Study On Algorithm For Infrared Hand Vein Image Identification

Biometric identification technology based on hand vein subcutaneous networkstructure appears as a promising technique for personal recognition.According to the characteristic of the infrared light that while the infrared lightis absorbed intensively by the blood, it is dispersed by other organs of the body. Theaverage infrared lighting is adopted, and the infrared vein image containinginformation of hand vein can be acquired by optical CMOS sensor.The contrast of hand vein image is enhanced by using Wiener filter, normalizedmode, gray level stretching, sharpening and so on. After binary image is obtained,four-domain region growing algorithm is presented to differentiate and get rid of thenoise areas. The quick thinning algorithm is adopted, which have many advantagessuch as complete thinning, quick speed, maintenance of minutiae, and so on. Afterhand vein image skeleton is analyzed, the method of extracting the crossing points isproposed to solve the problems meeting in the feature extraction. The burr other kindof noise is introduced by thinning algorithm. On the basis of the burr of the handvein image, an algorithm for rebutting is presented. Through experiments, the newalgorithm could make the skeleton express geometric structure of the hand veinimage.On the basis of geometric structure of the hand vein image, an algorithm forbiometric identification is presented, which can easily express topology structure ofthe hand vein image. Through experiments, image change such as horizontal moving,circumrotation and resizing has no impact on the hand vein identification algorithm.