With the development of new display technology and the continuous warming of the concept of metauniverse, micro-display has been widely used in near-eye display fields such as virtual reality and augmented reality. The silicon-based OLED micro-display has become the best choice for near-eye display devices due to its advantages of small area, high brightness, fast response, wide color gamut and high resolution. traditional silicon-based OLED micro-displays mostly adopt analog drive mode, but it is difficult to meet the requirements of high resolution and high refresh rate due to the limitation of the conversion speed and accuracy of digital-to-analog converter. The digital driving mode adopts pulse width modulation method, which makes the human eye perceive different gray levels by changing the light and dark time of pixels. With the advantages of fast switching speed, high stability and low noise, digital drive is more suitable for silicon-based OLED micro-displays with fast photoelectric response. However, with the further improvement of the requirements for the imaging quality of micro-display, the current scanning mode for digitally driven micro-display is difficult to meet the massive data transmission caused by the improvement of resolution and refresh rate. At the same time, digital drives mostly adopt the addressing display period separated sub-field method. The distribution of the luminous sub-field is not continuous within a frame time, and the dynamic false contour phenomenon occurs when the moving picture is displayed, which seriously affects the viewing quality of the human eye.Aiming at the imaging problems such as low resolution, low refresh rate of micro-display and dynamic false contour when displaying moving pictures, this paper proposes a digital driving scanning strategy based on super-pixel technology and digital driving principle. This method uses the integral property and visual persistence property of the human eye to quickly switch between two lower resolution images in the process of display. When the switching frequency is far beyond the critical flicker frequency that the human eye can perceive, the human eye will recognize two lower resolution images as one higher resolution image due to the spatial dislocation of the two images on the displays. A super-pixel model is built using MATLAB and the image quality of the super-pixel processed image is evaluated. Under the condition that the data transmission bandwidth is reduced by 50%, the average PSNR of the super-pixel image is about 34.086 dB, and the average SSIM is about 0.942. According to the integration method, the image quality of the super-pixel image perceived by the human eye is not significantly different from the original image. This scanning strategy provides an effective solution for improving the resolution and refresh rate of micro-display. In order to further prove the effectiveness of the super-pixel scanning strategy, this paper analyzes the mechanism of dynamic false contour and simulates the dynamic false contour of the image under the super-pixel scanning mode. Considering the existence of visual threshold in human eyes, the dynamic false contour of super-pixel is evaluated by combining and extending the just noticeable distortion integral method. Compared with traditional scanning methods, the super-pixel scanning method displays two sub-frame images in one gray period, and the display time of each sub-frame only accounts for half of the gray period. When the integration method is used to simulate the gray level of pixels perceived by the human eye, it is necessary to integrate the subfields of two sub-frame spans. The simulation results indicate that under the proposed scanning strategy, the probability of the error between the maximum perceived brightness of the human eye and the actual brightness of each pixel in the process of image movement is equal to 0 and not more than 8 gray levels is about 93.3% and 99.3%, respectively. Compared with 19 subfields and CGPWM scanning mode, the dynamic false contour phenomenon under the super-pixel scanning strategy has been significantly improved, and the display effect conforms to the human eye observation experience. According to the super-pixel scanning strategy, a digital driving type super-pixel micro-display controller was designed, and a system verification platform based on FPGA was built. The full-color digital driving type silicon-based OLED micro-display with resolution of 2 048×2 048 was successfully driven, which proved the feasibility of the proposed scanning strategy and laid the foundation for the application of the digital driving type silicon-based OLED micro-display based on super-pixel.