Title :
Research and design of digital validation image recognition algorithm based on contour features
Author :
Kang, Chun-ying ; Zhang, Wei ; Zhang, Tao
Author_Institution :
Heilongjiang Univ., Harbin
Abstract :
This paper proposes a simple and fast digit recognition algorithm based on exterior contour features of a digital validation image. First, we convert the validation image into a binary one and divide the binary image into some single number image. Then compute the exterior contour features of the single image from top, bottom, left and right 4 directions. Finally, we recognize the number according to their contour features. The test results show that the recognition accuracy of this algorithm is relatively high and the speed is fast. In addition, it can be used to recognize other kinds of characters.
Keywords :
edge detection; binary image; digital validation image recognition algorithm; exterior contour features; first-order differential; Algorithm design and analysis; Color; Cybernetics; Histograms; Image converters; Image recognition; Image segmentation; Machine learning; Machine learning algorithms; Testing; Contour features; First-order differential; Number recognition;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620925