Code inspection method for multi-dimensional classification and identification of shoe sole patterns

Abstract

The present invention provides a code inspection method for multi-dimensional classification and identification of shoe sole patterns. The method comprises the following steps: S1, after classifying each feature preset for the category of the shoe sole patterns, defining the pattern features; S2, pre-processing the shoe sole pattern images, carrying out normalization on the shoe sole pattern images to form normalized coordinates, and generating a map area under the normalized coordinates; S3, collecting and generating a feature set T of the whole map area, and calculating similarity of each dimension of the shoe sole pattern respectively; S4, gathering the similarity of each dimension of the shoe sole pattern, and calculating the score of the gathered dimensions; and S5, taking the target shoe sole pattern with highest score of similarity of each dimension as the approximate shoe sole pattern, and carrying out induction and combination.

Description

technical field [0001] The invention relates to the field of computer image recognition, in particular to a code checking method for multi-dimensional classification and recognition of shoe sole patterns. Background technique [0002] Shoe sole patterns (marks) are one of the most commonly encountered traces and physical evidence in case scene investigations, and play an important role in many investigative methods such as scene reconstruction, perpetrator characteristics analysis, and serial cases. In the public security criminal investigation work, searching and comparing massive shoe sole pattern pictures is the key technical difficulty of this technology. At present, the mainstream algorithm adopted in China is to compare the similarity of image data for image binarization conversion. However, due to the many influences of the on-site shoe sole pattern on the medium, carrier, extraction technology, and trace forming process, the on-site shoe sole pattern photo has the c...

AI Technical Summary

Problems solved by technology

However, due to the many influences of the on-site shoe sole pattern on the medium, carrier, extraction technology, and trace forming process, the on-site shoe sole pattern photo has the characteristics of light spots, noise, small trace contrast, unevenness, even deformation, and overlapping, which seriously affects the binary image. The accuracy of the algorithm, and this problem is the objective performance of the sole pattern on the spot. The pre-image correction of the binary image algorithm takes a long time, the workload is heavy, and this problem cannot be completely avoided, so many technicians gradually give up using this type of system comparison. Turning to manually select the sole pattern, resulting in a lot of labor costs and low output

Images