Digit recognition in a natural scene with skew and slant normalization

Abstract.This paper proposes a method to recognize digits in a natural scene, such as telephone numbers on a signboard. Candidate regions of digits are extracted from an image through contrast enhancement, edge extraction, and labeling. Since the target text patterns are in a 3D space, unlike traditional character recognition problems, we have to deal with the image transformation effect due to the orientation in the 3D space and projection. We have to cancel the effect as much as possible before digit recognition. In our method, the image transformation effect is modeled as skew and slant. In the proposed method, simplified Hough transform is used for the skew normalization. After the skew normalization, the remaining effect of image transformation is corrected by circumscribing digit patterns with tilted rectangles and affine transformation. In experiments, we tested a total of 1,332 images of signboards with 11,939 digits. We obtained a digit extraction rate of 99.2% and a correct digit recognition rate of 98.8%.

[1]  Yasuaki Nakano,et al.  Extraction of place-name from natural scenes , 2002, Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition.

[2]  Minoru Maruyama,et al.  Character extraction from natural scene images by hierarchical classifiers , 2004, ICPR 2004.

[3]  Majid Mirmehdi,et al.  Estimating the Orientation and Recovery of Text Planes in a Single Image , 2001, BMVC.

[4]  Hiroshi Sako,et al.  Information capturing camera and developmental issues , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[5]  Changming Sun,et al.  Skew and slant correction for document images using gradient direction , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[6]  Bidyut Baran Chaudhuri,et al.  Skew Angle Detection of Digitized Indian Script Documents , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  David S. Doermann,et al.  Progress in camera-based document image analysis , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[8]  Minoru Maruyama,et al.  A method to make multiple hypotheses with high cumulative recognition rate using SVMs , 2004, Pattern Recognit..

[9]  Hiromitsu Yamada,et al.  Effect of the perturbed correlation method for optical character recognition , 1993, Proceedings of 2nd International Conference on Document Analysis and Recognition (ICDAR '93).

[10]  Majid Mirmehdi,et al.  Recognising text in real scenes , 2002, International Journal on Document Analysis and Recognition.

[11]  Jin Hyung Kim,et al.  Color Texture-Based Object Detection: An Application to License Plate Localization , 2002, SVM.

[12]  Ching Y. Suen,et al.  Historical review of OCR research and development , 1992, Proc. IEEE.