tanabge6/Vehicle-Recognition-System-in-Tunisia

# Vehicle-Recognition-System-in-Tunisia This project is about developping a vehicle recognition system in Tunisia. It demonstrates the outcome of the summer internship (2 months) for second year students of the National School of Computer Science. **Problem setting :** The number of vehicles in the world is increasing rapidly every year, particularly in Tunisia. Therefore, several difficulties will be found to discover the vehicles identities when there is some problems such as parking violation enforcement, highway toll payment and also finding stolen cars. Added to that, it is impossible sometimes to identify the vehicle’s owner who escaped from authorities after causing an accident. This fact leads to develop an autonomous licence plate recognition system to identify vehicles and facilitate traffic management which is a popular and active research topic in the field of computer vision, image processing and intelligent transport systems. **Main Goals :** This project aims to solve the problem of vehicle identification in Tunisia using Deep Learning models. To do that, our mission is to find a solution based on DL models and image processing to accomplish the following tasks : - Detecting the presence of the vehicle’s licence plate. - Image segmentation and extraction of the text written on the licence plate. - Text classification with deep learning. - Development of a desktop application (UI) using .Net 📍 **Keywords :** Python, Computer Vision, Deep Learning, Image segmentaton, Yolo, Faster-RCNN, Tensorflow, Keras, OpenCV, c# (.Net) 📓 **Paper** : Report.pdf or Link . :round_pushpin: Requirements ```shell !pip install -r Deep-Learning/requirements.txt ``` 📍 Main scripts In the AI part of this project, we have : 1- Vehicle license plate detection using Yolo : The licence plate detection folder contains these necessary files : **lapi.weights , darknet-yolov3.cfg , classes.names** which will be called by for the ```object_detection_yolo.py``` script. …