摘要
This paper presents an efficient no-reference blur metric for images based on an artificial neural network (ANN) implemented by the back-propagation (BP) algorithm, named as BP-ANN. Three image features are extracted from spatial domains as the input vectors of the BP-ANN. The three features are No-Reference Structural Sharpness (NRSS), edge width and neighborhood correlation respectively. Experiments using the LIVE blur database demonstrate that the proposed algorithm correlates well with subjective quality evaluations.
- 出版日期2015
- 单位中国传媒大学