According to characteristics of the images obtained by imaging devices based on CMOS sensor

a fullreference image quality assessment algorithm for CMOS sensor is proposed. The histogram of oriented gradient and information entropy theory are used to build the structure regularity of image

and combine structure regularity with luminance information to represent image quality. Experimental results show that the correlation coefficients between objective and subjective scores are above 0. 95 for most common types of distorted images. The proposed algorithm can be used as the benchmark for evaluating and promoting the performance of CMOS sensor and improving imaging quality of camera.