Removing Image Artifacts of Wavefront Coding Technique to Improve Drone Detection

Abstract Wavefront coding is a useful technique which is employed to widen the depth of field. However, image artifacts are shown up on the restored image for asymmetric phase masks, which leads the degraded image quality. In this paper, we introduce a novel method based on the combination between the Inverse filter and the structure of Unet model to remove the image artifacts on the restored images. Another meaningfulness of this method is that the precision for drone detection will also be enhanced. The simulation will show the results for ability of drone detection by using the traditional optical system, the traditional wavefront coding and the proposed method. These results demonstrated that the combination between the Inverse filter with Unet model can be used to improve the precision of drone detecting over the big range of depth of field.