Efficient Method For Scratch Lines Noise Removal From Video

The digitalization and transfer of older films into high definition (HD) formats imply that high quality of restoration is necessary. Now a day�s Digital film restoration is an area under discussion of increasing interest to researchers and film archives alike. Old films, including cultural heritage masterpieces, are being digitally premastered and transferred into novel, higher quality formats and distributed through various means such as DVD, Blu-ray or HD pictures. Detection of Line scratches in old movies is a particularly difficult problem due to the variable spatiotemporal characteristics of this deficiency. Some of the main problems consist of sensitivity to noise and texture, and false detections due to thin vertical structures belonging to the scene. Automatic finding of image damaged regions is the key to automatic video image in-painting. Vertical scratches are the common damages in the old film. As the film is a collection of number of frames arrayed together to produce a motion sequence hence it becomes a lengthy and tedious work to process any video format in any manner. Normally if any scratch or noise generated on films it remains as it is on many frames in sequence in film which can be benefitted by the removal process by initially checking noise area on earlier slide. Hence proposed system is aimed at designing and developing of line scratches detection from old films and remove it. A line scratches detection algorithm based on edge detection is proposed. Edge detection is nothing but an image processing technique for finding the boundaries of objects inside images. The proposed algorithm first uses the operator which has the largest response to the vertical edge in Sobel operator to detect edges, and then uses canny operator to detect edges further. Third, we detect vertical lines in the image through probabilistic Hough transform. Finally, we obtain the true locations of the vertical lines scratches through morphology and width constraints. We contribute for removal of scratches using a new nonlinear continued fraction method dealing with both spatial and temporal information around the scratch is investigated in the restoration stage