Hand dorsum vein image identification processing method and hand dorsum vein image identification processing device

Abstract

The invention provides a hand dorsum vein image identification processing method and a hand dorsum vein image identification processing device. Binarization processing of acquired hand dorsum vein images is carried out to acquire binary images including hand dorsum venous lines. The image integration processing of a plurality of rotated binary images is carried out along the vertical axes of the binary images, and a plurality of integral vectors are acquired. The normalization processing of the plurality of integral vectors is carried out to acquire the binary integral images of the binary images. The two-dimensional discrete cosine transform processing of the binary integral images is carried out to acquire low-frequency images, and to-be-identified characteristics are extracted from the low-frequency images. The binary images are rotated a plurality of times, and the image integration processing of the binary images is carried out after each time of rotation, and therefore the characteristic accuracy of the finally-acquired to-be-identified characteristics is higher, the characteristic stability thereof is stronger, and therefore the robustness of identification results of identity identification adopting the to-be-identified characteristics is higher.

Description

technical field [0001] The invention relates to an image processing technology, in particular to a method and device for recognizing and processing vein images on the back of the hand. Background technique [0002] Vein recognition is a biometric technology. Since the vein image on the back of each person's hand is unique and persistent like a fingerprint, it can be used for identity authentication. [0003] Generally speaking, in the existing hand vein image recognition process, the hand vein image is first processed in grayscale according to the artificially preset segmentation threshold, and then the feature information of the key points is extracted from the processed hand vein image. Specifically, the key points of the veins may be such as vein detection endpoints and vein intersections, and the specific feature information may be such as information such as quantity and relative position. Substituting the feature information of the key points obtained after the recogn...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem that the feature information obtained in the existing method for recognizing and processing dorsal hand vein images cannot accurately reflect the image information of dorsal hand veins, the present invention provides a method and device for recognizing and processing dorsal hand vein images

Images