Point Cloud Filtering via 2D Histogram Slope Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for mapping physical spaces, such as warehouses, are time-consuming and prone to inaccuracies due to human error, leading to incomplete or outdated maps.
Innovation Solution
The technology refines point clouds by selecting an axis, compressing sections into 2D histograms, evaluating voxels, and applying filters like histogram filtering, k-means filtering, and slicing filtering to remove noise and enhance feature representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual mapping is performed by users walking around with scanners, then blueprints and maps can be generated, but the process is time-consuming and prone to human error
Solution Approach 1:
The system enables automatic point cloud filtering and map generation without continuous human intervention. The computer system automatically processes point cloud data, applies filtering algorithms, and generates updated maps, allowing the mapping system to serve itself and eliminating the need for manual updates.
Solution Approach 2:
The patent replaces manual mechanical scanning and mapping processes with automated computational methods. Instead of users physically walking around with scanners and manually creating maps, the system uses automated point cloud processing, histogram filtering, and algorithmic map generation to produce accurate maps efficiently.
2Measurement precision
If point clouds include noise and inaccurate points, then rough approximations of physical structures can be obtained, but the point clouds cannot accurately represent physical structures and surfaces
Solution Approach 1:
The patent segments the point cloud processing into distinct stages: initial noise removal through histogram filtering, followed by feature enhancement using k-means filtering, and final refinement with slicing filtering. Each segment addresses specific aspects of point cloud quality, systematically improving accuracy while managing complexity through structured processing steps.
Solution Approach 2:
The system changes parameters of the point cloud data through multiple filtering passes, adjusting density, distribution, and quality characteristics. By modifying parameters such as point density and spatial distribution through controlled filtering operations, the system transforms noisy point clouds into accurate representations of physical structures.
3Reliability
If maps are manually updated whenever changes are made to physical spaces, then blueprints can remain current, but the process is time-consuming and maps may still be outdated
Solution Approach 1:
The patent implements continuous automated map updating by continuously processing point cloud data and generating updated maps without interruption. The system maintains an ongoing process of data acquisition, filtering, and map regeneration, ensuring maps are continuously current with physical space changes without requiring discrete manual update actions.
Solution Approach 2:
The mapping system automatically detects changes in physical spaces and updates maps without human intervention. The computer system self-manages the entire update process from point cloud acquisition through filtering and map generation, eliminating the need for manual map updates and ensuring maps remain current with physical changes.
Data Source
AI summary
This specification describes systems and methods for refining point cloud data. Methods can include receiving point cloud data for a physical space, iteratively selecting points along an x, y, and z dimension, clustering the selected points into 2D histograms, determining a slope value for each 2D histogram, and removing, based on the slope value exceeding a predetermined value, points from the point cloud data. Methods can also include iteratively voxelizing each 2D histogram into predetermined mesh sizes, summating points in each voxelized 2D histogram, removing, based on determining the summation is below a predetermined sum value, points from the point cloud data, keeping, based on determining that a number of points in each voxelized 2D histogram exceeds a threshold value, a center point, selecting, for each histogram, a point, identifying, nearest neighbors in the point cloud data, removing the identified nearest neighbors from the data, and returning remaining points.