In the production of composts from organic raw materials, it is required to determine the quality of the fractional composition of the substrate or mixture. Determination of this index at raw material moisture values above 35-40% is accompanied by difficulties due to the specificity of the material, low mechanical strength of aggregates when using soil sieves, which substantially distorts the objectivity of the results. As an alternative to the current methodology, we have proposed a methodology for converting digitized images (high resolution) and constructing model surfaces based on model surfaces that reflect the structural formations of the composts being studied or compared to eliminate these discrepancies in the methodological approaches used. Studies were conducted to study the dynamics of composting of litter chicken manure using the proposed methodology, since the classical sieve analysis in this case was unacceptable. This was due to the fact that the fractional composition of the composts was compared from the raw material of the litter on the basis of sunflower seed husks and treated with water, liquid, humidity, respectively, 27.3% and 52.6%, and storage for 10-20 days. Divide into aggregates 52.6% of raw materials is quite difficult. During storage, the humidity and the aggregate state changed under the influence of natural thermal processes occurring in the composts. Methods of statistical processing and digitized surface obtained results characterizing the state of the substrate mixture.