Grating Interferometer CNN Enhancement for Contrast and Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing grating interferometers face challenges in simultaneously achieving high contrast sensitivity and spatial resolution due to independent factors influencing these qualities, making it difficult to acquire images with both high contrast and high resolution simultaneously.
Innovation Solution
A system utilizing a convolutional neural network (CNN) for machine learning is employed to enhance contrast sensitivity and resolution in grating interferometers by acquiring high-resolution and high-sensitivity images, performing image size rearrangement, generating numerical phantoms, and applying ReLu activation functions to correct convolution calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a grating interferometer is configured to achieve high contrast sensitivity, then the contrast sensitivity is improved, but the spatial resolution deteriorates
Solution Approach 1:
The patent segments the image processing task into multiple convolutional layers, where each layer extracts different features from the input image. This segmentation allows the system to separately optimize for contrast sensitivity and spatial resolution through dedicated processing stages, ultimately combining these features to produce an output image that achieves both high contrast sensitivity and high spatial resolution simultaneously.
Solution Approach 2:
The patent transforms the problem from the spatial domain to the frequency domain through Fourier transformation, and then applies machine learning in this transformed dimension. By processing images in the frequency domain and then transforming back, the system can enhance both contrast sensitivity and spatial resolution that are difficult to achieve simultaneously in the original spatial domain.
2Measurement precision
If machine learning processing is applied to enhance both contrast sensitivity and resolution, then the image quality is improved, but the processing complexity increases
Solution Approach 1:
The patent performs preliminary actions by dividing the complex machine learning processing into multiple sequential convolutional layers, where each layer performs a specific feature extraction task. This preliminary segmentation of processing tasks reduces the overall complexity by making each individual layer simpler and more manageable, while collectively achieving high image quality enhancement.
Solution Approach 2:
The patent replaces traditional mechanical or optical methods for enhancing image quality with machine learning-based computational methods. By using convolutional neural networks and Fourier transformation algorithms, the system achieves superior image quality enhancement without requiring complex physical modifications to the grating interferometer hardware.
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 system improves both contrast sensitivity and spatial resolution in grating interferometers, enabling the extraction of high-quality medical images with enhanced sensitivity and resolution, applicable beyond X-ray and neutron imaging fields.
Implementation Method 1
phase-contrast imaging using a grating interferometer in which an interference principle is applied to radiation
Data Source
AI summary
The present disclosure relates to an apparatus for enhancing contrast sensitivity and resolution in a grating interferometer by machine learning, which can improve both image contrast sensitivity and spatial resolution in a grating interferometer by machine learning, the apparatus including: a grating interferometer image acquisition unit that acquires a relatively high resolution image and a relatively high sensitivity image by linearly moving the position of a sample from the symmetrical grating interferometer; a numerical phantom generation unit that generates a numerical phantom for performing machine learning; a convolution layer generation unit that performs calculation processing of a convolutional neural network to extract features from input data; an activation function application calculation unit that can apply a ReLu (Rectified linear unit) activation function to an output value of the convolution calculation to perform smooth repetitive machine learning; a CNN repetitive machine learning unit that corrects a convolution calculation factor while repeatedly performing forward propagation and backward propagation processes; and an image matching output unit that matches and outputs features extracted by repetitive machine learning of the CNN repetitive machine learning unit.