Camera image quality assessment without reference information

Abstract

Blur plays an key part in evaluating of camera image quality. It leads to decrease of high frequency information and accordingly changes the image energy. Recent researches in quaternion singular value decomposition show that the quaternion's singular values and associated vectors can capture the distortion of color images, and thus singular values can be utilized to assess the sharpness of camera image. Based on this, a novel blind quality assessment method considering the integral color information and singular values of the blurred camera image is proposed for evaluating the sharpness of camera image. Pure quaternion is utilized to represent pixels of the blurred camera image and the energy of every block are obtained. Results confirm the superiority of the proposed blind algorithm in assessing camera images.