Low Quality Finger Vein Image Enhancement

With the rapid development of computer and network technology, the information security becomes more and more important. People hope that a more convenient and reliable method for identity authentication will be found. The biometrics makes this become possible. Fingerprint recognition is considered as the comparative maturity technology in biological feature recognition area. However, applicable area of fingerprint recognition has been restricted because it is easy to be forged, destroied and contaminated. Finger vein is in the body which has many merits such as living body recognition, interior characteristic, non-contacting and high safety and so on. So it can resolve effectively many problems of traditional biometrics. This dissertation studys low quality finger vein image enhancement. It is a pretreatment period and very important in the image processing area. Its function is that it not only can improve quality and effect of image but also make the image be observed easily by people and recognized effectively by machines.Main methods are as follows:(1) The Histogram Qqualization is a classical enhancement algorithm. Fist, some reasons of the Histogram Qqualization gray-level coalition are found by the study and analysis. And then, the merit and disadvantage for image enhancement are found. At last, A finger vein image enhancement algorithm based on advanced Histogram Qqualization is proposed. The experiment results show that it is effective not only in enhancing vein image but also in removing noise and false vein feature.(2) A finger vein image enhancement algorithm based on fuzzy theory is proposed for low quality finger vein image enhancement. It combined traditional fuzzy theory with the classical image enhancement algorithm. Different vein images are applied to test. It is proved that it is effective in vein feature enhancement by the muti-threshold method and it can advance recognition precision. In a word, The method is effectual and applicable for low quality finger vein image enhancement.(3) The disadvantage of the classical Morlet Wavelet is found by analysis and study by low quality finger vein image enhancement. Then, a advanced Morlet Wavelet image enhancement algorithm is proposed. The experiment results show that it is effective not only in vein image enhancement but also in removing noise and false vein feature. At the same time, it can improve the recognition precision of the whole system and the speed of image process. (4) In this dissertation, all experiment results reveal that the muti- threshold fuzzy enhancement algorithm is the best for low-quality finger vein image enhancement because it can improve the recognition precision best. But, Wavelet will be a progress area and studied for this paper. It is hoped that more methods based on Wavelet which can advance recognition precision are proposed in the future.