国产欧美精品一区二区,中文字幕专区在线亚洲,国产精品美女网站在线观看,艾秋果冻传媒2021精品,在线免费一区二区,久久久久久青草大香综合精品,日韩美aaa特级毛片,欧美成人精品午夜免费影视

Three-dimensional reconstruction technology has gradually become a key means of safety detection during the operation of water diversion tunnels. Affected by the special hydrological environment noise of the tunnel, the noise of the data acquisition equipment and the noise of the carrier motion, the collected point cloud data will inevitably suffer from noise interference, resulting in a lack of useful information, which is not conducive to the three-dimensional reconstruction. Therefore, this paper proposes a point cloud denoising algorithm based on sonar data feature points to achieve denoising of tunnel point cloud data. Firstly, according to the characteristics of the sonar point cloud data of the water diversion tunnel, the visual distance and perspective vector feature parameters are defined; secondly, the point cloud drift vector is estimated by coupling the perspective vector and the point cloud normal vector, and the probability density distribution of the perspective distance parameter is estimated by using the kernel function method to calculate the drift distance; finally, the drift algorithm is used to achieve noise filtering while maintaining the characteristics of the point cloud model. Experimental results show that the proposed algorithm can remove the data noise of the tunnel point cloud model while maintaining the detailed characteristics of the diversion tunnel model, and provide a high-precision point cloud data model for the detection of subsequent tunnel diseases.