Image Segmentation via Residual Polarity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image segmentation methods fail to effectively distinguish between object and background pixels, especially under uneven illumination conditions, leading to noise and incomplete object range determination.
Innovation Solution
An image segmentation method that involves obtaining a first gray-scale image, performing regression analysis to generate a residual image, determining an object backbone area, calculating average gray-scale values, and recursively expanding the object area based on pixel polarity and threshold comparisons to extract the target object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image segmentation is based on color space or gray level space distribution, then the segmentation process is simple, but it cannot distinguish whether pixels belong to background or object, leading to large amounts of noise or incomplete object range
Solution Approach 1:
The patent segments the image processing into distinct stages: background modeling, residual image generation, object backbone area determination, and recursive expansion. Each stage processes specific features independently, improving classification accuracy while maintaining manageable complexity through modular processing steps.
Solution Approach 2:
The patent transitions from traditional 2D spatial processing to incorporating residual polarity information as an additional dimension. By analyzing residual polarity alongside pixel intensity values during recursive expansion, the method creates a multi-dimensional classification space that improves object-background distinction without excessive complexity.
2Reliability
If conventional image segmentation does not consider uneven illumination or reflection brightness changes, then the processing is faster, but the determination information cannot distinguish object from background effectively
Solution Approach 1:
The patent performs background modeling and residual image generation before object extraction. By pre-processing the image to remove background and illumination effects beforehand, the subsequent object segmentation operates on simplified data, maintaining high reliability while improving processing speed through staged computation.
Solution Approach 2:
The patent transforms the image representation by computing residual images and applying recursive expansion with polarity-based thresholds. These parameter transformations convert raw pixel values into reliability-weighted residual values, improving segmentation reliability while the efficient recursive algorithm maintains acceptable processing speed.
3Manufacturing precision
If recursive expansion is performed based on residual polarity and pixel value thresholds, then the object boundary accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent performs recursive expansion only in regions where residual polarity indicates potential object boundaries, rather than processing the entire image uniformly. This partial action approach achieves high boundary precision at object edges while minimizing computational complexity in background regions.
Solution Approach 2:
The patent applies different processing strategies to different image regions: recursive expansion with polarity checking is applied locally at object boundaries identified through residual analysis, while uniform regions undergo simpler processing. This local quality approach optimizes boundary precision where needed without excessive overall complexity.
Data Source
AI summary
The invention provides an image segmentation method and an electronic device. The image segmentation method includes the following steps. Regression analysis is performed on a first gray-scale image to obtain a residual image having an object backbone area. A pixel value of each pixel in the object backbone area is defined as an average gray-scale value of the object backbone area in the residual image, and a second gray-scale image having the object backbone area is generated. It is recursively determined whether a residual polarity of each adjacent pixel adjacent to edge pixels of the object backbone area in the residual image is the same as a residual polarity of the corresponding edge pixel, and whether a pixel value of each adjacent pixel is greater than a first threshold, so as to expand the object backbone area in the second gray-scale image, which is extracted as a target object.