Comprehensive analysis method for multi-dimensional classification and identification of shoe sole patterns

Abstract

The invention provides a comprehensive analysis method for multi-dimensional classification and identification of shoe sole patterns. The method which comprises the following steps of: S-A, generating a shoe sole pattern database according to a shoe sole pattern multi-dimensional classification and identification encoding method, and querying a shoe sole pattern to be compared according to the encoding method of the database; and S-B, comparing the similarity between the shoe sole pattern to be compared and the shoe sole pattern data in the database according to the calculation methods of different feature areas, and obtaining the shoe sole pattern with the similarity according to a similarity score.

Description

technical field [0001] The invention relates to the field of computer image processing, in particular to a comprehensive analysis method for multi-dimensional classification and recognition of shoe sole patterns. Background technique [0002] At present, the serial and parallel support of on-site shoe prints is mainly realized by manual and binary image algorithms. There are many problems in the manual screening method, such as low efficiency, long time-consuming, and small series and parallel range; while the binary image algorithm is currently the most widely used footprint comparison scheme in China, but because the on-site footprint is affected by the medium, carrier and extraction technology and presents Due to the many influences of the trace process, the quality of on-site footprints is generally low, which seriously affects the accuracy of the binary image algorithm, and the improvement of the current binary image algorithm on the comparison rate of on-site footprint...

AI Technical Summary

Problems solved by technology

There are many problems in the manual screening method, such as low efficiency, long time-consuming, and small series and parallel range; while the binary image algorithm is currently the most widely used footprint comparison scheme in China, but because the on-site footprint is affected by the medium, carrier and extraction technology and presents Due to many influences of the trace process, the quality of on-site footprints is generally low, which seriously affects the accuracy of the binary image algorithm, and the improvement of the current binary image algorithm on the comparison rate of on-site footprints is slow and the effect is not obvious.

Through research and data analysis, it is found that at present, the main method for front-line criminal technicians in our city to carry out stringing and merging of on-site footprints is still manual stringing. meet work needs

At the same time, it is difficult to master the core technology of feature editing of binary images, and the quality of footprints cannot be monitored, which makes some personnel gradually give up using the footprint system and turn to artificial audition footprints, resulting in a lot of labor costs and low output.

Images