Automation Lab, Sungkyunkwan University

Lane Detection for CCTV-cameras using Yolov8 + SegmentAnything + Adaptive Hough Transform with UI

Prepare for an updated version, include:

• UI for the CCTV lane detection program

Introduction

• Lane Detection method for CCTV cameras using Yolov8 + SegmentAnything + Adaptive Hough Transform

Installation

Install conda and activate the environemnt

mkdir -p ~/miniconda3
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3 
rm -rf ~/miniconda3/miniconda.sh
~/miniconda3/bin/conda init bash

conda create -n cctv_lane_ui
conda activate cctv_lane_ui

Install pytorch

conda install pytorch torchvision pytorch-cuda=12.1 -c pytorch -c nvidia

Install sahi -> PyQT6 -> boxmot -> ultralytics

pip install sahi
pip install PyQt6
pip install boxmot
pip install ultralytics

Install Segment Anything

pip install git+https://github.com/facebookresearch/segment-anything.git

Create folder as the following structure

Edge-TSS
├── src
│   ├── dataset
│   │   ├── selected_vid
│   ├── results
│   │   ├── bg_sub
│   │   ├── grid
│   │
│   │  

Getting Started

Results

Reproduce results using F1 score metric.

CCTV-Camera Visualization

Acknowledgement

• open-mmlab/mmdetection
• pytorch/vision
• Turoad/lanedet
• ZJULearning/resa
• cfzd/Ultra-Fast-Lane-Detection
• lucastabelini/LaneATT
• aliyun/conditional-lane-detection