License Plate Location Algorithm Based on Multiresolution Edge Detection
Because of low precision of license plate location in the color vehicle images with complex background, a new License Plate Location Algorithm was proposed based on multi-scale edge detection method with wavelet transform. Firstly, we perform WT on two di
- PDF / 402,050 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 83 Downloads / 229 Views
License Plate Location Algorithm Based on Multiresolution Edge Detection Zhang Baohua and Gu Yu
Abstract Because of low precision of license plate location in the color vehicle images with complex background, a new License Plate Location Algorithm was proposed based on multi-scale edge detection method with wavelet transform. Firstly, we perform WT on two directions respectively, and obtain the magnitude of the 2-D WT. Then detect the maxima magnitude of WT as the image edge and get edge image. At last, the projection approach based on two-way back is adopted to examine up-down and left-right boundary of the car license. Experimental results demonstrate that the proposed method can enhance the images location speed and accurate capture license plate region. Keywords License plate location • Multiresolution • Edge detection
268.1
Introduction
The automatic license plate recognition technique has important practical value in the intelligent transportation system, in which the key points are to locate license plate and recognize the license plate characters [1, 2]. The license plates must be located from the background vehicle images before the character recognition. This task is considered as the most crucial step in license plate recognition. A lot of algorithms are proposed to solve this problem [3–5]: algorithm based on car license color characters, algorithm based on texture analysis, projection algorithm based on gray features, algorithm based on BP networks and so on. Different algorithms faced a same problem: poor adaptability. Any LPR system must be able to cope with: poor image resolution, weather conditions and different illuminations,
Z. Baohua (*) • G. Yu School of Information Engineering, Inner Mongolia University of Science and Technology, BaoTou City, Inner Mongolia 014010, People’s Republic of China e-mail: [email protected] S. Zhong (ed.), Proceedings of the 2012 International Conference on Cybernetics 2103 and Informatics, Lecture Notes in Electrical Engineering 163, DOI 10.1007/978-1-4614-3872-4_268, # Springer Science+Business Media New York 2014
2104
Z. Baohua and G. Yu
non-uniform lighting across the plate and so on. This paper present a locate algorithm based on multi-resolution edge detection, multi-resolution analysis combined with projection algorithm will improve adaptability of the locate algorithm. The edge detection is a very important technique of image processing, as it can locate significant variations of gray images [6]. Edge detection is applied to many image analysis fields. Edge features of the transportation image are very helpful to recognize localization of license plate, and edge density can be used to successfully detect a number plate location due to the characteristics of the number plate [7]. The standard 2D WT is an efficient image processing tool too; it is good at isolating horizontal and vertical edges, track edges in the coarser images, extract more reasonable edges and avoids the noised or blurred influence [8]. So 2D WT applied to edge detection will deco
Data Loading...
