ROI image extraction method in finger vein identification

Abstract

The invention discloses an ROI image extraction method in finger vein identification. The method comprises that an original finger vein image is obtained from a collector; superhigh pixel segmentation is carries out on the original finger vein image to obtain a superhigh pixel image; Sobel operator edge detection is carried out on the original finger vein image to remove noise from the image and obtain a finger edge image; overlapped points of the superhigh pixel image and the finger edge image are obtained; tracking points are determined according to the overlapped points, edge tracking is carried out on the superhigh pixel image, and a complete finger edge image is obtained; the complete finger edge image is corrected to obtain a finger vein image with a fuzzy background; and angle correction, height cutting, width cutting and normalization are carried out on the finger vein image with the fuzzy background to obtain an ROI image. The extracted ROI image includes a complete finger boundary contour, and the segmentation effect is good.

Description

technical field [0001] The invention relates to the technical field of pattern recognition, in particular to a method for extracting ROI images in finger vein recognition. Background technique [0002] The extraction of ROI in finger vein recognition refers to the pictures obtained directly through the acquisition equipment. The pictures contain noise and a lot of background information that has nothing to do with the recognition. Cut out what is needed, the image of the finger region (ROI). [0003] At present, the common method of extracting ROI in finger vein recognition is mainly based on Sobel operator edge detection, that is: first perform Sobel edge detection on the original image; Binarize the image; blur the finger background; correct the angle of the rotated image; use a fixed value to locate the height of the finger; the horizontal width is determined by the inscribed line of the finger edge; image normalization. [0004] However, the solution based on the Sobel...

AI Technical Summary

Problems solved by technology

[0004] However, the solution based on the Sobel operator is only ideal for the ROI of HKPU-FV data packet segmentation, but it is not suitable for data packets collected by SDUMLA-FV, UTFV, THU-FV, MMCBNU_600 and other types of collection equipment. Not completely suitable, there will be problems such as the extracted ROI contains a large amount of background or the vein texture information is not rich enough

Due to the diversity of acquisition devices, the distinction between fingers and backgrounds is ever-changing, and the Sobel operator is sensitive to noise. It is difficult for existing ROI extraction methods to detect the complete finger boundary contour, so the extracted ROI may still contain A large amount of background information or vein texture is not rich enough, which reduces the effectiveness of feature extraction and the accuracy of matching

Images