Film line scratch detection using neural network

Citations

SCOPUS

2

초록

Line scratches are one of the most common degradations in old films. To support a demand of high quality of multimedia service, these should be detected and restored automatically. However, although many detection and restoration algorithms have been researched, little have done in automatic scratch detection. This paper presents a texture-based object detection method for scratch detection. We use a multi-layer perceptron (MLP) to automatically make a texture classifier that discriminates between scratch regions and non-scratch ones in various environments. To assess the validity of the proposed method, it has been tested with all kinds of scratches, that is, principal/secondary scratches, alone/not-alone ones, and moving/static ones, and then experimental results show that the proposed approach leads to not only robust but also efficient scratch detection. © Springer-Verlag Berlin Heidelberg 2004.

키워드

Discrimination; Neural Network Film Restoration; Scratch Detection; Texture
제목
Film line scratch detection using neural network
저자
Kang, S.K.; Kim, E.Y.; Jung, K.C.; Kim, H.J.
발행일
2004
유형
Article
저널명
Lecture Notes in Computer Science
권
3332
페이지
810 ~ 817