A Local Information Entropy-Based Point Cloud Simplification Algorithm for Hazardous Chemical Warehouses

In automatic safety monitoring of hazardous chemical warehouses,point cloud acquisition with structured light cameras is one of the means to obtain scene data.The complex point cloud of the hazardous chemical warehouse scene has large sc-ale,uneven density and redundant information,which requires simplifying the point cloud.Inter-stack ranging is a key tech-nology for the safety monitoring of hazardous materials warehouses.Therefore,more attention should be paid to the conti-nuity of the edge of the stack point cloud and the integrity of the gap when the point cloud is simplified.The paper firstly a-dopts the optimal raster partition method for downsampling,which better retains the sparse point cloud information while sieving out the dense point cloud;then,aiming at the problem that the stack gap data is easy to be simplified,the local info-rmation entropy is calculated by the normal vector of the point cloud,and the point cloud is simplified based on it.The exp-erimental results show that the point cloud density after simplification is more uniform,and the stacking shape can be ret-ained more completely.At the same time, the method is also applicable to public datasets.