Method for determining central area of palm image

Abstract

The invention provides a method for determining a central area of a palm image, which comprises the following steps of: carrying out thresholding processing on a difference image of a natural light-near infrared light source palm image and a natural light source palm image, and obtaining four finger root concave points among five fingers by combining obtained angular points and adopting a clustering method; and positioning the four finger root pits, obtaining a center line and a center point of the palm area according to a positioning result, and finally determining the center area of the palm image according to the finger root pits and the center point of the palm area. The rectangular center area obtained by the determination method is calculated in a self-adaptive manner through the size of the palm, so that the method has relatively high adaptability and robustness, and the accuracy of a determination result is high.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a method for determining the central area of ​​a palm image. Background technique [0002] The determination of the central area of ​​the palm image is the first step in palmprint and palm vein image recognition, which determines the specific values ​​of subsequent palmprint and palm vein features and has an important impact on the recognition accuracy. Most of the existing palm positioning algorithms use Harris corner points to extract the key points of the fingers, and then select the center area of ​​the palm according to the experience value, which has the problems of poor robustness and great influence of light, and the robustness and accuracy of positioning need to be improved. Improve and improve. [0003] The following prior art documents are relevant to the present invention: [0004] [1] Chen Diandian et al., Palm Image ROI Extraction Method Based ...

AI Technical Summary

Problems solved by technology

[0009] The ROI extraction method proposed in literature [1] uses the Harris method to detect corner points, which will generate redundant corner points, resulting in poor detection robustness of finger pits, and does not use images generated by near-infrared sensors, palm images Thresholding needs to manually set the threshold, and the effect of binarizing the palm area is poor

[0010] In literature [2], it is proposed to calculate the circumscribed circle of a triangle by using the concave points between the fingers to obtain the central area of ​​the palm. The disadvantage is that the detection accuracy of the concave points of the fingers has a great influence on the subsequent circular palm area, and the Harris method is used in this paper to detect Dimpled fingers, results heavily influenced by lighting and palm posture

[0011] In literature [3], the height and width of the square area in the center of the palm are manually set, and the consistency of different palm center areas is poor.

[0012] In literature [4], the method of fixed direction difference is used to calculate the key points between the fingers. It does not use the near-infrared sensor, so the effect of the binarized area of ​​the palm obtained is poor, and this method is sensitive to the palm orientation.

[0013] The palm area determination method proposed in literature [5] needs to select control points in three situations to determine the concave point between the fingers, which is complex, and the width and height of the square area in the center of the palm need to be manually selected, and there is room for improvement.

[0014] In summary, the robustness and accuracy of the methods for determining the palm center area in the prior art need to be improved

Images