A predicted compensation model of human vision system for low-light image

Human visual perception determines the evaluation results of imaging system. This paper presents a nonlinear contrast resolution enhancement based on human vision system (HVS) characteristics. As human visual system has the characteristics of space and time threshold, the tool of gray flattening is used to analyze the gray scale distribution of low-light image. With the determination results of contrast resolution in low-light image, a nonlinear compensation method based on HVS characteristics is presented for improving visual low-contrast resolution. A compensation model prediction for machine vision system is also established to quickly obtain the appropriate image. Experiments show that the algorithm achieves simultaneously brightness and contrast enhancement for the low-light image, which not only improves background luminance, suitable for human visual observation, but also enhances the contrast of the image with more obvious layering.