Finger vein identification method and device

The method according to claim, wherein determining the finger vein blood vessel lines in said processed region comprises:

• sequentially adopting each pixel in a geometrized region as a center pixel, employing the center pixel as a center of a circle to mark the circle of radius r;
• determining a total number of pixels within said circle that have grayscale values not exceeding a grayscale value of said center pixel;
• when a determination is made that the total number satisfies a predetermined condition, determining that said center pixel is part of a finger vein blood vessel line and setting the grayscale value of said center pixel to a first grayscale value;
• and
• when a determination is made that the total number does not satisfy the predetermined condition, determining that the center pixel is not part of a finger vein blood vessel line and setting the grayscale value of the center pixel to a second grayscale value, such that when said first grayscale value is 255, said second grayscale value is 0, and when said first grayscale value is 0, said second grayscale value is 255.

A device for identifying finger veins is characterized by comprising:

• a collecting module implemented by a processor and configured for collecting a finger vein image;
• a region extracting module implemented by the processor and configured for employing a line fitting method to extract a region of interest in the finger vein image;
• an image processing module implemented by the processor and configured for:
• conducting geometric normalization and grayscale normalization processing of said region of interest to obtain a processed region, and
• determining finger vein blood vessel lines in said processed region to obtain a finger vein blood vessel image;
• and
• an identification module implemented by the processor and configured for conducting finger vein identification based on the finger vein blood vessel image,
• wherein said identification module is further configured for:
• determining a first total number of pixels in the finger vein blood vessel lines of the finger vein blood vessel image, a second total number of pixels in finger vein blood vessel lines of another finger vein blood vessel image, and a third total number of pixels at a particular coordinate position of the finger vein blood vessel lines in the finger vein blood vessel image and in the another finger vein blood vessel image
• calculating a matching rate based on the first total number of pixels, the second total number of pixels, and the third total number of pixels, and
• identifying a match between the finger vein blood vessel image and the another finger vein blood vessel image based on the matching rate satisfying a condition.