No-reference image blur assessment based on local total variation
Wenhua Wang, Zhiming Wang · 2016
In this Letter, we present a simple yet effective no-reference image blur assessment algorithm based on local total variation. Firstly, we calculate a local image blurriness metric by total variation. Then, the average of the largest 1% local total variation is computed. Finally, we obtain the blur score via a five-parameter logistic regression. The performance of the proposed algorithm is evaluated on four publicly available databases. Experimental results given by proposed algorithm are highly correlated with human assessment scores and competitive with the state-of-the-art techniques.