Adaptive line scratch detection in degraded films

In this paper, a robust, automatic and pixel-precision spatial line scratch detection algorithm is proposed. This algorithm deals with still images and may be followed by a temporal analysis to improve detection performances. By relaxing some of the hypotheses used in previous algorithms, detection of a wider range of scratch types is possible. The algorithm's robustness and lack of external parameters is ensured by the combined use of an a contrario methodology and local statistical estimation. In this manner, overdetection in textured or cluttered areas is greatly reduced. Experiments demonstrate the algorithm's ability to deal with difficult situations, in particular in the presence of noise, texture and slanted or partial scratches. Comparisons show the algorithm's advantages over previous work.