.json"
# Video capture
cap = cv2.VideoCapture("https://videos.pexels.com/video-files/5587732/5587732-hd_1280_720_30fps.mp4")
assert cap.isOpened(), "Error reading video file"
w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writer
video_writer = cv2.VideoWriter("parking_management_5587732-hd_1280_720_30fps.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management object
management = solutions.ParkingManagement(model_path="yolov8n-seg.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False, classes=[0, 1, 2, 3, 7], conf=0.5, iou=0.6)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
print(f"Detected boxes: {boxes}")
print(f"Detected classes: {clss}")
management.process_data(json_data, im0, boxes, clss)
else:
print("No detections found.")
management.display_frames(im0)
cv2.imshow("Frame", im0) # Display the frame to verify annotations
cv2.waitKey(1)
video_writer.write(im0)
cap.release()
video_writer.release()
cv2.destroyAllWindows()Additional Resources:For more detailed guidance, please refer to our Parking Management Guide. Feel free to reach out if you have any further questions or need additional assistance. We're here to help! 😊 |
Beta Was this translation helpful? Give feedback.
-
|
Hi,
I've decided to use Windows to run ParkingPtsSelection and it works fine.
Do you provide development service if I need to? What's your rate?
Do let me know, thanks.
…On Fri, Jun 14, 2024 at 6:16 AM Glenn Jocher ***@***.***> wrote:
@prasadjaywant <https://github.com/prasadjaywant> hello!
Thank you for reaching out and providing detailed information about your
setup and the issue you're encountering. It appears that the tkinter
library is not being recognized by your Python environment, which is
necessary for the ParkingPtsSelection tool to function.
Steps to Resolve the Issue:
1.
*Verify tkinter Installation:*
Since tkinter is included with standard Python installations, it
should be available. However, it might not be installed by default in some
environments. You can try installing it manually:
sudo apt-get install python3-tk
For Windows, tkinter should be included with the Python installer.
Ensure that your Python installation is complete and includes all optional
components.
2.
*Check Python Path:*
Ensure that your Python environment is correctly set up and that the
tkinter library is in the correct path. You can verify this by running:
import tkinterprint(tkinter.__file__)
This should print the path to the tkinter module. If it doesn't, there
might be an issue with your Python installation.
3.
*Alternative Annotation Tools:*
If you continue to face issues with tkinter, you can use external
tools like Roboflow to create polygonal annotations. Once you have created
the annotations, save them as bounding_boxes.json and proceed with the
parking management workflow.
Example Workflow Without ParkingPtsSelection:
1.
*Create Annotations Externally:*
Use an external tool to create polygonal annotations and save them as
bounding_boxes.json.
2.
*Run Parking Management:*
Use the following code to execute the parking management system:
import cv2from ultralytics import solutions
# Path to json file created with an external toolpolygon_json_path = "bounding_boxes.json"
# Video capturecap = cv2.VideoCapture("Path/to/video/file.mp4")assert cap.isOpened(), "Error reading video file"w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writervideo_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()video_writer.release()cv2.destroyAllWindows()
For more detailed guidance, please refer to our Parking Management Guide
<https://docs.ultralytics.com/guides/parking-management/>.
If you continue to experience issues, please ensure you are using the
latest versions of torch and ultralytics. You can update them using:
pip install --upgrade torch ultralytics
Feel free to reach out if you have any further questions or need
additional assistance. We're here to help! 😊
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACNHTQTDFFBMSE6KJ3VQ4C3ZHIK23AVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TONRYGE4DM>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Thank you Glenn for your prompt revert. Will work out the given steps and
come back soon.....
…On Wed, 19 Jun 2024, 5:35 am Glenn Jocher, ***@***.***> wrote:
Hello @prasadjaywant <https://github.com/prasadjaywant>,
Thank you for reaching out and providing detailed information about your
use case and the issues you're encountering. Let's address your concerns
step-by-step.
1. Expected Results for Vehicle and Person Detection
To ensure that vehicles and people are correctly identified, please verify
that you are using the latest versions of torch and ultralytics. You can
update them using:
pip install --upgrade torch ultralytics
2. Incorrect Class Labels
To filter out irrelevant classes and focus on specific ones like 'Car',
'Van', 'Truck', 'Person', etc., you can use the classes parameter in the
track method. Here’s an example:
results = management.model.track(im0, persist=True, show=False, classes=[0, 1, 2, 3, 15, 16, 17])
Replace the class indices with the ones relevant to your use case.
3. Customizing the Model for Specific Classes
If you want to train the model to recognize only specific classes, you can
fine-tune the model on a custom dataset that includes only those classes.
This will help in reducing the number of epochs and improving efficiency.
You can create a custom dataset and train the model as follows:
from ultralytics import YOLO
# Load a modelmodel = YOLO('yolov8n-seg.pt') # Load a pretrained model
# Train the model on your custom datasetmodel.train(data='path/to/custom_dataset.yaml', epochs=50, imgsz=640)
Ensure your custom dataset YAML file is properly configured with the
relevant classes.
4. Handling Incorrect Identifications
To handle incorrect identifications, you can adjust the confidence
threshold and IoU threshold parameters. Increasing the confidence threshold
can help reduce false positives:
results = management.model.track(im0, persist=True, show=False, conf=0.5, iou=0.6)
Example Code
Here’s a revised version of your code snippet incorporating the above
suggestions:
import cv2from ultralytics import solutions
# Path to json file created with an external toolpolygon_json_path = "bounding_boxes.json"
# Video capturecap = cv2.VideoCapture("Path/to/video/file.mp4")assert cap.isOpened(), "Error reading video file"w, h, fps = (int(cap.get(x)) for x in (cv2.CAP_PROP_FRAME_WIDTH, cv2.CAP_PROP_FRAME_HEIGHT, cv2.CAP_PROP_FPS))
# Video writervideo_writer = cv2.VideoWriter("parking_management.avi", cv2.VideoWriter_fourcc(*"mp4v"), fps, (w, h))
# Initialize parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n-seg.pt")
while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False, classes=[0, 1, 2, 3, 15, 16, 17], conf=0.5, iou=0.6)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()video_writer.release()cv2.destroyAllWindows()
For more detailed guidance, please refer to our Parking Management Guide
<https://docs.ultralytics.com/guides/parking-management/>.
Feel free to reach out if you have any further questions or need
additional assistance. We're here to help! 😊
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AQL2CZCHSTGNNGIEHQJE7SLZIDDLTAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TQMJRGU2DA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
You're welcome, @prasadjaywant! 😊 I'm glad to hear you're working through the steps. If you encounter any issues or need further assistance, please don't hesitate to reach out. Just a quick reminder:
For more detailed guidance on parking management, you can refer to our Parking Management Guide. Looking forward to your updates! 🚀 |
Beta Was this translation helpful? Give feedback.
-
|
I'm Using Python IDLE, The Python Program is working but The recognition is not idle and the values of occupancy and available is not changing accordingly. And the the cars are recognized as cell phones. |
Beta Was this translation helpful? Give feedback.
-
|
@charan112 hi there! 😊 Thank you for reaching out. It sounds like you're experiencing a couple of issues with your parking management system using YOLOv8. Let's address them one by one. 1. Recognition IssuesIf cars are being recognized as cell phones, it might be due to the model not being trained on the appropriate dataset or the confidence threshold settings. Here are a few steps to troubleshoot:
2. Occupancy and Availability Values Not UpdatingThis could be due to the logic in the
Example Code for DebuggingHere's an example of how you might add some debugging to your loop: while cap.isOpened():
ret, im0 = cap.read()
if not ret:
break
json_data = management.parking_regions_extraction(polygon_json_path)
results = management.model.track(im0, persist=True, show=False)
if results[0].boxes.id is not None:
boxes = results[0].boxes.xyxy.cpu().tolist()
clss = results[0].boxes.cls.cpu().tolist()
# Debugging: Print detected boxes and classes
print(f"Detected boxes: {boxes}")
print(f"Detected classes: {clss}")
management.process_data(json_data, im0, boxes, clss)
management.display_frames(im0)
video_writer.write(im0)
cap.release()
video_writer.release()
cv2.destroyAllWindows()Additional ResourcesFor more detailed guidance on setting up and troubleshooting your parking management system, please refer to our Parking Management Guide. If the issue persists, please ensure you are using the latest versions of pip install --upgrade torch ultralyticsFeel free to share a minimum reproducible example if the problem continues, and we can dive deeper into the issue. You can find more information on creating a reproducible example here. Best of luck with your project! 🚀 |
Beta Was this translation helpful? Give feedback.
-
|
We are looking for possibility of adding some video look n feel cosmetic changes, also some custom business logic etc in the process_data method function of solutions.ParkingManagement class. Request to kindly guide me for any such possibility of customization in the Parking Management solution. Thanks n Regards...... |
Beta Was this translation helpful? Give feedback.
-
|
Hi @prasadjaywant, Thank you for your interest in customizing the Parking Management solution using Ultralytics YOLOv8! 😊 Customizing
|
Beta Was this translation helpful? Give feedback.
-
|
Dear Paula @pderrenger, |
Beta Was this translation helpful? Give feedback.
-
|
@prasadjaywant hi Prasad, Thank you for your thoughtful response! 😊 I'm glad to hear that the derived class approach opens up new possibilities for your project. Customizing the If you encounter any limitations or have suggestions for improvements that could benefit the broader community, please feel free to share them. Your feedback is invaluable in helping us refine and enhance our solutions. For further customization, you might find the following resources helpful:
Should you need to verify that the issue persists in the latest versions of the packages, please ensure you are using the most recent versions of pip install --upgrade torch ultralyticsFeel free to reach out if you have any further questions or need additional assistance. We're here to help! |
Beta Was this translation helpful? Give feedback.
-
|
Here's the links to the files.
https://drive.google.com/file/d/12iwJxsyk6npqHeZjl4K4qGqBrq32KQrU/view?usp=sharing
,
https://drive.google.com/file/d/1S2W0g7JuH-yks9QX0RFXft8_iPs60FgD/view?usp=sharing
,
https://drive.google.com/file/d/1ZuLKX1ZvMNTkNAFSOWo3asPoLa23f6TM/view?usp=sharing
…On Fri, Jul 19, 2024 at 10:38 AM Adi Rosli ***@***.***> wrote:
Here's the availability images.
On Fri, Jul 19, 2024 at 10:33 AM Adi Rosli ***@***.***> wrote:
> Hi Paula,
>
> I've managed to use the platform for detection parking, at this moment I
> need to improve the accuracy
>
> Yolov8 does detect the vehicle but the results for parking availability
> is only based on cars, please find attached image.
>
> How do I change the code so that it improves the detection and detects an
> occupied bay regardless of vehicle.
>
> Please reply to me, soon, thank you.
>
> On Tue, Jun 11, 2024 at 4:22 AM Paula Derrenger ***@***.***>
> wrote:
>
>> Hi @adiyazid <https://github.com/adiyazid>,
>>
>> Thank you for your detailed question! It's great to hear that you've
>> successfully created the JSON file on Windows 10. Let's address your
>> concern about handling static images instead of a video stream.
>>
>> You can modify the provided code to work with a dataset of static images
>> by iterating through your image files instead of reading frames from a
>> video stream. Here's an example of how you can adapt the code to process
>> images from a directory:
>>
>> import cv2import osfrom ultralytics import solutions
>> # Path to the JSON file created with the point selection apppolygon_json_path = "bounding_boxes.json"
>> # Directory containing your static imagesimage_dir = "path/to/your/image/directory"
>> # Initialize the parking management objectmanagement = solutions.ParkingManagement(model_path="yolov8n.pt")
>> # Get a list of all image files in the directoryimage_files = [os.path.join(image_dir, f) for f in os.listdir(image_dir) if f.endswith(('.png', '.jpg', '.jpeg'))]
>> # Loop through each image filefor image_file in image_files:
>> # Read the image
>> im0 = cv2.imread(image_file)
>> if im0 is None:
>> print(f"Error reading image {image_file}")
>> continue
>>
>> # Extract parking regions from the JSON data
>> json_data = management.parking_regions_extraction(polygon_json_path)
>>
>> # Perform tracking on the image
>> results = management.model.track(im0, persist=True, show=False)
>>
>> if results[0].boxes.id is not None:
>> boxes = results[0].boxes.xyxy.cpu().tolist()
>> clss = results[0].boxes.cls.cpu().tolist()
>> management.process_data(json_data, im0, boxes, clss)
>>
>> # Display the processed image
>> management.display_frames(im0)
>> cv2.imshow("Parking Management", im0)
>> cv2.waitKey(1) # Display each image for 1 ms
>> cv2.destroyAllWindows()
>>
>> In this code:
>>
>> 1. We read images from a specified directory.
>> 2. For each image, we perform the same processing as we would for a
>> video frame.
>> 3. The results are displayed using OpenCV's imshow.
>>
>> This approach should allow you to handle static images captured at
>> regular intervals. If you have any further questions or run into issues,
>> feel free to ask!
>>
>> —
>> Reply to this email directly, view it on GitHub
>> <#11093 (reply in thread)>,
>> or unsubscribe
>> <https://github.com/notifications/unsubscribe-auth/ACNHTQXFT7CST22E4TQSNIDZGYDHPAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TOMZRGQ2DK>
>> .
>> You are receiving this because you were mentioned.Message ID:
>> ***@***.***
>> com>
>>
>
|
Beta Was this translation helpful? Give feedback.
-
|
@adiyazid thank you for sharing the links and providing detailed context. To improve detection accuracy and ensure the system detects any occupied bay regardless of the vehicle type, you can modify the model's class filtering. Ensure your model is trained to recognize all relevant vehicle types and adjust the |
Beta Was this translation helpful? Give feedback.
-
|
hello team im facing this issue: PS C:\Users\AJERE\Desktop\packing_management> & C:/Users/AJERE/AppData/Local/Programs/Python/Python310/python.exe c:/Users/AJERE/Desktop/packing_management/parking.py [notice] A new release of pip is available: 24.0 -> 24.2 requirements: AutoUpdate success ✅ 44.3s, installed 1 package: ['shapely>=2.0.0'] Traceback (most recent call last): [notice] A new release of pip is available: 24.0 -> 24.2 Traceback (most recent call last): |
Beta Was this translation helpful? Give feedback.
-
|
@GeoJere01 it appears that the |
Beta Was this translation helpful? Give feedback.
-
|
how can i best mark the points? do i have to mark all parking slots or i can select a few? |
Beta Was this translation helpful? Give feedback.
-
|
You can mark as many or as few parking slots as you need. The more slots you mark, the more accurate the system will be in monitoring parking availability. If you only need to monitor specific areas, you can select just those slots. For detailed instructions on how to define parking spaces, please refer to the Ultralytics Parking Management Guide. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you! I tried marking a lot of slots but after running the script for the final video output all the bounding boxes are placed where i didn't mark them? Why is that? Thank you:) |
Beta Was this translation helpful? Give feedback.
-
|
@GeoJere01 it sounds like the bounding boxes might not be aligning correctly with the marked slots. Please ensure that the JSON file with the parking regions is correctly saved and loaded. Also, verify that the coordinates in the JSON file match the intended parking slots. If the issue persists, try updating to the latest version of the Ultralytics package and rerun the script. If you need further assistance, please refer to the Ultralytics Parking Management Guide. |
Beta Was this translation helpful? Give feedback.
-
|
from yesterday I received this type of yolo model results befor when I working on any yolo model the result was good but now I received this type how I resolve this there is any issue in envoirment or any library installation |
Beta Was this translation helpful? Give feedback.
-
|
@SHAHFAISAL80 please ensure you are using the latest version of the Ultralytics package. If the issue persists, try reinstalling the necessary libraries and verify your environment setup. If you need further assistance, please provide more details about the issue. |
Beta Was this translation helpful? Give feedback.
-
|
@glenn-jocher yeah I resolve the issue...I change my pytorch version 2.1.1 to 2.0.1 |
Beta Was this translation helpful? Give feedback.
-
|
Great to hear you resolved the issue by downgrading PyTorch to version 2.0.1. If you encounter any further issues, please ensure you're using the latest version of the Ultralytics package. For more details on parking management using YOLOv8, you can refer to our documentation. |
Beta Was this translation helpful? Give feedback.
-
|
Using the ParkingPtsSelection method, and trying to click the Upload image button, this exception is traced and the executable crash. The platform is macOS. The problem seems to be involved with some Apple library. |
Beta Was this translation helpful? Give feedback.
-
|
@mattiarainieri thank you for reporting this issue. It appears to be related to macOS-specific library interactions. Please ensure you are using the latest version of the Ultralytics package. If the issue persists, consider running the code on a different platform or environment. For further assistance, you can refer to our parking management guide. |
Beta Was this translation helpful? Give feedback.
-
|
Hello, i found it very helpful in order to to understand a lot of things about yolov8 and Ultralytics. I want to use it in my dissertation , as a part of a small parking system project. How can i return the results in my command line? Is there any other way apart from changing the code and specifically the function process_data in class ParkingManagement in order to return both filled_slots, empty_slots. Thanks in advanced. |
Beta Was this translation helpful? Give feedback.
-
|
@chrisstem98 to return results in the command line without modifying |
Beta Was this translation helpful? Give feedback.
-
|
Hi, when i run from ultralytics import solutions
solutions.ParkingPtsSelection()get the error but interface running
i'm sure that my environment have tkinter. So I disable check_requirements on line no 16 of anaconda3/envs/ultralytics-env/lib/python3.8/site-packages/ultralytics/solutions/parking_management.py Next issue, when try to upload image file, the file on disk not listed on open dialog. I change line no 62 of anaconda3/envs/ultralytics-env/lib/python3.8/site-packages/ultralytics/solutions/parking_management.py from this self.image_path = filedialog.askopenfilename(filetypes=[("Image Files","*.jpeg";"*.jpg";"*.bmp")])to this self.image_path = filedialog.askopenfilename(filetypes=[('Image Files',('*.jpeg','*.jpg','*.bmp'))])and result is OK, file list showing on openfile dialog an I can start to draw the the boundingbox. maybe any wrong in my environment? here is result of my yolo check! Ultralytics 8.3.3 🚀 Python-3.8.19 torch-2.4.1 CUDA:0 (Quadro P2000, 5043MiB) OS Linux-6.8.0-45-generic-x86_64-with-glibc2.17 numpy ✅ 1.24.3<2.0.0,>=1.23.0 |
Beta Was this translation helpful? Give feedback.
-
|
@ali065119057 it seems you've resolved the issue with the file dialog by adjusting the code. Your environment appears correctly set up. If you encounter further issues, consider checking for any additional dependencies or updates. For more details on parking management, you can refer to the Ultralytics documentation. |
Beta Was this translation helpful? Give feedback.
-
|
I wanna ask if can i use segmentation instead of detection in implementing parking management solution? The thesis we are proposing is detection of motorcycle improper parking in parking areas. basically, improper parking gets detected when the motorcycle overlaps with the bounding box of the parking slot. |
Beta Was this translation helpful? Give feedback.
-
|
@expen6ve yes, you can use segmentation for detecting improper parking by identifying overlaps between motorcycles and parking slot boundaries. This approach can provide more precise localization compared to detection. For more details, you can refer to our parking management guide. |
Beta Was this translation helpful? Give feedback.
-
|
You want to use for motorcycle parking ?
…On Tue, Oct 29, 2024, 19:03 expen6ve ***@***.***> wrote:
I wanna ask if can i use segmentation instead of detection in implementing
parking management solution? The thesis we are proposing is detection of
motorcycle improper parking in parking areas. basically, improper parking
gets detected when the motorcycle overlaps with the bounding box of the
parking slot.
—
Reply to this email directly, view it on GitHub
<#11093 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A6MTUN3RYRQNMKV2QCULBATZ56IURAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCMBYG43TSNY>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Yes. |
Beta Was this translation helpful? Give feedback.
-
|
@expen6ve yes, you can use segmentation for detecting improper motorcycle parking by identifying overlaps with parking slot boundaries. For more details, please visit our guide on parking management: https://docs.ultralytics.com/guides/parking-management/. |
Beta Was this translation helpful? Give feedback.
-
|
Do you have any tips on how do i approach on that solution? |
Beta Was this translation helpful? Give feedback.
-
|
To approach your solution for detecting improper motorcycle parking, consider using segmentation to accurately identify and analyze overlaps with parking slot boundaries. This method can enhance precision in detecting parking violations. For further insights, refer to our parking management guide. |
Beta Was this translation helpful? Give feedback.
-
|
Yes sure I will share with you tomorrow
…On Wed, Oct 30, 2024, 17:03 expen6ve ***@***.***> wrote:
Do you have any tips on how do i approach on that solution?
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A6MTUN4GHN53ABLLYFF3NULZ6DDH7AVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCMBZHE2TCNI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Where can i possibly contact you in private? |
Beta Was this translation helpful? Give feedback.
-
|
@expen6ve thank you for reaching out. For any inquiries or support, please use the public discussion forums or GitHub issues. We do not offer private support. For more information on parking management with Ultralytics YOLO, please visit our guide. |
Beta Was this translation helpful? Give feedback.
-
|
+923125094307
Text me on what's app or on LinkedIn where you want
…On Wed, Oct 30, 2024, 21:17 expen6ve ***@***.***> wrote:
Where can i possibly contact you in private?
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A6MTUN2KXHOCXO6RKPQTHG3Z6EBCXAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCMJQGI2DIMA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
@SHAHFAISAL80 for any inquiries or support, please continue the discussion here on GitHub. We do not provide private support via WhatsApp or LinkedIn. Thank you for understanding. |
Beta Was this translation helpful? Give feedback.
-
|
Okay brother I got it I will participate here
…On Thu, Oct 31, 2024, 02:49 Glenn Jocher ***@***.***> wrote:
@SHAHFAISAL80 <https://github.com/SHAHFAISAL80> for any inquiries or
support, please continue the discussion here on GitHub. We do not provide
private support via WhatsApp or LinkedIn. Thank you for understanding.
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A6MTUNYK7UIQIP3NOUNBZVDZ6FH45AVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCMJQGU2DAMI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
Thank you for your interest in participating here. For any further inquiries or support, please continue the discussion on GitHub. We do not provide private support. |
Beta Was this translation helpful? Give feedback.
-
|
yeah don't worry i also believe everybody participate here inside of
personally
…On Wed, Oct 30, 2024 at 9:48 PM Glenn Jocher ***@***.***> wrote:
Thank you for your interest in participating here. For any further
inquiries or support, please continue the discussion on GitHub. We do not
provide private support.
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A6MTUN4DBZSNVXCBFT4LAWLZ6GZBRAVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCMJQG4ZTQNY>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
@SHAHFAISAL80 thank you for your engagement. For any further questions or discussions, please continue here on GitHub. We do not offer private support. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for your response.
…On Thu, Oct 31, 2024 at 4:59 PM Glenn Jocher ***@***.***> wrote:
@SHAHFAISAL80 <https://github.com/SHAHFAISAL80> thank you for your
engagement. For any further questions or discussions, please continue here
on GitHub. We do not offer private support.
—
Reply to this email directly, view it on GitHub
<#11093 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A6MTUNY5OHY3OFFOQHJMDHTZ6ILS3AVCNFSM6AAAAABHCDMGSGVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCMJRGA3TENQ>
.
You are receiving this because you were mentioned.Message ID:
***@***.***
com>
|
Beta Was this translation helpful? Give feedback.
-
|
@SHAHFAISAL80 thank you for your interest. For any questions or discussions, please continue here on GitHub. We do not offer private support. For more information on parking management with YOLO11, you can refer to the Ultralytics documentation. |
Beta Was this translation helpful? Give feedback.
-
|
Hi! what would be needed for a live feed of the parking management system? I have a thesis and this would be perfect but Im having trouble implementing my live webcam feed for a demo parking environment, any help would be great. Thanks! |
Beta Was this translation helpful? Give feedback.
-
|
@Skadadle for implementing a live webcam feed for a parking management system, you can use the |
Beta Was this translation helpful? Give feedback.
-
|
would a hikvision 4mp poe bullet type camera suffice to run this on a 3500x cpu, 1660gpu and 16gb ram? |
Beta Was this translation helpful? Give feedback.
-
|
@expen6ve yes, your setup with a Hikvision 4MP PoE camera, a Ryzen 3500X CPU, GTX 1660 GPU, and 16GB RAM is sufficient to run YOLO11 for parking management tasks. Ensure your camera's resolution does not exceed the supported 1920x1080 input size for optimal performance. For further details, refer to the Parking Management guide. |
Beta Was this translation helpful? Give feedback.
-
guides/parking-management/
Parking Management System Using Ultralytics YOLO11, Ultralytics YOLOv8, YOLOv9, YOLOv10
https://docs.ultralytics.com/guides/parking-management/
Beta Was this translation helpful? Give feedback.
All reactions