Research on Character Segmentation Method in Image Text Recognition

Abstract:

Character Segmentation is the key step for image text recognition. This paper presents a text tilt correction algorithm using tracked characteristics rectangle contour to extract angle, using line scan method based on the number of transitions to determine the character on the bottom. In order to meet the requirements of real-time and reliability, takes improved secondary single-character segmentation algorithm based on vertical projection method.

• Periodical: Advanced Materials Research (Volumes 546-547)
• Pages: 1345-1350
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.546-547.1345
• Online since: July 2012
• Authors: Lian Huan Li
• Keywords: Character Segmentation, Image Recognition, Text Recognition, Tilt Correction

References

[1] Yu Shiqi, Rui Zhen. Learning OpenCV (Chinese version) [M] Beijing: Tsinghua University Press, 2009. 10: 186.

[2] Gong Lixia, Bai Yanping. Radon Transform in the tilt plate image correction [J]. Test and Measurement Technology, 2009, 23 (5): 452-456.

[3] Du Yanhong, Zhang Weiyu, often Kwai such as the improvement of the line detection algorithm based on Hough transform [J]. Tianjin Agricultural College, 2007. 6, 14 (2) : 34-36.

[4] limits and based on digital image processing technology, Automatic Vehicle detection and segmentation algorithm [D]. Electronic Science and Technology University, 2007. 8: 14 -15.

[5] Yao Lei license plate recognition system software design and [D] Shanghai Jiaotong University, 2009. 12: 4.

[6] Wong Man-kit. Projection-based license plate character segmentation method [J]. Modern computer, 2009, 8: 57 -60.

[7] Hisch.J. W, Hu.W. F, Chang C.J. Shadow elimination for effective moving object detection by Gaussian shadow modeling[J]. International Journal of Image and Vision Computing. 2003, 21(6): 505-516.

DOI: 10.1016/s0262-8856(03)00030-1

[8] Yang. F, Ma. Z, Xie.M. A novel approach for license plate character segmentation [C]. The 1st IEEE Conference on industronics and Applications, Singapore, 2006, Institite of Electrical and Electronics Engineers Inc, l-6.

DOI: 10.1109/iciea.2006.257234

[9] Jiao Tingting, Hou Dewen. Analysis of character segmentation of license plates [J]. Shandong science, 2008, 21(2): 39-41.

[10] Fan Weiqi, Mu Changjiang. A method based on Chinese characters structure features of vehicle license plate character segmentation method [J]. Chinese Journal of scientific instrument, 2003, 24(4): 472-474.

[11] Xie Weisheng.

[11] The license plate location and character segmentation algorithm research and implementation [D]. Southwest Jiao Tong University, 2010. 6.

[12] Song Chenguang, Ye Haijian. The license plate character segmentation based on mathematical morphology algorithm [J]. Optoelectronic technology and information, 2005. 18, 18(6): 58-62.

[13] Wang Runmin, Qian Shengyou. Based on neural network and color characteristics of the license plate character segmentation method [ J]. Computer engineering and applications, 2006, 26: 57-59118.

[14] Zheng Wei, Zhao Hong Chan. The character image segmentation based on wavelet transform [J ]. Shanxi Electronic Technology, 2008, 3: 15-16.