Vehicle Identification and Inspection System to Improve Security in Restricted Access Areas

Abstract

The growing affluence of time has made the ownership of vehicles a necessity. This has resulted in an unexpected civic problem that is traffic control, vehicle theft, unavailable parking. As number plate is the unique identification of the vehicle it can be detected and recognized. Different methods have been developed to detect and recognize license plates. Due to the characteristic of the license plate that varies from country to country in terms of font and size, font style, color. For this reason, the detection and recogni- tion of license plates in different condition and several climatic conditions remains always dfficult to achieve good results. The goal is to design an efficient automatic vehicle iden- tification system to improve security control at the entrance of any restricted area. The proposed approach includes license plate recognition along with vehicle logo recognition for enhancing security. You Only Look Once(YOLO) and U-net techniques are imple- mented for detection and segmentation respectively. To recognize all upper case letters (A-Z) and digits (0-9), an extracted character is fed into the deep architecture. Local Binary Pattern(LBP) and template matching methods are implemented for vehicle logo localization and recognition.