In the rainy season most of the images are cannot able to see clearly, sometimes it may also lead to do crime. So inorder to avoid that problem this project is done to see the images clearly in the rain also. The aim of this project is to remove the rain streaks from the HD videos using the algorithm of convolution neural network (CNN). It is done using MATLAB. First the videos are converted into frames and the frames are converted into the images. Convolution neural network is used to remove the unwanted noises. Thus the same the unwanted rain streaks are removed. The input is given as the rain streak images, using the convolution neural network algorithm the rain streaks are removed and the output comes without the rain streak images. Such a new technique makes the proposed method capable of properly extracting rain streaks from videos, thus getting fine video with deraining effects. And the particular part can also be shown clearly while the car is in the rain. For example car number plate can be taken, if the car is travelling in the rain , the number plate cannot be detected in the normal camera , using the CNN algorithm the number plate of the car can be detected in the clear manner.