Design Radiosity Restoration Algorithm Based on Visual SLAM

Aiming at the problem that the camera is more sensitive to illumination changes during the AGV visual SLAM process of material conveying in the guide roller production workshop, a radiosity recovery algorithm based on visual SLAM is proposed. Firstly, the camera imaging model is established by analyzing the camera imaging principle, and the modeling analysis is carried out for the halation factor, exposure time and response function in the model, and the optimization model of the radiance parameters is established; Then, the camera radiosity recovery algorithm is designed according to the model, the optimization effect is improved by formulating matching point selection strategy, and the real-time performance of the optimization algorithm is improved by using step-by-step optimization; Finally, it is verified by experiments that this algorithm has a high improvement in the robustness of the visual SLAM system when the illumination changes are large.