Approaches to the identification of the state of $\mathrm{3D}$-surface by $\mathrm{2D}$-objects images and their mass number

The development of the decision support systems in the technological process control, quality control of products, photo separation etc., needs to provide additional to computer vision recognition systems for the type and kinds of objects (of natural origin and their mass quantity with high variability within the classes and the vicinity of the classes themselves) functionality of identification and detailed quantitative assessment of the state of the surface. The possibilities of refinement of $\mathrm{2D}$-identification of the $\mathrm{3D}$-object state (the curvature of the surface and invisible sides) by the set $k = 2, 3, \dots$ of different angle shots in case of using not only one, but several cameras. To make up for the loss of information about the invisible on the $\mathrm{2D}$-image parts of the surface of the $\mathrm{3D}$-object, the algorithm for the computer vision system, providing additional to recognition of the types and kinds of objects functionality of identification and quantitative assessment of their condition, sensitive to the details of the surface, is presented. The proposed algorithm of control of information on the missing parts of the object surface is more effective, fast and stable in comparison with the algorithm, based on the restoration of $\mathrm{3D}$-image.