Footprint expression method combining regional confidence and pressure distribution direction intensity

Abstract

The invention provides a footprint expression method combining region confidence and pressure distribution direction intensity. The method at least comprises the following steps: obtaining a to-be-identified footprint image and preprocessing the image; partitioning and integrating the to-be-identified footprint image, and calculating the integrated footprint image; calculating the confidence coefficient of the partitioned footprint image region; and carrying out feature selection on the obtained pressure direction gradient features of the regional confidence coefficient through a principal component analysis technology and a linear discriminant analysis technology based on the footprint feature expression of the regional confidence coefficient. The method does not directly extract the characteristics such as the length and the angle of the footprint, so that the method is not easily influenced by noise. Meanwhile, footprint heavy pressure distribution and contour shape information areconsidered, and the distribution characteristic of plantar pressure when people walk can be better embodied. Furthermore, the method adopted by the invention eliminates the influence of textures whenthe socks are worn, does not need to distinguish the barefoot from the socks, and is wider in applicability.

Description

technical field [0001] The present invention relates to the technical field of footprint recognition, in particular to a footprint expression method combining regional confidence and pressure distribution direction strength. Background technique [0002] At present, the texture information, geometric shape and pressure distribution of footprint images are mostly extracted in footprint-based biometric recognition. For the acquisition of footprint texture information, Fourier transform, discrete cosine transform, wavelet transform, and principal component analysis and independent principal component analysis are usually used to identify footprints. [0003] For geometric features, the main methods are: 1) extracting morphological features such as foot length, palm width, heel width, angle from the center of the heel to the center of each toe and the horizontal axis, and the area of ​​the sole of the foot; 2) extracting the pressure center of the front and rear feet respectivel...

AI Technical Summary

Problems solved by technology

[0007] According to the above mentioned technical problems that cannot take into account the contour features of footprints and need to capture hi...

Images