Study On Fingerprint Image Enhancement And Preprocessing Algorithms

Computer fingerprint recognition technology has been used in many fields. Especially in the investigation of criminal cases, it has been the essential equipment in such institutions and applied in security checking, office fingerprint card punching, car lock and bank coffer etc. Since there are some practical problems in recognition precision, speed and reliability, its application is limited greatly. Actually, the main function of fingerprint recognition in the investigation of criminal cases is to extremely reduce the search scope of the fingerprint database, the final recognition is mainly finished by fingerprint specialists. One of the important elements that influence the recognition efficiency is that the preprocessing of the fingerprint image in recognition prophase can't provide reliable characteristic description for recognition. Moreover, fingerprint image information is also a very important vision information source of web multimedia field, such as encryption of web works, electronic commerce and security authentication of electronic bank.The process of computer vision can be divided into primary stage and advanced stage. The primary stage provides basic element and its description for recognition to the advanced stage and the advanced stage completes the object recognition. The primary stage is also called preprocessing stage. A good output in the primary stage is the basis of a successful recognition. This paper discusses some improved methods to the preprocessing of fingerprint recognition, and it also does some research work on fingerprint image enhancement for manual vision discrimination.Firstly, because of the low quality input fingerprint image, later processing wastes time and the recognition precision decreases, this paper gives a quality assessment algorithm of fingerprint image. Secondly, a fingerprint image enhancement method based on Gabor filter is presented and realized. This method estimates the orientation and orientation certainty at each pixel, computes the average ridge frequency, and adjusts the parameters of Gabor filter using the orientation and average ridge frequency at each pixel to realize adaptive filtering. The unrecoverable region is segmented basedon the orientation certainty. On this foundation, the patent of Dalian Everspy Image Technology Corporation is applied in fingerprint image enhancement and a Gabor filter enhancement method based on sector subarea is presented. Finally, a fingerprint image description method based on finger code is provided and realized, and this paper discusses some application of this method in fingerprint recognition field.