Starburst Algorithm for LiDAR Point Cloud Object Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
LiDAR sensors generate large amounts of data due to their ability to detect smaller particles, requiring significant processing power for object detection and classification, which is inefficient and resource-intensive.
Innovation Solution
A system and method utilizing a starburst algorithm that processes three-dimensional point cloud data from LiDAR sensors to identify clusters of points, determine a center point, project rays to generate shapes, and compare these shapes to candidate shapes for classification, thereby reducing data processing needs and improving classification efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR sensors are used to detect smaller particles, then measurement precision is improved, but productivity deteriorates due to significantly larger amounts of data generated requiring significant processing power
Solution Approach 1:
The patent segments the point cloud data into multiple regions of interest (ROIs) based on spatial location and object type. By dividing the large dataset into smaller manageable segments, the processing system can handle each region independently, reducing the computational burden on any single processing unit while maintaining comprehensive object detection capabilities across the entire scene.
Solution Approach 2:
The patent extracts only the essential features and characteristics needed for object classification from the massive point cloud data, rather than processing all raw data points. By taking out and retaining only the critical information (such as geometric features, intensity values, and spatial relationships), the system achieves efficient processing while preserving the ability to accurately classify objects.
2Measurement precision
If significant processing power is applied to process LiDAR data, then object classification accuracy is improved, but use of energy increases
Solution Approach 1:
The patent performs preliminary actions by pre-processing the point cloud data to identify and filter out irrelevant points before the main classification process. By conducting initial filtering, noise removal, and feature extraction in advance, the system reduces the amount of data that requires intensive processing during the classification stage, thereby lowering energy consumption while maintaining classification accuracy.
Solution Approach 2:
The patent applies partial action by processing only the portions of data that are necessary for accurate classification, rather than exhaustively processing all data points. By identifying and focusing computational resources on critical regions and features, the system achieves sufficient classification accuracy with reduced energy expenditure.
3Reliability
If all point cloud data is processed for object detection, then reliability of detection is improved, but loss of time increases due to extensive processing requirements
Solution Approach 1:
The patent segments the point cloud data into multiple regions of interest (ROIs) based on spatial location and object type. By dividing the large dataset into smaller manageable segments, the processing system can handle each region independently, reducing the computational burden on any single processing unit while maintaining comprehensive object detection capabilities across the entire scene.
Solution Approach 2:
The patent performs preliminary actions by pre-processing the point cloud data to identify and filter out irrelevant points before the main classification process. By conducting initial filtering, noise removal, and feature extraction in advance, the system reduces the amount of data that requires intensive processing during the classification stage, thereby lowering energy consumption while maintaining classification accuracy.
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
The system effectively classifies objects by reducing unnecessary data processing and improving processing performance, enabling efficient object detection and classification in real-time applications such as autonomous vehicles.
Implementation Method 1
A LiDAR sensor operates by emitting laser pulses and detecting the laser pulses reflected back by objects towards the unit
Data Source
AI summary
A system for classifying an object may include one or more processors, a sensor and a memory device. The memory device may include a data collection module, a starburst module, and an object classifying module. The modules have instructions that when executed by the one or more processors cause the one or more processors to obtain three dimensional point cloud data from the sensor, identify at least one cluster of points representing the object within the three dimensional point cloud data, identify a center point of the at least one cluster of points, project a plurality of rays from the center point to points of the at least one cluster of points to generate a shape, compare the shape to a plurality of candidate shapes, and classify the object when the shape matches at least one of the plurality of candidate shapes.