Palm print image region-of-interest extraction method

Abstract

A palmprint image region-of-interest extraction method comprises the following steps: performing reverse binarization processing on a palmprint image to obtain a palmprint binarization image; a connected region marking algorithm is adopted, all connected regions in the palm print binarization image are obtained, the areas of the connected regions are calculated, and the connected region with the largest area is obtained as a palm connected region; a vertical line is adopted to carry out scribing operation to detect eight finger seam feature points of the palm connected area, a connected area marking algorithm is adopted to obtain three finger seam feature areas of the palm connected area according to the eight finger seam feature points, and a finger seam point in each finger seam feature area is obtained; and performing rotation correction on the palm communication region based on the finger seam points, and performing region-of-interest extraction on the palm communication region after rotation correction. The palmprint image region-of-interest extraction scheme provided by the invention has extremely high consistency and stability.

Description

technical field [0001] The invention relates to the technical field of biometric identification, in particular to a method for extracting a region of interest of a palmprint image. Background technique [0002] When the palm print is collected, the position and direction of the palm of each person are different because the position of the palm is not limited. Even for the same person, it is difficult to keep the same position and direction every time. There will be differences in the position and direction of the , and these differences will have a greater impact on the subsequent identification. Therefore, before the feature extraction and recognition of the palmprint image, it is necessary to extract the region of interest of the palmprint image, and retain the palmprint region with high stability and consistency, thereby improving the palmprint recognition rate. At present, palmprint ROI extraction methods are generally a location extraction method based on an inscribed ...

AI Technical Summary

Problems solved by technology

Among them, the former needs to repeatedly search for the center of the circle and adjust the radius, the preprocessing time is long, the complexity is high, and cannot be used in real-time systems; the main difficulty of the latter is that the segmentation positioning point is not easy to determine, especially in the complex environment background, it is difficult to confirm the segmentation positioning point , resulting in low consistency and stability of palm area extraction, which affects the success rate of palm feature recognition

Images