Research on Infrared Target Tracking and Detection Based on Least Squared Method

In order to compress calculation, the background estimation algorithm for the infrared target detection was proposed. Firstly, a Least Square Matrix is proposed and an image background is estimated. Then the target is detected by self-adaptive threshold detection in different images. It is shown by nonlinear function regression experiment and sequence infrared image detection experiment whose methods improve the performance of nonlinear function regression and the infrared background estimation. Experimental results show that this method has good operability and effective suppression of drift and objectives in matching with points of non-rigid deformation, as well as the accuracy of the infrared target detection tracking.