Online non-contact palm vein region-of-interest extraction method

Abstract

The invention discloses an online non-contact palm vein region-of-interest extraction method, which comprises the following steps of: S1, acquiring an image acquired by front-end acquisition equipment, and performing hand region segmentation on the acquired image to form a binary image; s2, in the binary image, separating a non-zero pixel region and a zero pixel region through a boundary tracking method to track a palm contour, and then supporting simplification of palm contour concavity analysis through adaptive sampling from edge detection to reduce the number of scanning points; s3, selecting feature points by using a concavity analysis method to construct key positioning points extracted from the region of interest; and S4, constructing reference points between valley points by using the index finger and the middle finger, the middle finger and the ring finger, and the ring finger and the little finger through the coordinates constructed in the step S3, and extracting a relatively large region of interest from the acquired image, thereby completing extraction of the palm vein region of interest.

Description

technical field [0001] The invention belongs to the field of pattern recognition, relates to the technical fields of biological recognition and computer graphics, and in particular relates to an online non-contact palm vein interest region extraction method. Background technique [0002] The region of interest extraction of palm veins is mainly determined by establishing a concave analysis model, which is a key step in palm vein recognition and an important basis for subsequent feature extraction and recognition. Compared with traditional fingerprint recognition, palm vein recognition under "living recognition" will face more palm lines, including mastoid lines, main lines and folds, etc., while scaling, rotation and irregular boundaries of the palm line area Changes are more complex. When performing region-of-interest extraction, the performance of the above-mentioned difficulties is also more prominent. As an important part of palm vein recognition, palmprint is extracte...

AI Technical Summary

Problems solved by technology

[0004] (2) In the case of natural shooting, the palm itself has obvious deformations such as scaling, rotation and irregularity, which brings difficulties in extracting the region of interest for the palm;

[0008] (1) The collection of palm veins mainly includes contact collection and non-contact collection. No matter which collection method is used, the collection process will be affected by factors such as light, collection background and temperature, which will increase the time for vein image recognition;

[0009] (2) The accuracy of feature extraction is low and the recognition rate is not high due to the lack of a suitable region of interest positioning and segmentation method;

[0010] (3) The palm vein image has its own palmprint, the existing algorithm still cannot completely remove the interference of the palmprint, and the robustness of the algorithm needs to be improved

Images