Spatial Block Data Structure for LIDAR Point Cloud Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in efficiently processing, storing, and manipulating extremely large amounts of three-dimensional scan data, such as those generated by LIDAR systems, due to limitations in computer memory and existing data compression techniques, leading to difficulties in registration, measurement, and modeling of complex objects.
Innovation Solution
The system organizes and processes data by partitioning it into separately addressable blocks based on spatial density and area, allowing for efficient storage and retrieval, enabling the construction of a unified data structure that can handle arbitrary numbers of points and support real-time queries and visualization, using a spatial block structure that can be built 'out of core' on secondary storage, with data compression and lazy evaluation techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LIDAR scanning technologies are used to collect billions of point samples, then the measurement precision and data quantity are improved, but the device complexity and storage requirements worsen
Solution Approach 1:
The patent divides the massive point cloud data into smaller, manageable blocks that can be processed and stored independently. This segmentation allows the system to handle billions of points by organizing them into discrete units that fit within standard memory constraints, resolving the contradiction between high measurement precision and device complexity.
Solution Approach 2:
The patent introduces a hierarchical data structure that adds an organizational dimension to the point cloud data. By structuring the data in multiple levels of hierarchy, the system can efficiently manage and query billions of points without requiring proportional increases in memory or processing complexity.
2Ease of operation
If standard off-the-shelf CAD programs are used to process data, then the ease of operation is improved, but the productivity worsens when handling large data volumes
Solution Approach 1:
The patent creates a universal data structure that can handle both small and large point cloud datasets using the same processing framework. This multi-functional approach allows standard CAD operations to work efficiently on massive datasets without requiring specialized software, maintaining ease of operation while dramatically improving productivity.
Solution Approach 2:
The patent changes the fundamental parameter of data organization from traditional flat arrays to a hierarchical block structure. This parameter change enables the same software to process data of any size efficiently, resolving the contradiction between ease of operation and productivity for large datasets.
3Ease of manufacture
If data is stored in traditional formats, then the ease of manufacture and storage is improved, but the loss of information occurs during compression and manipulation
Solution Approach 1:
The patent segments the data into blocks that maintain their individual integrity and can be processed independently. This segmentation prevents information loss by ensuring that no data is discarded during compression or manipulation, while still maintaining ease of storage and manufacture through efficient block-based organization.
4Productivity
If the data is processed in real-time, then the productivity is improved, but the use of energy worsens due to the large volume of data
Solution Approach 1:
The patent segments the massive dataset into smaller blocks that can be processed in parallel or sequentially as needed. This segmentation enables real-time processing of only the necessary data portions, improving productivity while reducing energy consumption by avoiding the need to process the entire dataset at once.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for unprecedented efficiency in manipulating and visualizing large data sets, reducing processing time and effort, and enabling the handling of previously intractable data volumes by allowing actual point data to be queried and displayed quickly, with scalable geometric representations and dynamic simplification of subsets.
Implementation Method 1
Recent advances in technology such as LIDAR scanning technologies have resulted in the ability to collect billions of point samples on physical surfaces
Implementation Method 2
In a LIDAR process, a laser beam scans across a view that encompasses the structure of interest
Data Source
AI summary
Large data sets can be stored and processed in real time by combining and registering the large data sets into a single data set. The data can be stored in a data tree structure formed of layers of spatially organized blocks of data. Such storage allows portions of the data to be viewed efficiently, displaying actual point data at an acceptable resolution for the viewing mechanism. Density limited queries can be executed that allow sub-sampling to be done directly and evenly without geometric constraint, to provide a subset of points that is limited in size and includes a spatially-even decomposition of that set of points. This allows the system as a whole to support arbitrarily large point sets while allowing full partitioning functionality, which is efficient to use in both time and space.