README.md

Object Detection Using YOLO with Self-Trained Data

This repository provides a guide on setting up and using YOLO (You Only Look Once) for object detection using self-trained data and a webcam. YOLO is a popular object detection algorithm that can efficiently detect objects in real-time.

Prerequisites

• Python (>=3.6)
• YOLOv5: This guide uses YOLOv5 for object detection. Install it via:

  pip install yolov5

Getting Started

1. Clone the Repository:

   Clone this repository to your local machine:

   git clone https://github.com/Hk669/yolov5.git
   cd cap-detection-yolo-webcam

2. Prepare Your Custom Dataset:

   Collect and annotate images for your custom object detection task. You'll need to create annotations in YOLO format. Use labelImg for labeling the images

3. Training:

   Train the YOLO model on your custom dataset. You can follow the YOLOv5's official training guidelines.

4. Export Model:

   After training, export the trained YOLO model.

5. Webcam Object Detection:

Run the webcam:

 cap = cv2.VideoCapture(0)
 for label in labels:
     print('Collecting images for {}'.format(label))
     time.sleep(5)

     # loop through images
     for img_num in range(number_images):
         print('Collecting images for {}, image number {}'.format(label,img_num))
 
         # webcam feed
         ret, frame = cap.read()
 
         # naming out image path
         imgname = os.path.join(IMAGES_PATH, label+'.'+str(uuid.uuid1())+'.jpg')
 
         # writes image to file
         cv2.imwrite(imgname, frame)
         
         # render to the screen
         cv2.imshow('Image collection', frame)
         time.sleep(2)
         if cv2.waitKey(10) & 0xFF == ord('q'):
             break
 cap.release()
 cv2.destroyAllWindows()

Project Structure

cap-detection-yolo-webcam/
│
├── data/                # Custom dataset and annotations
│   ├── images/
│   └── labels/
│
├── weights/             # Trained model weights
│
├── dataset.yml  # Webcam object detection script
|          
│
└── README.md

Acknowledgments

• This project is built upon the YOLOv5 repository: YOLOv5 GitHub

License

This project is licensed under the MIT License - see the LICENSE file for details.

---