Point Cloud Densification Using RGB Image Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional three-dimensional scanning range finders struggle to capture detailed, colored representations of physical surfaces due to limited resolution and lack of color data, making it difficult to distinguish fine architectural details in urban areas and other complex terrains.
Innovation Solution
Integrating image data with measured surface points to create additional 'implied' points, enhancing the resolution and adding color to both measured and implied points, resulting in a higher-resolution, colored three-dimensional representation of the physical surface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional LiDAR is used for three-dimensional scanning, then measurement range and depth accuracy are achieved, but resolution and color information are insufficient
Solution Approach 1:
The patent combines LiDAR depth measurement data with RGB image data to create a unified three-dimensional representation. The LiDAR point cloud provides accurate depth information while the RGB images provide color information, and the two data sources are merged through coordinate transformation and point cloud registration to produce a colored three-dimensional model that preserves both depth accuracy and color information.
Solution Approach 2:
The patent uses feature matching and coordinate transformation as intermediary processes to bridge the LiDAR depth data and RGB image data. By identifying corresponding features in both data sources and establishing coordinate relationships, the system enables accurate integration of depth and color information without direct physical interaction between the sensors.
2Area of stationary object
If LiDAR scanning is performed over large areas, then coverage is achieved, but resolution of fine details deteriorates
Solution Approach 1:
The patent merges LiDAR point cloud data with high-resolution RGB image data to compensate for the resolution limitations of LiDAR when surveying large areas. The RGB images captured by cameras provide detailed visual information that supplements the coarser LiDAR measurements, enabling fine architectural details to be resolved even when the LiDAR point density is low.
Solution Approach 2:
The patent applies different data sources with different strengths to different regions of the survey area. In areas where fine detail is critical, the high-resolution RGB image data provides enhanced local quality, while the LiDAR data provides overall structural accuracy. This allows the system to maintain high resolution for important features while covering large areas.
3Loss of information
If flash LiDAR is used to obtain both range and color information, then color data is achieved, but cost and surface detail sufficiency worsen
Solution Approach 1:
The patent segments the data acquisition functions between separate devices: LiDAR for depth measurement and RGB cameras for color capture. This segmentation allows the use of成熟, cost-effective technologies for each function rather than requiring expensive integrated flash LiDAR systems. The segmented approach maintains color information capability while reducing overall system cost.
Solution Approach 2:
The patent employs multi-functional sensor platforms that can perform both LiDAR scanning and RGB imaging using the same mounting and positioning infrastructure. This universality reduces redundant hardware and simplifies system integration, thereby reducing complexity and cost while achieving both depth and color information from a single coordinated system.
Data Source
AI summary
Image data obtained from an image sampling of a physical surface is integrated with position data obtained from a three-dimensional surface sampling of the same physical surface by combining data from the images with the measured surface points from the surface sampling to create additional “implied” surface points between the measured surface points. Thus, the originally obtained point cloud of measured surface points is densified by adding the implied surface points. Moreover, the image data can be used to apply colors to both the implied data points and the measured data points, resulting in a colored three-dimensional representation of the physical surface that is of higher resolution than a representation obtained from only the measured surface points.