In recent years, computer vision and automation technologies have become an integral part of military and sports training, especially in shooting training. Modern shooting ranges require high accuracy and an automated scoring system. The purpose of this study is to create a shooting range system based on computer vision algorithms and an automatic target changer using a multi-copy roller, which reduces the gap between shots. A configurable neural network accurately detects hits and analyzes the results in real time. The prototype system improves accuracy, automates processes and increases safety. Future goals include integration with learning platforms and enhanced functionality.