Analysis of the traces on the discharged bullets by the congruent matching profile segments method and the $k$-nearest neighbors

The article discusses the problem of classification into the categories of “match” and “non-match” for compared images of land impressions on discharged bullets. The aim of the research is to increase the effectiveness of the comparison of images of land impressions by the congruent matching profile segments (CMPS) method. The scientific novelty is in adding to the analysis an additional independent feature, as well as the use of the $k$-nearest neighbors algorithm at the final stage of traces comparison. The research shows that the accuracy of the classification for the compared pairs of the land impression images by the combined method is about 87%. The work establishes that the analysis of the images by the CMPS method makes it possible to effectively compare images of land impressions with high resolution (less or about 1$\mu$m per pixel). The research is of interest to software developers of automated ballistic identification systems