A magnification algorithm of underground mine surveillance images based on PDE

摘要

The mine surveillance system can record the information of the production field timely and accurately. In order to reduce the production cost, the resolution of the system is lower. High resolution image can provide more information. In addition, surveillance images contain a lot of noise because of the poor quality of air and inadequate lighting. This paper presents a novel image magnification algorithm based on PDE. The algorithm uses WaDi interpolation result as initial image and corrects the image edges with total variation (TV) model. Experiment results show that the algorithm can protect the details of image and remove the noise effectively. Compared with the traditional image interpolation results, the edges of magnified images are clear. The PSNR and edge sharpness of images are also improved.