THERMAL RADIATION SCANNER METHOD FOR THE JOINT DETECTION OF DIFFERENT TYPES OF MINES

There are many methods of mine detection, such as X-ray radiography, neutron-radioactive method, ultrasound method, acoustic signature method, thermographic method, etc. With the advent of unmanned aerial vehicles, numerous studies have been conducted to develop safe procedures for detecting unexploded mines. When investigating and searching for mines using electro-optical sensors, various mathematical models are used to assess the accuracy of target detection. At the same time, a simple Johnson model is widely used, which relates the probability of detection to the spatial resolution of the sensor used. According to this criterion, the search for mines can yield the following results: (1) detection; (2) classification; (3) recognition; (4) identification. The objective of the study in this article is to determine the conditions for detecting the maximum number of the above two types of mines when using a thermal imaging scanner with a fixed spatial resolution. The problem is formalized under the assumption that maximum spatial resolution is the only limiting condition for solving this problem. At the same time, such restrictive conditions as the variety of shapes and sizes of mines, the variety of types and characteristics of the underlying surface and the location of the mine, etc. they are not counted. The problem of optimizing the joint search for two types of mines placed in a minefield using the thermal imaging method is formulated and solved. The Johnson criterion is used, which determines the number of necessary pixels for a specific size of the desired detail in the object image. The optimization problem of detecting the maximum total number of two types of mines has been solved when using a single scanner with a fixed spatial resolution index. The problem is solved by linear programming, taking into account a single restrictive condition, which is the limited spatial resolution of the scanner. It is shown that the use of the linear programming method makes it possible to determine the maximum possible number of detectable antipersonnel and antitank mines, provided that antipersonnel mines are detected with maximum probability.