No-reference image blur assessment based on gradient profile sharpness

The increasing number of digital images and videos has boosted the need of no-reference objective image and video quality assessment (QA). In this paper, we focus on proposing a perceptual-based no-reference image blur assessment using the feature of gradient profile sharpness (GPS). We first build a triangle model to represent the gradient profiles of edge pixels, and propose GPS based on the triangle model. Then a metric for blur assessment is extracted from the distribution histogram of GPS. Experimental results show that the proposed blur metric correlates well with the perceived blurriness, and it can achieve competitive performance with the state-of-the-art no-reference metrics on public image dataset.