Autonomous Flow Network Image Segmentation via Contrast Enhancement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately analyze and quantify the structure of objects with apparent random patterns or pathways, such as blood vessels or river systems, which are difficult to predict and compare, especially for predictive purposes and network connectivity improvement.
Innovation Solution
A method involving image processing techniques like low pass filtering, wavelet transformation, and neural network analysis to enhance contrast and identify features in images of flow networks, allowing for the approximation and reassembly of line-like structures, thereby improving the analysis of random patterns and pathways.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image processing techniques are applied to enhance contrast and identify features in flow networks, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the image processing into distinct sequential steps: low pass filtering to reduce noise, wavelet transformation to enhance contrast, local mean calculation to normalize intensity, and line approximation to identify structures. Each step processes a specific aspect of the image independently, making the complex overall process manageable and systematic.
Solution Approach 2:
The patent applies preliminary processing actions before main analysis by first reducing noise through low pass filtering and then enhancing contrast through wavelet transformation. These preliminary actions prepare the image data in advance, making subsequent structure identification more accurate and reliable.
2Measurement precision
If low pass filtering and wavelet transformation are applied to minimize visible light intensity differences, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent applies wavelet transformation which can be computationally intensive, but only to the extent necessary to achieve contrast enhancement. The local mean calculation is performed only at pixels where intensity variation exceeds a threshold, avoiding unnecessary computations throughout the entire image.
3Manufacturing precision
If line approximation testing is performed on network sections, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent changes the parameter space from raw image intensity values to transformed wavelet coefficients and normalized intensity ratios. This parameter transformation simplifies the line approximation testing by converting complex visual patterns into mathematical relationships that are easier to evaluate systematically.
Data Source
AI summary
Disclosed herein is a method that comprises obtaining an image of a network section through which flow occurs; where the flow is selected from a group consisting of fluid, electrons, protons, neutrons and holes; subjecting the image to a low pass filter to increase contrast in portions of the network sections; computing a local mean of visible light intensity at each pixel that is present in the image; calculating a visible light intensity difference between each pixel and the local mean of visible light intensity and producing a differentiated image using this calculation; creating a base image of the differentiated image; where the base image comprises a hand segmented gold standard dataset; removing objects below a minimum threshold size from the base image; and retaining the remaining objects if they approximate the line or spine.