Finger vein image segmentation method based on directional valley detection, system thereof and terminal

Abstract

The invention discloses a finger vein image segmentation method based on directional valley detection, a system thereof and a terminal. The method comprises the following steps: 1, a finger vein imageacquired by an image acquisition module is received; 2, ridgelet transformation is carried out on an intercepted finger vein image and image enhancement processing is carried out; 3, a valley area isextracted for the vein image after enhancement processing; 4, the obtained vein image is divided to three areas: a background area, a confusion area and a foreground area, and three times of threshold segmentation processing are carried out on the three areas; and 5, the obtained segmented image is subjected to edge detection, and a segmented image after processing is outputted. According to thesystem and the method of the invention, an algorithm is put forward according to some intrinsic information contained in vein images such as the shape, the step property and the direction, the vein features of interest in the image can be accurately extracted, and extracted error features can be reduced.

Description

technical field [0001] The invention relates to the field of image segmentation, in particular to a finger vein image segmentation method based on direction valley shape detection, a system and a terminal thereof. Background technique [0002] Image segmentation refers to separating the part of the image that contains important information from its background, so as to extract the objects with important meaning and the target object of interest in the image, and then understand the image well. [0003] Segmenting finger vein images can not only speed up feature extraction, but also increase the accuracy of feature extraction for low-quality finger vein images, which will be very beneficial to improve the recognition rate. Although many image segmentation algorithms have been researched, in terms of finger vein recognition, image segmentation algorithms are still very lacking, because there are some low-quality texture areas in finger vein images, these areas are not easy to ...

AI Technical Summary

Problems solved by technology

Although many image segmentation algorithms have been researched, in terms of finger vein recognition, image segmentation algorithms are still very lacking, because there are some low-quality texture areas in finger vein images, these areas are not easy to segment, if the algorithm is not effective enough It is better to classify these areas into the background, then some finger vein features will be lost. Even if these areas are classified as feature areas, because of their poor quality, not only can not extract useful feature information from them, but will also affect the feature extraction results. Causes a lot of interference, which reduces the recognition rate of finger veins

Images