3D Model Creation from Sparse Point Cloud Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in creating three-dimensional models of target objects with unevenly spaced inter-point distances and only partial point cloud data, particularly for objects like cables on utility poles.
Innovation Solution
The method involves creating a three-dimensional model from point cloud data, superimposing it onto an image of the target object, and correcting the model based on the superimposed image, allowing for the creation of models even with uneven point spacing and partial data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If point cloud data is acquired using a fixed three-dimensional laser scanner, then dense point cloud is produced at short distance, but sparse point cloud is produced at long distance
Solution Approach 1:
The patent divides the point cloud processing into multiple stages: first creating a preliminary three-dimensional model from available point cloud data, then superimposing it on images to identify missing portions, and finally generating additional models only for those missing portions. This segmentation allows the system to handle uneven point spacing by processing data in manageable segments rather than requiring uniform distribution throughout the entire object.
Solution Approach 2:
The patent introduces images as an intermediary medium between the sparse point cloud data and the final three-dimensional model. The images serve as a mediator that provides visual information about the target object, allowing the system to infer and generate missing portions of the model that cannot be directly obtained from the sparse point cloud data alone.
2Manufacturing precision
If points are interpolated to form scan lines, then three-dimensional model can be created, but no interpolation can be performed when distance between point clouds is large
Solution Approach 1:
The patent implements a feedback mechanism where the preliminary three-dimensional model is superimposed on images of the target object, and the results are used to identify missing portions. This feedback loop allows the system to detect areas where point cloud data is insufficient and adjust the modeling process accordingly, enabling accurate model creation even when inter-point distances are large in certain regions.
Solution Approach 2:
Instead of requiring complete interpolation across the entire object, the patent applies partial action by generating three-dimensional models only for the missing portions identified through image superposition. This approach allows the system to create accurate models in critical areas without requiring uniform interpolation throughout the entire object, thereby overcoming the limitation of large inter-point distances.
3Adaptability or versatility
If three-dimensional model is created from partial point cloud data, then modeling of small diameter objects becomes possible, but model accuracy may be compromised
Solution Approach 1:
The patent uses images as an intermediary to bridge the gap between partial point cloud data and accurate three-dimensional modeling. The images provide additional visual information that compensates for the sparsity of point cloud data, enabling the system to generate accurate models of small diameter objects even when only partial point cloud data is available.
Solution Approach 2:
The patent changes the approach from relying solely on point cloud data density to using a combination of point cloud data and image information. By altering the data sources and processing parameters, the system can accurately model small diameter objects with unevenly spaced point clouds, overcoming the traditional limitation that required dense, uniformly distributed point data.
Data Source
AI summary
An object of the present disclosure is to enable a three-dimensional model to be created even for a target object which has unevenly spaced inter-point distances and only a part of a point cloud.According to the present disclosure, there are provided an apparatus and a method in which, a three-dimensional model of a target object is created from point cloud data in which each point represents three-dimensional coordinates, the three-dimensional model is superimposed on an image in which a target object of the three-dimensional model is photographed, a superimposed image generated by the superimposition is displayed, and when a range of the target object in the superimposed image is input, a three-dimensional model is created again using point cloud data in which a point is located in a range of the superimposed image.