A ground point segmentation system and method based on vector iteration

Abstract

A ground point segmentation system and method via vector iteration includes a light source and a central processing unit. The light source is installed in a vehicle and is used to acquire a three-dimensional model. The central processing unit is electrically connected to the light source and includes an analysis and calculation module and a storage module. The analysis and calculation module receives the three-dimensional model and analyzes and calculates at least one reference vector value and a target vector value using a vector iteration ground point segmentation method. An angle is formed between the reference vector value and the target vector value. The angle is calculated and compared with a preset threshold of the storage module to determine whether the angle matches the preset threshold and to generate a ground point or a non-ground point.

Claims

1. A ground point segmentation system based on vector iteration, comprising: A light source, installed in a vehicle, is used to acquire a three-dimensional model; and a central processing unit, electrically connected to the light source, the central processing unit including an analysis and calculation module and a storage module. The analysis and calculation module receives the three-dimensional model and analyzes and calculates at least one reference vector value and one target vector value through a vector iterative ground segmentation method. An angle is formed between each reference vector value and the target vector value. The angle is calculated and compared with a preset threshold of the storage module to determine the result of the comparison between the angle and the preset threshold and generate a plurality of ground points or a plurality of non-ground points.

2. The ground point segmentation system via vector iteration as described in Request 1, wherein the 3D model is a point cloud model.

3. The vector-based iterative ground point segmentation system as described in claim 1, wherein the central processing unit further comprises a classification module that receives a portion of the ground points and reclassifies them to form an unknown label point.

4. A ground point segmentation method based on vector iteration, comprising the following steps: A vehicle is equipped with a laser, which acquires a three-dimensional model. A central processing unit receives the three-dimensional model and uses a vector iterative ground segmentation method to analyze and calculate at least one reference vector value and one target vector value through an analysis and calculation module of the central processing unit. An angle is formed between each reference vector value and the target vector value. The angle is calculated and compared with a preset threshold in a storage module of the central processing unit to determine the result of the comparison between the angle and the preset threshold and generate a plurality of ground points or a plurality of non-ground points.

5. The vector iterative ground point segmentation method as described in claim 4, wherein when the angle is less than the preset threshold, the target vector value after iteration is iterated again through the vector iterative ground segmentation method to iterate the reference vector value after iteration.

6. The ground point segmentation method via vector iteration as described in request item 4, wherein when the angle is greater than the preset threshold, the iteration of the target vector value after iteration stops on the reference vector value after iteration.

7. The ground point segmentation method via vector iteration as described in request item 4, wherein the 3D model is a point cloud model.

8. The vector iterative ground point segmentation method as described in claim 4, wherein after the step of comparing the result of determining the angle with the preset threshold and generating a plurality of ground points or a plurality of non-ground points, it further comprises: receiving a portion of the ground points through a classification module of the central processing unit and reclassifying them to form an unknown label point.

9. The vector-based iterative ground point segmentation method as described in claim 8, further comprising, after the step of receiving a portion of the ground points through a classification module of the central processing unit and reclassifying them into an unknown label point: the analysis and calculation module of the central processing unit receives the unknown label point and performs analysis and comparison with a portion of the ground points to define the unknown label point and relabel it as the ground point.

10. The vector iterative ground point segmentation method as described in claim 9, further comprising, after the step of defining the unknown label point and relabeling it as the ground point: the analysis and calculation module retrieves the closest unknown label point relabeled as the ground point, compares and analyzes the closest unknown label point with a certain value stored in the storage module of the central processing unit, and calculates the closest unknown label point using the vector iterative ground segmentation method, thereby determining whether the closest unknown label point is a ground point or a non-ground point.