Anisotropic Filter for X-ray Talbot Interferometry Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing techniques struggle to effectively apply anisotropic filtering to images with low signal-to-noise ratios, particularly in X-ray Talbot interferometry, where reducing exposure dose leads to insufficient contrast and difficulty in detecting edge details in soft tissues like cartilage.
Innovation Solution
An image processing apparatus and method that acquires measurement data from electromagnetic waves transmitted through an object, generates images representing different physical quantities, and applies an anisotropic filter with varying filter characteristics determined based on an absorption image to enhance edge detection and noise reduction in images with low SN ratios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If exposure dose is reduced to minimize radiation damage, then safety is improved, but signal-to-noise ratio deteriorates leading to insufficient contrast and difficulty in detecting edge details
Solution Approach 1:
The patent segments the image processing into multiple stages: acquiring multiple images with different exposure conditions, separately processing high-exposure images for edge detection and low-exposure images for detail preservation, then combining results. This segmentation allows optimizing each stage for its specific purpose while avoiding the need to use high exposure throughout.
Solution Approach 2:
The patent changes exposure parameters across multiple images - acquiring some images with higher exposure doses and others with lower exposure doses. By varying this parameter and processing different images differently, the system achieves both adequate signal-to-noise ratio for edge detection and low radiation damage overall.
2Reliability
If conventional filtering is applied to low SN ratio images, then noise reduction is achieved, but edge details are lost due to insufficient contrast
Solution Approach 1:
The patent applies different processing qualities to different parts of the image based on local characteristics. Edge regions are processed with filtering optimized for edge preservation, while other regions receive different treatment. This local quality approach ensures edge details are maintained while noise is reduced in appropriate areas.
Solution Approach 2:
The patent performs preliminary edge detection on high-exposure images before applying filtering to low-exposure images. By identifying edge locations in advance from the higher quality images, the system can then apply noise reduction to low-exposure images without compromising edge detection accuracy, as edges are already known from the preliminary action.
3Reliability
If multiple images with different exposure conditions are acquired and processed separately, then image quality is improved, but processing complexity increases
Solution Approach 1:
The patent merges the results from processing multiple images with different exposure conditions into a single final image. By combining the edge information from high-exposure images with the detail information from low-exposure images through a unified processing framework, the system achieves improved image quality while managing complexity through integration rather than separate independent processes.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The approach effectively enhances edge visibility and noise reduction in images with low SN ratios, allowing for more accurate measurement of cartilage thickness and improved diagnostic capabilities in medical imaging.
Implementation Method 1
X-ray absorption imaging in which contrast resulting from absorption by an object is imaged
Implementation Method 2
detecting, based on phase interference, a change in the length of an optical path of incident light formed during transmission of X rays through the object
Implementation Method 3
Light transmitted through the object is diffracted by a grating referred to as a diffraction grating and having a periodic pattern
Implementation Method 4
The shield grating blocks a portion of the first interference pattern to form a second interference pattern with a period of approximately several hundred micrometers, that is, a moiré pattern
Data Source
AI summary
An image processing apparatus includes: a measurement data acquiring unit that acquires data obtained by using an imaging apparatus to capture an image formed by an electromagnetic wave transmitted through an object, as measurement data of the object; an image generating unit that generates a first image and a second image that represent information on different physical quantities, from the measurement data; a filter characteristics determining unit that determines, based on the first image, filter characteristics to be set when an anisotropic filter is applied, for each position in the image; and a filtering unit that applies the anisotropic filter to the second image while varying the filter characteristics of the anisotropic filter for each position in the image in accordance with the filter characteristics determined based on the first image.