Space-Time Intensity Peak Detection for Real-Time 3D Triangulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional peak finding algorithms in 3D machine vision systems using light triangulation suffer from artefacts due to laser line width covering multiple pixels, especially at object discontinuities, and existing space-time analysis methods are computationally heavy and require full space-time volumes, making them impractical for real-time implementation.
Innovation Solution
A method that determines intensity peak positions in a space-time volume by locally analyzing a partial subset of image frames, using hypothetical peak positions and iterative space-time analysis to refine peak positions, reducing computational requirements and enabling real-time processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional peak finding algorithms are used to detect intensity peaks in image data, then the detection process is simple and fast, but artefacts occur due to laser line width covering multiple pixels especially at object discontinuities
Solution Approach 1:
The patent transitions from analyzing single 2D image frames to analyzing a 3D space-time volume formed by multiple image frames over time. By adding the time dimension and performing space-time analysis, the system can distinguish true peak positions from artefacts caused by laser line width, as true peaks maintain consistency across time while artefacts do not.
2Measurement precision
If space-time analysis is performed on the full space-time volume to improve peak position accuracy, then measurement precision improves, but computational complexity and data storage requirements increase significantly
Solution Approach 1:
The patent divides the full space-time volume into multiple smaller sub-volumes or regions of interest. By performing space-time analysis on these segmented portions rather than the entire volume, the system maintains measurement precision while significantly reducing computational complexity and data storage requirements.
Solution Approach 2:
The patent performs space-time analysis on a partial subset of the space-time volume rather than the complete volume. This partial action approach provides sufficient peak position accuracy for practical applications while avoiding the excessive computational burden of analyzing the entire space-time volume.
3Reliability
If space-time analysis is performed on the full space-time volume to reduce artefacts, then reliability improves, but processing time increases making real-time implementation impractical
Solution Approach 1:
The patent segments the space-time volume into smaller sub-volumes that can be processed independently and more quickly. This segmentation maintains the reliability benefits of space-time analysis by preserving the temporal dimension for artefact rejection, while enabling parallel processing and reducing overall processing time for real-time implementation.
Solution Approach 2:
The patent performs preliminary processing steps such as identifying candidate peak positions in individual frames before performing the more computationally intensive space-time analysis. This preliminary action reduces the amount of data requiring full space-time analysis, thereby improving processing speed while maintaining reliability.
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
Improves peak position accuracy and reliability while facilitating practical implementation by reducing computational resources and data storage needs, allowing for real-time 3D image generation with reduced artefacts.
Implementation Method 1
sensing of light reflected from a measure object as part of light triangulation
Implementation Method 2
light triangulation performed by an imaging system
Data Source
AI summary
Determination of information regarding an intensity peak position in a space-time volume (360; 361) formed by image frames generated by an image sensor (331) from sensing of light reflected from a measure object (320) as part of light triangulation. The space-time volume (360; 361) is further associated with space-time trajectories relating to how feature points of the measure object (320) map to positions in the space-time volume (360; 361). A first hypothetical intensity peak position, HIPP1, (551a; 651a) is obtained (701 in said space time volume (360; 361). A first space time analysis position, STAP1, (552a; 652a) is computed (702) based on space-time analysis performed locally around the HIPP1 (551a; 651a) and along a first space time trajectory associated with the HIPP1 (551a, 651a). Said information regarding the intensity peak position is determined (703) based on the HIPP1 (551a; 651a) and the STAP1 (552a; 652a).