Abstract <p>The spinal cord may be divided into segments. The neural networksof different segment groups control, in particular, locomotion andvisceral functions. Spinal cord segments serve as critical topographiclandmarks for both experimental and therapeutic interventions. However,accurate identification of segment positions in vivo, particularlythrough automated methods, remains challenging: in mammals, somespinal cord segments are displaced rostrally (ascend) relative totheir corresponding vertebrae, and the extent of this displacementvaries even within a single species. One solution to this problemmay be the use of reference images of slices of segments taken,for example, from histological atlases of the spinal cord. In thispaper, we investigate various methods for analyzing the similarityof gray matter contours in transverse slices of the mammalian spinalcord, which allow us to determine whether a slice belongs to a certainsegment. We consider the methods for analyzing slices obtained fromone animal (based on the Jaccard coefficient, the metric of distancesbetween contours, correlation analysis of <i>R-φ</i> curves,or Hu invariant moments), as well as the methods for comparing imagesof spinal cord segments with reference ones (correlation analysisof <i>R-φ</i> curves, Hu invariant moments).The results obtained allow us to assume that the method of determiningsegments by comparing tomographic or histological images of spinalcord transverse slices at different levels with a certain databasecontaining a set of reference images of slices of specific segments,based on Hu invariant moments, is the most effective of those considered.</p>