# Traffic-light-detection

**Repository Path**: liuchao01/Traffic-light-detection

## Basic Information

- **Project Name**: Traffic-light-detection
- **Description**: Detection of traffic signals as `stop` and `go` using YOLOv3 and openCV
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-06-13
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

### Steps to run the scripts

1. Download and convert the weights for yolov3

`wget https://pjreddie.com/media/files/yolov3.weights`

`python3 yad2k.py yolov3.cfg yolov3.weights yolo.h5`

2. Paste the images on which traffic light detection is to be performed in `images/test` directory

3. To view the results of YOLOv3 algorithms for object detection run script utils.py

`python3 utils.py`

4. To test traffic light detection run script red_light_detection.py

`python3 red_light_detection.py`

The results will be displayed in the terminal against the name of every test image, to view the cropped images of traffic light browse thwough the `images/res` folder