Research On Extraction Algorithm For Check Image Information

In the oriental countries, such as China, Japan and Korea, seals are extensively used for the purpose of entity identification. In China, seals play a more important role. With the development of social economy, the portfolio of the bank increases day by day. So the traditional method of manual check seal identification cannot readily meet the needs of banking transactions. While there is only a single criterion in the existing bank seal automatic recognition system of market with the deficiencies of low precision. Thus, carrying on further research of extraction and registration algorithm for check seal identification has higher practical meaning and theoretical meaning.Based on the collection and analysis of relevant studies and papers based on the combined knowledge of image processing and pattern recognition, focusing on the extraction of check images and check seal image information. We first check seal on the complex characteristics of the image analysis. In order to eliminate the impact of back conditions, and check their background floral patterns, fonts and other written or printed form of noise disturbance, this paper, color component transform, gray linear, median filtering, Otsu method, morphological closing operation and labeling law series method process seal on the check image is preprocessed, get help to identify the two values of low-noise images.Seal extract, after winning the second value based on the seal image, this paper, seed dispersal, hole filling algorithm and improved multistage Hough transform circle detection methods and to use its seal of feature extraction of the reserve information as prior knowledge a check in the round seal and turn to extract a square seal. The seal image extraction process, to correct the existing methods are mostly processed for the shortcomings of a single seal.This article also check the logo of bank, check number, bank institution code, the amount of information in the case of extraction. First, the location of positioning based on information cut, color components using the threshold to remove seals, shading and other noise interference, use of horizontal and vertical projection of the information on the target character cut, get a single binary character image, easy identification later. After the large number of check processing and extraction of samples obtained more satisfactory results.