Convex Hull Feature Extraction for Dynamic Spatial Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional localization and object recognition technologies are inadequate for dynamic environments, as they assume a static environment, making it difficult to separate and recognize structures, objects, and dynamics in indoor spaces where these elements are mixed.
Innovation Solution
A method for extracting outer space feature information from spatial geometric data using three-dimensional point cloud data, employing a convex hull method to distinguish and separate static and dynamic elements, allowing conventional technologies to be applied in dynamic environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional localization and object recognition technologies are used in dynamic environments, then the technologies can be applied in real-world scenarios, but the assumption of static environment causes inaccurate feature extraction and recognition
Solution Approach 1:
The patent segments spatial geometric data into multiple sampling planes and processes each plane independently using convex hull algorithms. This segmentation allows the system to handle dynamic environments by treating different spatial regions separately, maintaining measurement precision while adapting to real-world complexity.
Solution Approach 2:
The patent transforms three-dimensional point cloud data into two-dimensional sampling planes for processing. By projecting 3D spatial geometric data onto 2D planes and applying convex hull methods in this reduced dimensionality, the system achieves efficient feature extraction that works accurately in dynamic environments while reducing computational complexity.
2Loss of information
If all spatial geometric data including structures, objects, and dynamics are processed together, then complete environmental information is obtained, but calculation load increases and processing efficiency decreases
Solution Approach 1:
The patent extracts only the essential outer boundary features of structures using convex hull algorithms on sampling planes. By taking out and processing only the critical geometric boundaries rather than all detailed spatial data, the system maintains complete structural information while significantly reducing calculation load and improving processing efficiency.
Solution Approach 2:
The patent applies different processing approaches to different spatial regions by creating multiple sampling planes with local orientations. Each sampling plane processes local geometric features with appropriate algorithms, ensuring that important structural information is preserved while reducing overall computational requirements through localized processing.
3Device complexity
If conventional methods process spatial data without separating static and dynamic elements, then implementation is simpler, but the ability to distinguish structures from moving objects deteriorates
Solution Approach 1:
The patent performs preliminary processing by organizing spatial geometric data into multiple sampling planes before applying convex hull algorithms. This preliminary structuring of data into manageable planes simplifies the subsequent processing steps while enabling accurate distinction between static structures and dynamic elements through systematic geometric analysis.
Solution Approach 2:
The patent creates simplified two-dimensional representations (copies) of three-dimensional spatial data through sampling planes. These 2D copies preserve the essential geometric characteristics needed for distinguishing static from dynamic elements while being computationally simpler to process, thus reducing implementation complexity without sacrificing measurement precision.
Data Source
Figure 1A
Figure 1B~1C
Figure 1D
AI summary
The present invention provides a method for extracting outer space feature information from spatial geometric data, the method comprising: an input step S10 of inputting spatial geometric data for a target region; a sampling step S20 of determining a sample by selecting an arbitrary area for the spatial geometric data input in the input step using a preset selection method; a feature extraction step S30 of acquiring feature information for a corresponding sampling plane using a convex hull method based on sampling information including sampling plane information of the spatial geometric data for a sampling plane selected in the sampling step. The sampling step and the feature extraction step are repeatedly performed in a preset manner.