The extraction of characters from cover images using mathematical morphology
Lixu Gu, Naoki Tanaka, Toyohisa Kaneko, R.M. Haralick · Systems and Computers in Japan · 1998
Since characters are composed from thin and long lines, the character regions in an image can be easily detected and extracted according to their thin and long features. In this paper, we propose the new approach of detecting regions with different widths (widths of character lines) so as to extract characters from cover images using mathematical morphology. The proposed approach is efficient for more complicated images where characters are embedded in a complex background. Our experimental results for extraction of characters from cover images show that for gray-scale images, 81% of the characters are correctly extracted, while 95% of the characters are extracted from color images. The characters that are smaller than our smallest structure elements cannot be extracted so far owing to the low resolution of our scanned images. If this problem could be solved, the extraction rates should be 92% and 100%, respectively. © 1998 Scripta Technica, Syst Comp Jpn, 29(4): 33–42, 1998