Description
This project achieves some functions of image identification for Self-Driving Cars.
First, use yolov5 for object detection whose class includes car, truck, pedestrian, bicyclist, traffic light, traffic sign, motor and large vehicle.
Second, crop the images of traffic light and traffic sign to execute the image classification respectively.
Furthermore, the GUI of this project makes it more user-friendly for users to realize the image identification for Self-Driving Cars.
For example: input source (e.g., Folder, Image, YouTube, DroidCam, WebCam), image display, parameter adjustment, information page, etc.
It is written in Python and uses Tkinter for its graphical user interface (GUI).
Software and Hardware Environment
| IDE (optional) | Visual Studio Code |
|---|---|
| Extensions | Python |
| Programming Language | Python |
| Python Version | Python 3.7.10 |
| Python Package | Refer to requirements.txt |
| GPU (preferred) | GTX 1080 Ti or higher |
Quick Start Examples
Install
Install Visual Studio Code and Python 3.7.10 required with all requirements.txt dependencies installed:
$ git clone https://github.com/JeffWang0325/Image-Identification-for-Self-Driving-Cars.git
$ cd Image-Identification-for-Self-Driving-Cars
$ pip install -r requirements.txtExecute GUI
- Step1: Install all models and images
- Step2: Move the files to the appropriate path
- Step3: Execute detect_Main_Jeff.py
GUI Demo & Report:
Please click the following figures or links to watch GUI demo videos or report:
自駕車影像辨識系統 (Image Identification for Self-Driving Cars using Python Tkinter )-English Version
專題報告: 自駕車影像辨識系統 (Image Identification for Self-Driving Cars)-HD
專題報告: 自駕車影像辨識系統 (Image Identification for Self-Driving Cars)
GUI Demo1 using Python Tkinter (Image Identification for Self Driving Cars)-中文版
GUI Demo2 using Python Tkinter (Image Identification for Self Driving Cars)-中文版
GUI Demo3 using Python Tkinter (Image Identification for Self Driving Cars)
※Outline:
0. Introduction
●4 Parts of GUI: Model Input Setting, Input Sources, Image Display, Information Page
1. Input Source - Folder
●Inside the folder, they can be either images or videos.
2. Input Source - Image
3. Information Page
※More Information:
●Image height and width
●Object detection result
●Computation time of Yolov5, traffic light and traffic sign
4. Input Source - YouTube
5. Input Source - DroidCam
●Write the IP Cam Access in the textbox.
6. Parameter Adjustment - Yolo v5
7. Parameter Adjustment - Sign (DL)
8. Information Page - Others
Contact Information:
If you have any questions or suggestions about code, project or any other topics, please feel free to contact me and discuss with me. 😄😄😄
