Image Radiation Compensation Method Based on Two-dimensional Attention Module

With the diversification of projection scene requirements and the continuous development of projection technology, various scenes containing color textures are used as projection surfaces in daily life. These projection surfaces lead to color distortion of projection image affected by color texture and affect projection effect. The radiation compensation of the projector is to counteract the impact of the textured surface on the projected image by processing the input image of the projector. Aiming at this problem, this paper proposes a color compensation network model for projected images based on a two-dimensional attention module. The model uses a convolutional neural network to learn a complex mapping between the projector input image and the image captured by the camera. An attention module is added to the feature extraction network to generate attention map refinement adaptive features. Finally, a combination of loss functions is set for the specific task of color compensation for network optimization. The experimental results show that the method proposed in this paper effectively solves the problem of color distortion of projected images caused by colored textured surfaces and improves the quality of the projection.