3D virtual scene images have many features and are affected by factors such as changes in viewpoint and occlusion,which makes it difficult to extract image features.To address this problem,a multi-feature parallel tracking algorithm for 3D virtual scene images was proposed.Firstly,the algorithm calculated the image gradient to denoise the image,and then used the gray co-occurrence matrix to perform gray-level processing,thus calculating feature values.Moreover,a particle filter tracking algorithm was used to fuse multiple features.Meanwhile,a weighted fusion rule was introduced to determine the final position of each feature of the 3D virtual image,thus achieving multi-feature parallel tracking.Simulation results show that the proposed algorithm can accurately track multiple features of 3D virtual scene images with good robustness,and the tracking speed is only 23.4 frames per second.
关键词
三维虚拟场景图像/多特征/并行式跟踪/灰度共生矩阵
Key words
3D virtual scene image/Multiple features/Parallel tracking/Gray co-occurrence matrix
维普
万方数据