Vehicle-License-Plate-Recognition

Overview

This repository contains the code and documentation for a license plate segmentation system based on the YOLOv8 model. The system is designed to accurately detect and segment license plates from various traffic scenes, providing a crucial input for intelligent traffic systems.

Dataset

The system utilizes two datasets for training and testing:

CCPD2020: Contains 11,774 images of new energy vehicle license plates.
CCPD2019: Consists of two subsets with 20,967 images of regular fuel vehicle (blue plate) and 25,467 images of fuel vehicles with higher recognition difficulty (blurry, high tilt, etc.).

Data Preprocessing

The plate_dataset_preprocess.py script is responsible for preparing the CCPD datasets for training the YOLOv8 model. The script performs the following tasks:

Extracts车牌位置信息 from image filenames.
Creates corresponding label (.txt) files for each image, including class labels (blue plate as 1, green plate as 0), plate center coordinates, and plate dimensions.
Splits the dataset into training, validation, and testing sets in a 5:3:2 ratio.

Model Training

The model.py script is used to train and use the YOLOv8 model on the preprocessed dataset. The training process involves:

Loading the pre-trained YOLOv8 model.
Adjusting hyperparameters such as learning rate, batch size, and number of epochs.
Monitoring loss functions to ensure optimal model performance.