Boundary Surface Estimation Using Brightness Gradient and Cloth Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual removal of superficial blood vessels from medical images is time-consuming, and existing methods struggle to accurately determine a smooth boundary surface, especially when image information on the boundary surface is limited.
Innovation Solution
An image processing apparatus and method that utilize brightness gradient distribution to estimate the boundary surface of an object, employing cloth simulation to model the surface and suppress the influence of varying brightness values, allowing for continuous surface estimation even with limited image information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual operation is used to remove superficial blood vessels from PAT images, then the image interpretation quality is improved, but the processing time becomes excessively long
Solution Approach 1:
The patent replaces manual mechanical operations with an automated image processing system that uses differential value calculation and clustering algorithms to automatically remove superficial blood vessels from PAT images, thereby eliminating the time-consuming manual process while maintaining interpretation quality
Solution Approach 2:
The system enables self-service by allowing the image processing apparatus to automatically perform vessel removal without requiring continuous human intervention, using automated algorithms to identify and process superficial vessels based on brightness characteristics
2Speed
If PTL 1 method is used to detect body surface region by calculating differential values on line units, then the processing speed is improved, but the boundary surface smoothness deteriorates
Solution Approach 1:
The patent merges multiple line-based differential value calculations into a comprehensive three-dimensional clustering process that integrates information from all directions, transforming discrete line results into a continuous smooth boundary surface representation
Solution Approach 2:
The patent transitions from one-dimensional line-based analysis to three-dimensional volumetric clustering, adding spatial dimensions to the processing to achieve both speed and smoothness by considering the entire volume rather than individual lines
3Adaptability or versatility
If image data from imaging methods with limited boundary surface information is used, then the imaging versatility is improved, but the boundary surface estimation accuracy deteriorates
Solution Approach 1:
The patent introduces brightness distribution characteristics as an intermediary that bridges the gap between limited direct boundary information and accurate surface estimation, using the brightness patterns within the volume data to infer boundary surface properties
Solution Approach 2:
The patent changes the approach by not relying on direct boundary pixel information but instead using internal brightness distribution parameters and differential value patterns to indirectly determine boundary surface characteristics, thereby achieving accurate estimation from limited data
Data Source
Figure 1
Figure 2
Figure 3(A)~3(B)
AI summary
The present invention allows to estimate a continuous surface that defines a boundary surface of an object using image data, even if the image data was acquired by an imaging method in which image information on the boundary surface of the object is limited. An image processing apparatus according to the present invention includes an estimating unit that, using the brightness gradient of a voxel group constituting volume data which represents a reconstructed image, estimates a continuous surface which defines a boundary surface of an object corresponding to the volume data.