Workpiece Image Search Using Shape Deformation for Aspect Ratio Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing workpiece image search methods struggle to accurately detect similar workpiece shapes due to reliance on simple pixel comparisons, which are inadequate for identifying shapes with varying aspect ratios.
Innovation Solution
A workpiece image search apparatus that deforms one workpiece image to match the shape of another, using a similarity calculation unit to compare the deformed image with the original, allowing for shape feature extraction and deformation to enhance similarity detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple pixel comparison between bitmaps is used, then the comparison process is simple, but the similarity detection accuracy deteriorates when aspect ratios differ
Solution Approach 1:
The patent applies preliminary action by performing deformation processing on workpiece images before similarity comparison. The system pre-processes images by deforming them to standardized aspect ratios or transforming them to compensate for dimensional differences, ensuring that subsequent pixel-based comparisons accurately reflect true shape similarity rather than being confounded by aspect ratio variations.
Solution Approach 2:
The patent changes parameters by applying deformation transformations to workpiece images, modifying their dimensional parameters (width, height, aspect ratio) while preserving shape characteristics. This allows the system to normalize images for comparison purposes, enabling accurate similarity detection across workpieces with different original dimensions by transforming them to a common reference frame.
2Measurement precision
If workpiece images are deformed to reduce shape differences, then shape similarity detection accuracy is improved, but the processing complexity increases
Solution Approach 1:
The patent applies copying by creating deformed copies of original workpiece images for comparison purposes. Instead of modifying the original stored images, the system generates transformed copies with adjusted dimensions and aspect ratios, preserving the source data while enabling accurate similarity comparisons through these derived representations.
Solution Approach 2:
The deformation processing acts as an intermediary step between image acquisition and similarity comparison. This intermediate transformation layer converts images into a standardized form that facilitates accurate comparison, serving as a mediator that bridges the gap between varied input formats and the requirements for precise similarity measurement.
3Measurement precision
If deformation processing is applied to all workpiece images, then shape matching accuracy is improved, but the processing time increases
Solution Approach 1:
The patent applies partial action by selectively performing deformation processing only on images that require it for accurate comparison. Rather than uniformly processing all images, the system identifies which images need dimensional adjustment based on their relationship to the target workpiece, applying deformation only where necessary to achieve shape matching, thus reducing overall processing time while maintaining accuracy.
Data Source
AI summary
A workpiece image search apparatus includes: a workpiece image deformation unit that generates a third workpiece image by deforming a second workpiece image so that a difference in workpiece shape between a first workpiece image and the second workpiece image becomes smaller, wherein the first workpiece image is obtained by projecting a first workpiece shape of a first workpiece on a two-dimensional plane, and the second workpiece image is obtained by projecting a second workpiece shape of a second workpiece on a two-dimensional plane; and a similarity calculation unit that calculates a similarity between the first workpiece shape and the second workpiece shape by comparing the third workpiece image with the first workpiece image.


