TBT-L: A Shooting Training Methodology Based on Optical Sensors and Computer Vision.

This paper presents the development of a low-cost Basic Laser Shooting Trainer (TBT-L) designed to optimize initial military instruction. The solution integrates the Unity game engine for creating immersive scenarios with the OpenCV library for real-time computer vision processing. The system utilizes Commercial Off-The-Shelf (COTS) hardware, such as projectors and monocular cameras, replacing expensive proprietary infrastructures. The proposed methodology includes rigorous calibration steps for optical distortion correction and a laser detection algorithm based on frame subtraction. Experimental results demonstrated a 96.71\% hit rate and an average grouping of 4.38 cm, with a stable latency of 0.88 seconds. It is concluded that the simulator is an effective and scalable tool, capable of improving technical shooting skills in a safe and economical manner.