Redundant point cloud automatic elimination method and system based on photogrammetry technology

Abstract

The invention provides a redundant point cloud automatic elimination method and system based on a photogrammetry technology, and the method comprises the steps: firstly obtaining a sequence image of a dynamic object, carrying out the aerial triangulation matching through the photogrammetry technology, and obtaining an exterior orientation element of each image; and then, based on the exterior orientation element of each image, identifying point clouds corresponding to a static environment in the image point clouds through a preset algorithm, and automatically removing the point clouds. When a dynamic object is shot and modeled, the point cloud corresponding to the static environment in the image point cloud is automatically removed. The point cloud corresponding to the dynamic object is always in a changing state, and the point cloud value corresponding to the static environment is unchanged, so that the homonymy point pairs with similar point cloud coordinate values are searched as the point cloud corresponding to the static environment through the preset algorithm, and the point cloud corresponding to the static environment is automatically eliminated. According to the method and system, redundant point cloud data can be accurately removed, the requirement of dynamic object modeling in the later period can be met, the efficiency of dynamic object modeling is improved, and the precision of the model is improved.

Description

technical field [0001] The invention relates to the field of photogrammetry modeling, in particular to a method and system for automatically removing redundant point clouds based on photogrammetry technology. Background technique [0002] At present, the modeling methods of dynamic objects based on photogrammetry mainly include fully automatic modeling and semi-automatic modeling. [0003] The fully automatic modeling method is based on photogrammetry technology, which can directly perform automatic spatial three-matching on the captured images, and construct network modeling; the whole process does not require manual intervention, and is automatically completed by the program; it improves the efficiency of modeling, but due to redundancy Too much data leads to a large amount of data and takes a long time; and affects the accuracy of the model. [0004] The semi-automatic modeling method is on the basis of fully automatic modeling, using manual labor to manually remove the ...

AI Technical Summary

Problems solved by technology

Compared with fully automatic modeling, this method can greatly reduce the amount of redundant data and improve the accuracy of the model, but the manual operation is cumbersome, so it takes longer and less efficient than the fully automatic modeling mode

Images