Edge Feature Video Tool Robust Edge Discrimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine vision inspection systems face challenges in reliably detecting edges in regions of interest with multiple closely spaced edges, as users find it difficult to adjust edge detection parameters, particularly for unskilled operators, due to the complexity of edge conditions and part-to-part variations.
Innovation Solution
A method and system that improve edge detection reliability by using a video tool with a region of interest (ROI) and edge detection parameters such as edge gradient threshold, profile scan direction, and gradient prominence-counting parameter, allowing automatic determination of these parameters to accurately locate edges within a robust extremum margin, thereby simplifying the detection process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual adjustment of edge detection parameters is used, then flexibility in handling different edge conditions is improved, but ease of operation deteriorates due to complexity for unskilled users
Solution Approach 1:
The system performs self-adjustment of edge detection parameters by automatically analyzing the image data and setting appropriate threshold values and scan directions without requiring user intervention. The computer executes algorithms that adaptively determine optimal parameters based on the actual edge characteristics in the image, making the system serve itself rather than requiring expert operation.
Solution Approach 2:
The system automatically changes detection parameters such as gradient thresholds and scan directions based on the analyzed image characteristics. By dynamically adjusting these parameters according to the specific edge conditions detected in the image, the system achieves adaptability to different edge types and orientations without manual parameter tuning.
2Ease of operation
If automatic determination of edge detection parameters is used, then ease of operation is improved, but measurement precision may deteriorate due to loss of manual control
Solution Approach 1:
The system uses feedback from the image analysis process to automatically adjust parameters. By analyzing the detected edges and their characteristics, the system feeds this information back to refine parameter selection, ensuring that automatic parameter determination achieves high measurement precision through iterative optimization rather than fixed preset values.
Solution Approach 2:
The patent replaces manual mechanical adjustment of parameters with automated computational algorithms. The computer-based system substitutes human operator intervention with algorithmic parameter determination, using image processing and pattern recognition to automatically set optimal detection parameters, thereby maintaining precision while improving ease of operation.
3Measurement precision
If complex edge detection algorithms are used to handle multiple closely spaced edges, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system segments the edge detection process into distinct computational stages: gradient calculation, threshold determination, scan direction selection, and edge location refinement. By dividing the complex detection task into modular segments that can be processed sequentially, the system achieves high precision for closely spaced edges while managing algorithmic complexity through structured organization of detection operations.
Solution Approach 2:
The system introduces additional computational dimensions by analyzing edges in multiple scan directions and using multi-parameter optimization. By extending the detection approach from simple thresholding to multi-dimensional gradient analysis and prominence counting, the system achieves superior precision for complex edge patterns while organizing the complexity across multiple analytical dimensions rather than increasing single-algorithm complexity.
Data Source
AI summary
A reliable method for discriminating between a plurality of edges in a region of interest of an edge feature video tool in a machine vision system comprises determining a scan direction and an intensity gradient threshold value, and defining associated gradient prominences. The gradient threshold value may be required to fall within a maximum range that is based on certain characteristics of an intensity gradient profile derived from an image of the region of interest. Gradient prominences are defined by limits at sequential intersections between the intensity gradient profile and the edge gradient threshold. A single prominence is allowed to include gradient extrema corresponding to a plurality of respective edges. A gradient prominence-counting parameter is automatically determined that is indicative of the location of the selected edge in relation to the defined gradient prominences. The gradient prominence-counting parameter may correspond to the scan direction.