Blind Quality Assessment for Cartoon ImagesAbstract—Current blind image quality assessment (BIQA) algorithms are mainly designed for natural images. Unfortunately, cartoon and cartoon-like images are quite different from natural images. Hence, recent BIQA methods are not very robust to cartoon images. In this paper, we propose a specific BIQA algorithm designed for cartoon images, which consists of the following terms. First, a multi-order gradient statistic term is used to measure the quality of edges, and a gradient histogram prior model of high-quality (HQ) cartoon images has been built. Second, a local binary pattern statistic term is adopted to describe the textural complexity on the non-edge area. Third, a blocking and blurring assessment term is utilized to measure some common artifacts of low quality (LQ) cartoons. Experimental results on cartoon image dataset demonstrate that the proposed method can accurately evaluate the visual quality of cartoon images and is more suitable for cartoon scenario than traditional BIQA algorithms. |
AuthorLove Computer Vision Archives
December 2018
Categories |