115 / 2023-05-31 18:31:10

Judgment method of tunnel section shape based on point cloud

irregular tunnel,point cloud data,section shape,tunnel classification,computer vision

The safety of mineral resources is crucial to the national economy and the people's livelihood, and monitoring the shape of tunnels is one of the important means to ensure the safe mining of minerals and avoid illegal mining. With the development of technologies such as computer vision and machine learning, scene analysis and modeling using 3D point cloud data has become a widely used research direction. In underground mine operations, the use of 3D point clouds provides help for detecting tunnel deformation and avoiding illegal overexploitation. The noise inside the mine tunnel seriously affects the accuracy of point cloud analysis. The reasons for the poor effect of traditional denoising work come from two aspects. One is that the noise caused by the underground environment such as humidity, smoke, engineering equipment, pipelines and cables during the data collection process is not easy to identify and eliminate. The second is that there are holes in the tunnel caused by the irregular operation of the collection personnel, and there are silhouettes of construction workers, etc. In the actual denoising work, using the same algorithm to denoise the tunnels of different shapes is not effective. Therefore, this paper proposes an algorithm for judging tunnel shape based on point cloud to implement different denoising algorithms for different shapes of tunnel, which provides an algorithm basis for the realization of targeted mine tunnel point cloud data denoising. This technology uses handheld SLAM to obtain the original point cloud data in the tunnel. The data comes from different mine types, different mining areas, different scales, and different tunnel data sets in Shandong Province. This paper preprocesses and extracts the features of the original point cloud data. , and the recognition effect of the method is verified through experiments, and a classification test is also carried out. The results show that the method can effectively identify different types of tunnel section shapes, including rectangle, circle, ellipse, trapezoid, and the composite forms of these basic shapes, such as circle+trapezoid, circle+rectangle, ellipse+rectangle etc., which can well meet the requirements of performing subsequent denoising algorithms. The technology proposed in this paper, based on the point cloud tunnel section shape judgment method, is the premise of subsequent denoising, and provides a new idea and technical means for targeted denoising and automatic processing of different types of tunnels. This technology will help improve the efficiency and accuracy of mine safety mining, and has important application value and promotion significance.