Urine Sediment Image Processing via Color Histogram Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current urine sediment image processing methods face challenges in precision due to reliance on shape and space features, which are influenced by variations and location changes of visible elements, leading to low processing accuracy and inefficiency.
Innovation Solution
A method and apparatus that approximate pixel colors in a urine sediment image to a code book of clustered colors from urine sample blocks, using a distribution histogram with an occurrence frequency correction factor and standardization, to generate features for classification or retrieval, independent of shape and space information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If shape and space features are used for processing urine sediment images, then the processing can be performed using conventional methods, but the precision is low due to variations and location changes of visible elements
Solution Approach 1:
The patent changes the feature parameters from shape and space characteristics to color distribution characteristics. By extracting color features and constructing histograms based on color frequency distribution, the method achieves invariance to shape and position variations while maintaining high processing accuracy for urine sediment image analysis
Solution Approach 2:
The patent replaces the conventional shape-based processing approach with a color-based statistical approach. Instead of relying on geometric features that are sensitive to transformations, the system uses color histogram analysis which is inherently more robust to variations in object position, orientation, and shape
2Measurement precision
If conventional color processing is used without correction, then the processing is simple, but frequent colors dominate the features reducing processing precision
Solution Approach 1:
The patent introduces an intermediary correction mechanism using occurrence frequency correction factors. These factors act as mediators that adjust the raw color histogram data to compensate for the dominance of frequent colors, thereby improving feature representation without requiring complete redesign of the processing system
Solution Approach 2:
The patent performs preliminary correction of the color histogram by applying occurrence frequency correction factors before the main classification or retrieval operations. This preliminary action pre-adjusts the feature data to reduce the dominance effect of frequent colors, improving subsequent processing accuracy
Data Source
AI summary
Concepts herein relate to processing urine sediment images. An example method comprises: approximating the color of a pixel in a block to be processed to one of the kc colors in a code book, wherein the code book is a set of the kc colors generated in a set of urine sample blocks; obtaining a distribution histogram of the number of pixels the color approximation results of which fall on each color of the kc colors; using an occurrence frequency correction factor to correct the number of pixels the color approximation results; standardizing the corrected number of pixels the color approximation results; and processing the block to be processed.