A method for mapping pressure footprint image by optical footprint image

Abstract

The invention discloses a method for mapping a pressure footprint image by an optical footprint image. The method comprises the following steps: S1, acquiring the optical footprint image; S2, acquiring a pressure footprint image; S3, filtering and denoising the pressure footprint image acquired in the step S2; S4, combining the optical footprint image in S1 and the denoised pressure footprint image in S3 into one image as a training image; S5, importing the plurality of training images into a computer, and utilizing a Pix2Pix algorithm to obtain an optimal model through the continuous game ofa generator and a discriminator of the Pix2Pix algorithm; S6, collecting a single optical footprint image, importing the single optical footprint image into the optimal model obtained in S5, and mapping a pressure footprint image. The optical footprint image can be mapped into the pressure footprint image, and therefore the reliable pressure footprint image of the target individual is provided forpublic security and scientific research staff and used for analyzing morphological characteristics of the pressure footprint image.

Description

technical field [0001] The invention relates to the technical field of image processing and biological feature recognition, in particular to a method for mapping a pressure footprint image from an optical footprint image. Background technique [0002] Pix2Pix can achieve direct conversion of matching images in the two fields, and the obtained results are relatively clear; the main idea of ​​the U-net network is derived from FCN, which uses a fully convolutional network to classify images pixel by pixel, which can be achieved in the field of image segmentation. Good effect, because the network structure is similar to U-shaped, so it is named after this, its composition is composed of the convolution compression layer on the left end, and the transposed convolution amplification layer on the right end; PatchGAN discriminator is a kind of discriminant model, Intuitively, it is completely composed of convolutional layers, and the final output is an n*n matrix, and finally the me...

AI Technical Summary

Problems solved by technology

[0003] At present, the pressure footprint image is obtained by the person being collected standing or walking on the pressure plate, which requires manual operation, high machine cost, and regular maintenance is required to prevent the damage of the pressure collector from affecting the normal collection of the pressure footprint

And for the crime scene, it is impossible to obtain the pressure footprint image of the suspect in this way

Images