MATHEMATICAL MODELING AND IMAGE PROCESSING ALGORITHMS IN ARTIFICIAL INTELLIGENCE-BASED TECHNICAL VISION SYSTEMS

This paper presents a comprehensive study on the development of a mathematical model and image processing algorithms for artificial intelligence-based technical vision systems. The research focuses on analyzing the structural organization and functional components of the system, including image acquisition devices, illumination subsystems, and data processing modules that ensure stable and accurate operation. A systematic and integrated approach is applied to model the transformation of object parameters and to enhance the accuracy, robustness, and reliability of object detection processes. Particular attention is given to image preprocessing, segmentation techniques, and contour detection algorithms, as these stages play a crucial role in improving the quality of visual data and minimizing noise-related distortions. The experimental results demonstrate that the use of filtering techniques and centroid-based methods significantly increases measurement precision and reduces localization errors. In addition, the integration of stereo vision methods allows accurate estimation of object distance and spatial coordinates, which is essential for real-time applications. The proposed system also shows устойчивость under varying environmental conditions, including changes in illumination and background complexity. The findings confirm that artificial intelligence-based technical vision systems can be effectively implemented in industrial automation, robotics, and agricultural product sorting. The developed mathematical model and algorithms contribute to improving processing speed, accuracy, and system reliability, providing a strong foundation for further scientific research and practical applications in intelligent vision technologies.