Hardness Tester Indentation Region Extraction Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hardness testers face challenges in accurately extracting the indentation region from measurement images, especially for samples with complex brightness distributions or surface features like engravings and cutting traces, due to the simplicity of background subtraction methods.
Innovation Solution
A hardness testing system that employs multiple indentation region extraction methods, including binary conversion, reduction and expansion processes, and pattern matching, to accurately identify the indentation region based on images acquired before and after indentation formation, ensuring accurate hardness calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If background subtraction method is used to extract indentation region, then the extraction process is simple, but the extraction accuracy deteriorates for samples with complex brightness distribution
Solution Approach 1:
The patent segments the indentation region extraction process into multiple independent methods (gradient method, threshold method, contour method), each handling different aspects of the extraction task. This segmentation allows the system to select the most appropriate method for each specific sample type, resolving the contradiction between simplicity and accuracy.
Solution Approach 2:
The patent changes the processing parameters and approaches by implementing multiple extraction methods with different algorithms and criteria. Instead of using a single fixed method, the system adapts parameters such as gradient thresholds, binary conversion levels, and contour detection sensitivity to match the specific characteristics of the sample being measured.
2Measurement precision
If multiple indentation region extraction methods are employed, then the extraction accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent implements a dynamic selection mechanism that automatically chooses the most appropriate extraction method based on the characteristics of the measurement image. The system evaluates image properties such as brightness distribution, noise levels, and indentation depth to dynamically select from multiple extraction methods, thereby managing complexity while maintaining high accuracy.
Solution Approach 2:
The system incorporates feedback mechanisms where the results from different extraction methods are evaluated and compared. The determination unit uses feedback from the extraction results to select the most reliable method and can adjust processing parameters accordingly, ensuring accurate extraction without requiring all methods to be simultaneously active, thus controlling system complexity.
3Loss of time
If simple background subtraction is used, then the processing time is short, but the measurement accuracy deteriorates for samples with engravings or cutting traces
Solution Approach 1:
The patent performs preliminary analysis of the measurement image characteristics before selecting the extraction method. By pre-evaluating image properties such as brightness distribution patterns, presence of engravings, or cutting traces, the system can quickly select the appropriate extraction method without requiring exhaustive processing of all possible methods, thus minimizing processing time while ensuring accuracy.
Solution Approach 2:
The system applies different extraction methods to different regions or aspects of the image based on local characteristics. For example, gradient-based methods may be applied to regions with clear edges, while threshold methods are used for uniform regions, and contour methods handle complex boundaries. This localized approach improves accuracy for specific sample features without requiring complete reprocessing of the entire image with all methods.
Data Source
AI summary
A hardness tester that loads a predetermined test force and forms an indentation in a surface of a sample using an indenter, and measures the hardness of the sample by measuring dimensions of the indentation includes a CCD camera that acquires an image of the surface of the sample before and after the indentation is formed. The CPU can execute a plurality of indentation region extraction processes that use mutually distinct methods, the indentation region extraction processes each extracting an indentation region based on the images acquired by the CCD camera. The CPU also makes a determination determining whether the indentation region extracted by the plurality of indentation region extraction processes matches a predefined reference indentation region, and based on an indentation region that is determined to match, the CPU calculates the hardness of the sample.