Rain or Snow Removing in a Color Image Using Three Stage Approach Based on Interpolation

Abstract

The image restoration is one of the widest technique in image processing, sometimes images are degraded due to rain and snow such situations image components are not perfectly visible . Restoring such images, low pass filters are not preferred because which eliminates some of the object components. To avoid that a new three stage restoration technique for removal of rain and snow is implemented in this project. First of all guided filter for high pass components is implemented from that low pass components are also obtained, by using neighbourhood processing finding the position of rain and snow such that remove the area and finally the unknown components are estimated by using interpolation technique. Further enhancement can be achieved by using the low pass median filter.