Introduces a real-time color imaging system can be operated at night and uses a low-cost, simple and convenient CMOS sensor. In this imaging system, the sensor is driven by the FPGA, the differential data output by the sensor is decoded by the FPGA, and a series of image processing algorithms are implemented in the hardware description language through FPGA programming to improve the imaging quality of the system at night. The main problems solved by this imaging system are as follows: (1) For night time light energy is weak, and after adding an IR cut filter, the sensor can capture the problem of insufficient light. This paper proposes an imaging method without an IR cut filter. This method requires a color correction algorithm in the FPGA to correct the color distortion caused by the transmission of near-infrared light. (2) The problem of low contrast and insufficient color saturation for night time imaging results. This paper combines the gamma correction algorithm and the defogging algorithm to propose a method that can effectively improve the image color saturation. The combination of the two algorithms can improve the brightness of the nighttime image and make the image more vivid. (3) When imaging a single color sensor at night, the details and brightness of the captured object still do not meet the needs of the human eye. In this paper, a mono and color dual-camera system is used to improve imaging quality. In this system, two sensors are simultaneously imaged, so that the imaging results retain both the color information of the color sensor and the clarity of the mono sensor.