Moving targets detection algorithm based on background subtraction and frames subtraction

Aimed at the current difficulties of moving targets detecting in the settled scene, a new method for moving targets detection based on background subtraction and frames subtraction is proposed. At first, the background model for background subtraction is established with an average method, thus, one binary foreground image is obtained through the background with the background that can be updated real-time with the average method, then, several binary foreground images gotten by continual frames subtraction are ready. The other one is acquired with Boolean OR operator on these foreground images that are gotten by continual frames subtraction. Finally, the primitive moving targets are attained with Boolean AND operator on both of foreground images. The final results are acquired with the shadows detection and morphological filtering that work on the primitive moving targets. The experiments show that the proposed method is not only simple and efficient, but also a real-time one with the high operation speed and the high accuracy.