AN EFFICIENT METHOD TO LOCATE LICENSE PLATES UNDER DIVERSE ENVIRONMENTS

Abstract

Advances in artificial intelligence (AI) have made car and road object identification more efficient. Researchers face a difficult problem in detecting license plates (LPD), which necessitates the use of a dependable and precise automatic detection method. Recent approaches to deep learning have showed promise, although they are frequently limited to specific locales or private datasets. This study focuses on accurately finding license plate locations using machine vision and deep learning approaches, especially under adverse weather situations such as blurriness, nighttime, and high brightness. A modified RCNN model is developed for LPD, which uses the China City Parking Dataset (CCPD) to forecast nine different stages. The suggested method includes data preprocessing, weight generation, and model training. The model is taught using self-generated weights rather than pre-learned weights, and it detects number plates with 98% accuracy under a variety of weather situations. After 50 epochs of training, the detection module predicts suitable bounding boxes. The Region Convolutional Neural Network (RCNN) is highly effective in detecting number plates under diverse environments.