3D Ray-Based Background Detection for Embedded Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional background detection techniques for 3D object detection systems are inaccurate, inefficient, and require excessive computational and memory resources, making them unsuitable for embedded systems.
Innovation Solution
A method and system that divide a 3D space into rays, define peaks and catchment regions, and update catchment distances to accurately distinguish between static and dynamic objects using minimal computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional background detection techniques are used, then background detection can be performed, but accuracy is poor and computational resources are excessively consumed
Solution Approach 1:
The patent segments the 3D space into multiple rays extending from the sensor, with each ray processed independently to identify background portions. This segmentation allows the system to focus computational resources on specific spatial regions rather than processing the entire point cloud uniformly, thereby improving detection accuracy while reducing overall computational complexity
Solution Approach 2:
The patent transforms the background detection problem from traditional 2D image space to 3D point cloud space by introducing radial distance as an additional dimension. By defining background based on spatial distribution patterns in three dimensions (azimuth, elevation, and range), the system achieves more accurate background identification while maintaining computational efficiency through the structured ray-based approach
2Productivity
If conventional background detection techniques are used, then background detection can be performed, but processing speed is slow and efficiency is poor
Solution Approach 1:
By dividing the point cloud into multiple rays and processing each ray independently, the system enables parallel processing of spatial regions. This segmentation strategy significantly improves processing speed while maintaining detection reliability through consistent application of background detection logic across all rays
Solution Approach 2:
The patent performs preliminary classification of points as background or foreground based on their spatial distribution patterns before subsequent processing steps. This preliminary action filters out background portions early in the pipeline, reducing the computational load for later object detection tasks while ensuring accurate classification through the ray-based spatial analysis
3Quantity of substance
If conventional background detection techniques are used, then background detection can be performed, but memory usage is excessive
Solution Approach 1:
The patent processes each ray independently and maintains minimal state information for background detection, avoiding the need to store entire point cloud datasets in memory. This segmented processing approach dramatically reduces memory requirements while preserving detection accuracy through localized spatial analysis
Solution Approach 2:
The patent extracts and removes background portions from the point cloud data through the ray-based detection process. By identifying and separating background points based on their spatial distribution characteristics, the system reduces the amount of data that needs to be retained in memory for subsequent processing, thereby optimizing memory usage without sacrificing detection precision
Data Source
AI summary
A method for determining a background includes: receiving, via a sensor, a plurality of data points for a 3D space; dividing the 3D space into a plurality of rays extending from the sensor; defining a plurality of peaks for each ray; defining a catchment region for each peak for each ray and including a catchment distance from the corresponding peak towards and away from the sensor; and updating the catchment distance of the catchment region. Each data point is enclosed by a corresponding ray. Each peak is located at a peak distance from the sensor and includes a peak height. For each data point from the plurality of data points, the method includes: determining the corresponding ray enclosing the data point; determining a containing peak for which the data point lies within the catchment region; and incrementing the peak height of the containing peak by a peak increment value.