The first monograph by Russian researchers dedicated to digital processing of aerospace images has been published in London. The co-authors were from the TSU Faculty of Innovative Technologies scientists and the rector and two graduate students of the Southwest State University (Kursk). The book outlines new algorithms and mathematical models that automatically decrypt aerial and space imagery data, while minimizing the number of restrictions and providing high-precision recognition of terrestrial objects.
- The monograph was based on the data obtained during a large-scale project that was carried out at the TSU computing cluster SKIF Cyberia, and unites nine organizations, including JSC Gazprom Space Systems, JSC Academician M.F. Reshetnev Information Satellite Systems, and other structures, - says Vladimir Syryamkin, one of the main authors, the editor of the book, and a professor at the Faculty of Innovative Technologies TSU. - The models, methods, and algorithms proposed in the monograph are the basis for a wide range of software and hardware-software tools that provide image recognition with minimum errors.
Aerospace and satellite photographs are widely used today in urban planning, road construction, monitoring oil pipelines, controlling fire-danger conditions, protecting an area, tracking the movement of objects, and other purposes. For each of these, the accuracy of pattern recognition is critical.
- There are two basic requirements for the model used for automatic data analysis: it must have noise immunity and interference protection, - says Vladimir Syryamkin. - This means that it works effectively, regardless of the different types of restrictions, and it eliminates interference, in the event that someone creates it intentionally. The monograph, which we wrote in co-authorship with colleagues from Southwest State University, offers special ready-made models, the effectiveness of which has already been tested.
The book is intended for a wide range of specialists - scientists, builders, geologists, military, and others.