License Plate Detection Based on Sparse Auto-Encoder

Abstract

In modern society, automatic license plate recognition (ALPR) plays an important role in the field of Intelligent Transport Systems (ITS). In order to recognize the license plate efficiently, the location of the license plate must be detected first. In consequence, the detection of the license plate becomes a crucial stage in an ALPR system, affecting the performance of the whole system enormously. In this paper, we propose a novel method based on Sparse Auto-Encoder (SAE) to detect the vehicle license plate. The proposed method consists of three main stages: (1) A block-based image segmentation technique used for dividing the image into several blocks. (2) Deep learning model (SAE) trained for candidate block selection. (3) Accurate extraction of the license plate. Unlike other existing license plate detection methods, the proposed algorithm use a deep learning model to learn the features of the license plate. Experiment results demonstrate that our method can detect various types of license plates with a high accuracy and a relatively short running time.