Stress Distribution Image Normalization for Abnormality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to compare stress distribution images under varying conditions such as different vehicle sizes and weights, making it difficult to identify abnormal portions due to differences in measurement conditions, date, and time.
Innovation Solution
A stress distribution image processing device that normalizes images by designating a region with stress equal to or greater than a predetermined threshold value as the normalization region, allowing for comparison of images under different conditions by adjusting pixel values to a standardized range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If stress distribution images are compared directly under different measurement conditions, then the comparison process is simple, but the ability to identify abnormalities is poor
Solution Approach 1:
The patent applies parameter changes by normalizing stress distribution images through mathematical transformations. Specifically, it converts raw stress values into standardized parameters using reference data from normal states, enabling meaningful comparisons across different measurement conditions while maintaining abnormality detection accuracy
Solution Approach 2:
The patent introduces reference data from normal states as an intermediary element. This reference data serves as a mediator between different measurement conditions, providing a common baseline that enables accurate comparison and abnormality identification across varying experimental setups
2Measurement precision
If stress distribution images are normalized using a predetermined threshold, then the abnormality identification accuracy is improved, but the processing complexity increases
Solution Approach 1:
The patent transforms the complex normalization problem into a simpler parameter transformation task. By using predetermined thresholds and reference-based scaling, it converts varied stress distributions into standardized parameters through straightforward mathematical operations, reducing processing complexity while maintaining accuracy
Solution Approach 2:
The patent performs preliminary normalization using predetermined thresholds before detailed analysis. This preliminary action prepares the data in advance, establishing a standardized basis that simplifies subsequent abnormality detection and reduces the complexity of the overall processing pipeline
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A stress distribution image processing device including: a region designation unit configured to designate a normalization region which includes a portion of stress equal to or larger than a predetermined threshold value in an entire screen of a stress distribution image of a target object, and the normalization region being to be used for performing normalization; and a normalization unit configured to normalize all pixels in the normalization region based on stress values in all pixels in the normalization region to obtain a normalized image.