Adaptive Distance Metric Selection for Image Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image segmentation methods using clustering techniques face challenges in achieving accurate boundary separation between regions with similar and dissimilar color and texture characteristics due to the fixed distance calculation methods, which are not adaptive to the characteristics of the input data.
Innovation Solution
A data processing method that dynamically selects a distance calculation method based on the input data, integrating coordinate and color distances using various metrics such as Euclidean, Manhattan, and Chebyshev distances, to improve the accuracy of region segmentation by enhancing the difference between similar and dissimilar regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If Euclidean distance is used for color distance calculation, then the calculation is simple and fast, but the boundary accuracy between regions with different color characteristics becomes poor
Solution Approach 1:
The patent dynamically switches between Euclidean distance and Manhattan distance based on the characteristics of the input image data. When the image contains regions with significant color differences, Manhattan distance is selected to enhance boundary accuracy. When color variations are gradual, Euclidean distance is used for faster processing. This dynamic adaptation resolves the contradiction between calculation speed and boundary accuracy.
Solution Approach 2:
The patent changes the distance calculation parameter (metric type) based on the input data characteristics. By analyzing the color distribution and variance in the input image, the system selects the appropriate distance metric (Euclidean or Manhattan) to optimize both processing efficiency and segmentation accuracy for different image types.
2Productivity
If fixed weight coefficients are used in distance calculation, then the processing is simple and fast, but the boundary accuracy becomes poor for certain processing target images
Solution Approach 1:
The patent employs dynamic weight coefficient adjustment based on the statistical characteristics of the input image. The system calculates color variance and distribution patterns to adaptively determine optimal weights for different color components and distance metrics. This dynamic weighting allows the system to achieve high boundary accuracy for diverse image types while maintaining processing efficiency through automated parameter selection.
Data Source
AI summary
In a data processing method of determining a class to which each of a plurality of input data items belongs, a distance between input data and each of a plurality of representative data items is calculated, a distance calculation method is selected based on the input data, and the input data is allocated to a class to which representative data, from which a shortest distance has been obtained out of the calculated distances using the selected distance calculation method, belongs.