INFRARED IMAGE STRIPE FILTERING METHOD BASED ON SEGMENTATION BETWEEN TARGET AND BACKGROUND

摘要

Provided is an infrared image stripe filtering method based on segmentation between a target and a background. The method comprises the following steps: S1, dividing an infrared image into a background region and a target region; S2, calculating the number line_nb of the background points and the number line_nd of target points at each line; S3, calculating an average value avg_ib of the background region rows and an average value avg_id of the target region rows; S4, calculating a local mean value and a local mean square variance of the average value of the target region rows, and calculating a local mean value and a local mean square variance of the average value of the background region rows; S5, calculating the stripe noise of both the target region and the background region; S6, calculating the final stripe noise; and S7, removing the final stripe noise from the infrared image. In the present invention, a target is separated from a background to respectively estimate mean values of the target region and the background region, so that the method is suitable for more scenarios, especially for the scenario for stripe filtering of video images.