Image Portion Detection via Filtering and Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for automatically detecting portions of interest in image data, such as ultrasonic NDT scans and road lane markings, face challenges due to noise and variability in data, leading to incomplete defect detection and increased manual inspection time.
Innovation Solution
A computer-implemented method that identifies parts of the image or matrix data satisfying a preset threshold, applies filters to isolate likely portions of interest, sorts objects into clusters based on predefined criteria, and uses known characteristics to accurately identify the portion of interest, enabling efficient detection of layers or boundaries regardless of discontinuities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual visual inspection is used to identify portions of interest in ultrasonic NDT scans, then detection accuracy can be maintained through operator expertise, but inspection time and human effort increase significantly
Solution Approach 1:
The system performs self-inspection by automatically identifying the back wall and portions of interest in ultrasonic NDT scans without requiring manual visual inspection. The computer-implemented method autonomously processes scan data, applies filtering and clustering algorithms, and generates inspection results, enabling the system to serve itself rather than relying on human operators for routine detection tasks.
2Productivity
If automated detection methods are applied to identify portions of interest in image data, then inspection speed and productivity improve, but detection reliability deteriorates due to noise and data variability
Solution Approach 1:
The detection process is segmented into distinct processing stages: initial object identification from scan data, application of filtering operations to remove noise, clustering of filtered objects based on spatial and characteristic criteria, and final selection of portions of interest. This segmentation allows each stage to focus on specific tasks, improving overall reliability while maintaining automated processing speed.
Solution Approach 2:
Filtering operations serve as an intermediary step between raw scan data and final detection results. The filter acts as a mediator that processes intermediate objects, removing noise and variability before the clustering stage. This intermediary processing enhances detection reliability by ensuring that only quality-filtered data proceeds to subsequent analysis stages.
3Measurement precision
If filtering operations are applied to remove noise from detected objects, then detection precision improves, but processing complexity increases
Solution Approach 1:
The filtering operation modifies parameters of detected objects, such as removing objects below certain size thresholds or adjusting object properties based on scan data characteristics. By changing parameters systematically, the filter improves detection precision while maintaining a manageable processing complexity through rule-based parameter adjustment rather than complex algorithmic processing.
Data Source
Figure 1(a)
Figure 1(b)
Figure 1(c)
AI summary
A computer-implemented method of automatically locating a portion of interest in image or matrix data derived from an item under consideration comprises: identifying parts of the image or matrix data that satisfy a preset threshold as objects which are possibly parts of the portion of the interest; applying at least one preselected filter to the data corresponding to the objects to find a set of objects consisting of the objects most likely to be part of the portion of interest; sorting the objects of the set into clusters according to a predefined criterion; and using a known characteristic of the portion of interest to identify which one of the clusters corresponds to the portion of interest.