Application of computer vision and image prepro-cessing technologies in decision support systems

The article focuses on the development of a decision support system based on computer vision technologies and image preprocessing methods. It presents a comparative analysis of algorithms used to enhance the quality of visual data and to automatically identify key objects in images. The study emphasizes various methods such as filtering, equalization, and image segmentation, which are designed to improve the accuracy of computer vision algorithms in detecting important structures. The research tested and evaluated several image preprocessing techniques, including adaptive and standard histogram equalization, median filtering, and gamma correction. The results demonstrate that applying image preprocessing significantly improves the quality of data analysis. The Shi-Tomasi algorithm showed the highest efficiency in object recognition, especially when used with equalization techniques, allowing for precise identification of structural landmarks. The paper highlights the critical role of preprocessing in boosting the performance of computer vision systems across various domains, such as industrial applications, quality control, safety, and autonomous systems. These systems enable the automation of image analysis while reducing the impact of human error in decision-making processes. A practical example of these technologies is the development of a system for diagnosing musculoskeletal injuries and pathologies based on X-ray images. The implementation of such systems in medical practice accelerates diagnostic processes and improves the accuracy of diagnoses, ultimately reducing the workload of specialists and enhancing the quality of healthcare services.