Applied photogrammetry systems, which act as measuring instruments, are often influenced by the external environment and operating conditions, which determine the accuracy of the results. In this regard, the problem of dynamic adjustment of algorithms for these changing conditions arises. To avoid an increase in the likelihood of human error and to reduce the requirements for personnel qualifications, you can resort to the tools of intelligent systems. For these purposes, the development of appropriate components is required, including machine learning tools. This article proposes a method and procedure for machine learning of the photogrammetric algorithm based on the observation of operator actions and a system of production rules.

Keywords: inductive learning, machine learning, photogrammetry, pattern recognition, photogrammetry, forestry, pipe industry, measurement, mobile app, automation

Today one of the topical issues is the practical application of the theory of pattern recognition, face recognition, including the definition of a person in the image and its subsequent identification. Such tasks include the extraction of indirect signs of interaction between users of social networks. By identifying the user's face and further searching it in photographs and video files in other profiles, an expansion of the list of basic data available to the analyst is achieved. It is known that the system of computer vision, which is able to recognize faces in a natural and artificial environment under any conditions, has not been implemented to date. One of the obstacles is the variation of the scene - the area where the face is being identified. The article proposes a model of an intelligent information system using bion-spirited methods for preprocessing images in order to expand the range of allowable scene variations.

Keywords: pattern recognition, object identification, photometry, intellectual system, social networks, analysis