The work aims to enhance the image quality of the laser three-dimensional imaging system, effectively filter out noise in the image, and propose a filtering algorithm for adaptive mean shift. The mean shift filtering algorithm was improved based on the traditional algorithm. The mean square error of pixels within the area was selected as the control parameter for adaptive control of the size of bandwidth matrix h. According to the size of the broadband matrix h, the appropriate pixel element value was selected to participate in the process of calculating the mean, so as to improve the computational accuracy of results. The experimental results showed that the improved algorithm could effectively filter out noise in the image and improve the image clarity. The proposed algorithm has good edge-preserving and denoising characteristics.