Research And Implementation Of Handprint Recognition Method Based On Measure Of Medium True Degree

Biological recognition technology is more and more widely used for its unique stability, uniqueness and convenience. Among these, as an emerging biometric technology, the palm print recognition has the characteristics of simple sampling, rich image information, high user acceptance, not easy to forge, less noise interference and so on, it has been widely concerned by researchers at home and abroad. The fingerprint shape controlled by genes, can be considered irregular, but a texture in the one and only among individuals. The palm print contains rich features, line features, texture features, geometric features and the diversity of information. Because of the complexity and diversity of handprint and ambiguity, at present, based on the current research and application of biometric technology in the hand there are still some problems to be solved.Based on in-depth research and analysis of the image of hand, the intermediary true value measurement(measure of medium truth degree, MMTD) is introduced into the palm print image filtering, edge detection, amplification, localization, recognition and so on. The main research contents are as follows:(1) Study the image filtering method based on MMTD. According to the characteristics of the noise, correspond the gray level of the normal pixels and the noise points to the corresponding logical true value interval, distance ratio function is used to remove noise of the image, the experimental results verify the validity of the method.(2) Based on the analysis of characteristics of fingerprints image edge, through the comparison of the center pixel and around four direction pixel gray value, improved MMTD palm print image edge detection method based on. Simulation results show that this method can effectively extract the fingerprint image edge information.(3) Study on the image recognition method based on MMTD fingerprints similar.(4) At the end of this paper, design and implement a palm print recognition system based on MMTD. After positioning the key points of palm print image、handprint filter based on MMTD、hand edge detection and processing based on MMTD, principal component analysis(PCA) was used to extract the feature of feature region. MMTD similarity criterion was used to identify palm print, the experimental results show that the better recognition effect of hand.