关键词: 身份证, 图像分块, 光照校正, 字符分割, 投影

Abstract:  In order to solve the difficulties of binarization caused by uneven illumination of the acquired images, the digital image uneven illumination correction algorithm based on the probability is applied to illumination correction for the ID card. This algorithm uses a statistical method when extracting the background. It first calculates the mean and standard deviation of the pixels’ brightness, then calculates the grayscale of the background, according to the statistical value. This is conducive to exclude the interference of noise points, and enhance robustness. The horizontal projection is used to determine the line position of the character, and the vertical projection is used to determine the position of the single character. Experimental results show that the method can effectively binarize and accurately segment characters on the ID card.

Key words: ID card, image partitioning, illumination correction, character segmentation, projection