Published Paper Details:

In computer science digital image processing is the part of the computer algorithms to perform image processing on digital images. As a subcategory or field of advanced sign handling, computerized picture preparing has numerous points of interest over simple picture handling. Visual bends on pictures brought about by awful climate conditions can negatively affect the exhibition of numerous open air vision frameworks. One regularly observed terrible climate is downpour which causes huge yet complex nearby force variances in pictures. The paper targets building up a viable calculation to expel special visualizations of downpour from a solitary downpour picture, for example separate the downpour layer and the de-down-poured picture layer from a downpour picture. Most picture preparing strategies include disengaging the individual shading planes of a picture and regarding them as two-dimensional sign and applying standard sign handling procedures to them. There are a great deal of strategies have been proposed to decrease the raindrops and snowflakes in recordings. In this paper, two popular techniques are employed, namely image decomposition and dictionary learning Built upon a non-linear generative model of rain image. We propose a word reference learning based calculation for single picture de-pouring. The rot is practiced by the blend of storm/snow disclosure and a guided divert in the data picture.