USING REGION BASE IN-PAINTING TO REMOVAL UNWANTED OBJECT FROM VIDEO
Priyanka Prabhakar Waman, Rajashri Baban Khilari, Laxmi Ramesh Taware, Roshan Trimbak Awari · International journal of advance research and innovative ideas in education · 2017
Image in-painting, is sometimes it is also known as “image completion” is the reconstruct and recovers removed or corrupted pixel of an image in a given section so that the reassemble image looks real. These image in-painting techniques recover the eliminated pixel in images by increase continues structures into the target section via diffusion. One of the main problems of most related video completion and video in-painting techniques is the processing time. Many self-starting techniques for video in-painting are convenient, but most of them are compellingly demanding and fail to repair the damaged areas. Also, from the existing video in-painting algorithms a cryptic video classification has to be provided. To decrease this problem, in-painting process is introducing by using the background registration method which is suggested in this paper. The video is converted into no of image frames and the frame is registered. After, the edges of an object to be removed are detected by matching the roster frame with each adjacent frame of the vide o. Later an outline is reproduced for each time frame. Then the in-painting process is winding up uniquely for each and every time frame of the image. Finally, these reconstructed image frames are displayed one by one or continuously, so that it appears as a better quality video.