This Article 
 Bibliographic References 
 Add to: 
Recognizing Characters in Scene Images
February 1994 (vol. 16 no. 2)
pp. 214-220

An effective algorithm for character recognition in scene images is studied. Scene images are segmented into regions by an image segmentation method based on adaptive thresholding. Character candidate regions are detected by observing gray-level differences between adjacent regions. To ensure extraction of multisegment characters as well as single-segment characters, character pattern candidates are obtained by associating the detected regions according to their positions and gray levels. A character recognition process selects patterns with high similarities by calculating the similarities between character pattern candidates and the standard patterns in a dictionary and then comparing the similarities to the thresholds. A relaxational approach to determine character patterns updates the similarities by evaluating the interactions between categories of patterns, and finally character patterns and their recognition results are obtained. Highly promising experimental results have been obtained using the method on 100 images involving characters of different sizes and formats under uncontrolled lighting.

[1] T. Kawatani, T. Tsutsumida, S. Akamatsu, and N. Nakajima, "A new precise recognition method for handprinted Kanji," inProc. CVPR'85, 1985, pp. 538-543.
[2] I. Masuda, N. Hagita, T. Akiyama, T. Takahashi, and S. Naito, "Approach to smart document reader system," inProc. CVPR'85, June 1985, pp. 550-557.
[3] Y. Nakamuraet al., "An optical character recognition system for industrial applications: TOSEYE-1000," inProc. Int. Workshop Industrial Applications of Machine Vision and Machine Intelligence, 1987, pp. 364-368.
[4] M. Takatooet al., "Gray scale image processing technology applied to vehicle license number recognition system," inProc. Int. Workshop Industrial Applications of Machine Vision and Machine Intelligence, 1987, pp. 76-79.
[5] L. C. Howington, "Automated license plate reading,"Advanced Imaging, vol. 4, no. 8, pp. 46-50, Sept. 1989.
[6] H. Hoshinoet al., "A method for extracting specific characters from complex background," inNat. Convention Rec., IECE Japan, vol. 1152, 1983 (in Japanese).
[7] K. Funakoshiet al., "Study of character extraction from scene images," inNat. Convention Rec., IEICE Japan, vol. 1503, 1987 (in Japanese).
[8] A. Perez and R. C. Gonzalez, "An iterative thresholding algorithm for image processing,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-9, no. 6, Nov. 1987.
[9] K. V. Mardia and T. J. Hainsworth, "A spatial thresholding method for image segmentation,"IEEE Trans. Pattern Anal. Machine Intell., vol. 10, no. 6, pp. 919-927, Nov. 1988.
[10] T. Taxt, P. J. Flynn, and A. K. Jain, "Segmentation of document images,"IEEE Trans. Pattern Anal. Machine Intell., vol. 11, no. 12, pp. 1322-1329, Dec. 1989.
[11] J. Ohya and A. Shio, "A study of recognizing characters in scene images,"Trans. IEICE Japan, vol. J71-D, no. 6, pp. 1037-1047, June 1988 (in Japanese).
[12] J. Ohya, A. Shio, and S. Akamatsu, "A relaxational extracting method for character recognition in scene images," inProc. CVPR'88, June 1988, pp. 424-429.
[13] C. K. Chow and T. Kaneko, "Automatic boundary detection of the left ventricle from cineangiograms,"Comput. Biomed. Res., vol. 5, pp. 388-410, 1972.
[14] S. W. Zucker, "Region growing: Childhood and adolescence,"Comput. Graphics Image Processing, vol. 5, no. 3, pp. 382-399, 1976.
[15] N. Otsu, "A threshold selection method from gray-scale histograms,"IEEE Trans. Syst. Man Cybern., vol. SMC-9, no. 1, pp. 62-66, 1979.
[16] J. Ohya and A. Shio, "Detection of character candidate regions from scene images," inNat. Convention Rec., IECE Japan, vol. S13-1, 1986 (in Japanese).
[17] M. Umeda, "Recognition of multi-font printed Chinese characters," inProc. 6th Int. Conf. Pattern Recognition, 1982, pp. 793-796.
[18] A. Rosenfeld, R. A. Hummel, and S. W. Zucker, "Scene labeling by relaxation operations,"IEEE Trans. Syst. Man Cybern., vol. SMC-6, no. 6, pp. 420-433, 1976.

Index Terms:
optical character recognition; image segmentation; character recognition; scene images; image segmentation; adaptive thresholding; multisegment characters; relaxational approach
J. Ohya, A. Shio, S. Akamatsu, "Recognizing Characters in Scene Images," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 16, no. 2, pp. 214-220, Feb. 1994, doi:10.1109/34.273729
Usage of this product signifies your acceptance of the Terms of Use.