Abstract: Sniper aiming point trajectory vision measurement system is a high-precision optical measurement device used in modern sniper training. To address system's challenges of high-resolution image processing latency and trajectory localization difficulties caused by outdoor strong light, a fast centroid localization algorithm for aiming point based on adaptive threshold region of interest (ROI) projection was proposed and implemented on a field programmable gate array (FPGA) platform. The algorithm dynamically calculates adaptive threshold for ROI extraction by analyzing proportion of spot pixels in image in combination with relevant scene parameters, enabling efficient and accurate ROI localization while significantly reducing computational load. Experimental results showed that for images with a resolution of 2 448 × 2 048, proposed algorithm reduced the volume of processed data by 98.7%. Meanwhile, adaptive threshold mechanism effectively mitigated the impact of varying outdoor lighting conditions, and maintained stable centroid extraction across different shooting scenarios, significantly improving the system's real-time performance and robustness.