Human Finger Vein Recognition Method

Abstract

The invention discloses a human finger vein recognition method, which includes reducing the target image to a set ratio, performing primary clustering on all pixels in the target image according to their gray values; deleting the maximum central value and The clustering set corresponding to the smallest central value, and re-clustering the gray values ​​in the remaining clustering sets; delete the class with the largest central value in the re-clustering, and merge the remaining classes in each category into a gray-scale set , according to the maximum gray value and the minimum gray value in each category, the gray value in the gray set is updated to obtain the feature map data; the pixel values ​​greater than zero in the feature map data are extracted, and for each feature map data respectively Form a set of spatial points; use the ICP algorithm to calculate the total registration effect value of the target image and the fingerprint image template; when the minimum value of all total registration effect values ​​is less than the set threshold, the target image and the minimum registration effect value The collector corresponding to the fingerprint image template is the same person.

Description

technical field [0001] The invention relates to fingerprint identification technology, in particular to a human finger vein identification method. Background technique [0002] In the fingerprint recognition process, because the acquisition of the finger vein image requires the photosensitive chip to work in a high ISO (high sensitivity) mode, the high sensitivity of the image will cause strong shot noise to be superimposed on the acquired image, resulting in Many digital image processing algorithms are difficult or impossible to work; at the same time, under the strict constraints of recognition time and accuracy, it is also difficult to use conventional large-scale calculation algorithms such as noise removal and image matching. [0003] Therefore, a finger vein recognition algorithm that is insensitive to shot noise and light and is suitable for computing platforms with low power consumption and low performance needs to be developed urgently. Contents of the invention ...

AI Technical Summary

Problems solved by technology

[0002] In the fingerprint recognition process, because the acquisition of the finger vein image requires the photosensitive chip to work in a high ISO (high sensitivity) mode, the high sensitivity of the image will cause strong shot noise to be superimposed on the acquired image, resulting in Many digital image processing algorithms are difficult or impossible to work; at the same time, under the strict constraints of recognition time and accuracy, it is also difficult to use conventional large-scale algorithms such as noise removal and image matching

Images