Processing method of indoor three-dimension model data

Abstract

The invention discloses a processing method of indoor three-dimension model data. The method comprises the steps that 1, multiple pictures shot by a camera are arranged and combined sequentially; 2, ascanning process is played back, and a 2D map is generated through an SLAM algorithm based on loopback detection; 3, the scanning process is played back again, an AMCL algorithm is adopted to conductlocating on the 2D map, and a three-dimension point cloud model is built; 4, the three-dimension point cloud model is denoised and smoothed to obtain the optimized three-dimension point cloud model;5, the arranged pictures are subjected to panoramic splicing, and the actual position of a panoramic map is obtained according to locating. The processing method has the advantages that the SLAM algorithm with the loopback function is adopted to obtain an optimal map, the optimal map is precisely located through the AMCL algorithm, the locating precision is improved, an entire indoor three-dimension model is obtained, and good interaction among users can be achieved.

Description

technical field [0001] The invention relates to the technical field of three-dimensional measurement, in particular to a method for processing indoor three-dimensional model data. Background technique [0002] For the existing indoor 3D data collection, it is necessary to plan the data collection points in advance according to the floor plan. There must be other data collection points within a radius of several meters (usually 3-5 meters, depending on the sensor of the device) of each data collection point, otherwise the The data at the collection points cannot be spliced; according to the planned sequence, data is collected at each data collection point in turn. At the same time, all moving objects and people in the scene need to be removed during the collection process. The equipment moves to the next collection point; after all the data is collected, each data packet is spliced ​​into a complete indoor 3D model. After the 3D model is modeled, install the client on the co...

AI Technical Summary

Problems solved by technology

After the 3D model is modeled, install the client on the computer, load the data and perform further interaction or processing, and provide cross-visualization services through page technology, while mobile devices cannot use it

At the same time, in the existing technology, the point cloud model is directly displayed, and the panoramic image is used as the texture material of the 3D model, and is not directly displayed to the user, so that the interactive user interface cannot realize other functions

[0003] In addition, the data processing program is developed for the designed hardware shape and the sensor used, which is easy to modify. After the equipment is finalized, the hardware shape and sensor must be strictly consistent with the design during production and use. Any error will affect the data. Mass produces irreversible effects

Therefore, it is difficult to adjust such devices for usage scenarios or functional requirements

Images