An MTF-Inspired Underwater Image Quality Assessment Method

Due to the light attenuation in water, underwater image qualities are affected and various in different imaging systems. In this paper, a modulation transform function (MTF)- inspired image quality assessment method is proposed and in which a point-by-point slanted-edge algorithm is introduced to evaluate the underwater systems. Four systems with different light sources are installed to verify the method at different distances in the same water quality. The light sources of these systems are a customized 500W lamp, biconvex lens cluster (BILC) light, Fresnel lens cluster (FNLC) light and digital light processing (DLP) compensated floodlighting, respectively. Then, experiments demonstrate the superiority of DLP compensated floodlighting than the other systems and the obtained MTF curves show that when the spatial frequency is 0.5 cy/pixel, the MTF value of the DLP imaging system at the imaging distance of 1m and 3m are 0.05 and 0.1, respectively. The experiments show that the proposed method can be used to evaluate imaging quality effectively.

I. Introduction

Due to the absorption and scattering of water medium, the energy of light is exponentially attenuated when it is transmitted under water, which causes color distortion and edge blurring of the underwater images. Different imaging equipment performs differently, and it is affected by water quality. Therefore, establishing an effective image quality assessment method is of great significance to assess the performance of an imaging system.