Method of the image de-occlusion by using focal plane camera

In order to address the problem that the foreground occlusions in the light path affect the acquisition of information of interest, this paper explores the de-occlusion algorithm based on the application of the camera array through experiments. The array light field camera is used to collect 4D light field data, after which the digital refocusing technology is employed to perform refocusing at different depths so as to highlight the detailed features of target objects. The experimental results support the feasibility of applying the array light field camera to remove occlusions, as well as its ability to enhance the quality of images, reproduce the image of occluded areas, improve the readability of images, and mitigate the effects of noise. According to the no-reference image quality assessment, in terms of the quality of reconstructed images, the algorithm in this paper can improve the signal to noise ratio (SNR) and peak signal to noise ratio (PSNR) by 17. 3% and 77. 6%, respectively.