期刊名称：International Journal of Image Processing (IJIP)
电子版ISSN：1985-2304
出版年度：2009
卷号：3
期号：5
页码：252-264
出版社：Computer Science Journals
摘要：This paper presents a real-time and robust method for license plate location and recognition. After adjusting the image intensity values, an optimal adaptive threshold is found to detect car edges and then algorithm used morphological operators to make candidate regions. Features of each region are to be extracted in order to correctly differentiate the license plate regions from others. It was done by analysis of percentage of Rectangularity of plate in decision system .usage of color filter makes algorithm more robust on LPL, too. The algorithm can efficiently determine and adjust the plate rotation in skewed images. It finds the optimal adaptive threshold corresponding to the intensity image obtained after adjusting the image intensity values. To segment the character of the license plate, a segmentation algorithm base on profile is proposed. An optical character recognition (OCR) engine has then been proposed. The OCR engine includes characters dilation, resizing input vector of ANN. To recognize the characters on the plates, MLP neural networks have been used and compared with Hopfield, LVQ and RBF. The results show that MLP outperforms. According to the results, the performance of the proposed system is better even in case of low-quality images or in images with illumination effects and noise
