Research on image preprocessing method of long-wave infrared polarization imaging

Abstract

Infrared polarization image has excellent applications in target detection, tracking and warning, etc., but the original infrared polarization image has problems such as cold reflection and line blurring, which results in the low quality of the final image and affects the overall effect of infrared polarization detection. To address the above shortcomings, we propose an improved image correction method based on single-pixel non-uniformity differential image correction to remove the cold reflection. In addition, we use the two-dimensional discrete cosine transform method to Sharpen the image. The results show that the proposed method can effectively suppress the cold reflection problem and improve the robustness of the image.