A weak flash target detection algorithm based on background subtraction

This paper proposes an enhanced target detection algorithm based on background subtraction, designed to facilitate the rapid identification of weak flash targets in complex backgrounds. The algorithm establishes a background model devoid of foreground elements by subtracting the current frame from the background model, thereby facilitating the iterative updating of the background model and enabling the reflection of the overall contour of the moving object in a more comprehensive manner. Additionally, the algorithm exhibits a fast detection speed, a low false alarm rate, a significant improvement in ghosting and voiding, and a superior effect in moving target detection.