License Plate Detection and Segmentation Using Cluster Run Length Smoothing Algorithm

For the different types of license plates being used, the requirement of an automatic license plate recognition system is different for each country. In this paper, an automatic license plate detection system is proposed for Malaysian vehicles with standard license plates based on image processing and clustering. Detecting the location of license plate is a vital issue when dealing with uncontrolled environments and illumination difficulty. Therefore, a proposed algorithm called Cluster Run Length Smoothing Algorithm (CRLSA) was applied to locate the license plates at the right position. CRLSA consisted of two separate proposed algorithms which applied run length edge detector algorithm using 3 3 × kernel masks and 128 grayscale offset plus a threedimensional way to calculate run length smoothing algorithm, which can improve clustering techniques in segmentation phase. Six separate experiments were performed; Morphology, CRLSA, Clustering, Square/ Contour Detection, Hough, and Radon Transform. From those experiments, analysis based on segmentation errors was constructed. The prototyped system has accuracy more than 96%. DOI: 10.4018/jitr.2012070103 Journal of Information Technology Research, 5(3), 46-70, July-September 2012 47 Copyright © 2012, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. tion (LPD) were for contour detection (Han, Han, Wang, & Zhai, 2003), Hough transform (Soh, Chun, & Yoon, 1994), Radon transform (Shapiro, Gluhchev, & Dimov, 2006), Morphology (Xu & Zhu, 2007), clustering (Siah, 2000; Abdullah, Khalid, Yusof, & Omar, 2007; Abdullah et al., 2010; PirahanSiah, Abdullah, & Sahran, 2011; Abidin, Abdullah, Sahran, & PirahanSiah, 2011) and Speed Up Robust Features (SURF) (Bay, Tuytelaars, & Van Gool, 2006). However, this paper will elaborate on the theories and applications of only some of the mentioned techniques for object detection. Later, these techniques are experimented and compared. The rest of the paper is organized as follows. Related works and proposed method are discussed in subsequent sections. Then, we justify our proposed work in the Results and Analysis section. This paper concludes our findings in the Conclusion section.