Radiation Image Frequency Band Segmentation for Tissue Differentiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical image systems using radiation, such as X-rays, face challenges in distinguishing between different tissues within soft tissue due to similar attenuation coefficients, leading to difficulties in differentiating multiple tissues in radiation images.
Innovation Solution
An image processing apparatus that extracts low-frequency and high-frequency band components from radiation images, applies contrast enhancement and detail enhancement processes, and synthesizes these components to generate a diagnosis image, using techniques like wavelet transform and contrast stretching based on brightness levels and edge intensities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single energy band X-ray is used for imaging, then the image acquisition process is simple and fast, but the ability to distinguish different soft tissues is poor due to similar attenuation coefficients
Solution Approach 1:
The patent segments the image processing into distinct frequency bands (low-frequency and high-frequency components) and applies different enhancement strategies to each. Low-frequency components undergo contrast enhancement to improve tissue differentiation, while high-frequency components receive detail enhancement to preserve edge information. This segmentation allows targeted processing that resolves the contradiction between differentiation ability and processing complexity.
Solution Approach 2:
The patent applies different quality enhancement methods to different frequency components of the image. Low-frequency regions receive contrast stretching to improve tissue differentiation, while high-frequency regions receive detail enhancement to preserve structural information. This local quality approach enables simultaneous optimization of both tissue differentiation and image detail without requiring complex multi-energy imaging systems.
2Measurement precision
If contrast enhancement is applied to the entire image, then tissue differentiation is improved, but image details and edges may be lost
Solution Approach 1:
The patent divides the image into low-frequency and high-frequency components through frequency decomposition. Contrast enhancement is selectively applied only to the low-frequency component, while the high-frequency component retains its original detail information. This segmentation prevents the loss of image details that would occur if contrast enhancement were applied to the entire image.
Solution Approach 2:
After separate processing of low-frequency and high-frequency components, the patent merges them back together through synthesis to produce the final enhanced image. This combining process integrates the improved tissue contrast from the low-frequency component with the preserved detail information from the high-frequency component, achieving both goals simultaneously.
3Measurement precision
If frequency decomposition is performed to separate low-frequency and high-frequency components, then selective enhancement becomes possible, but the processing time and computational load increase
Solution Approach 1:
The patent uses frequency decomposition to segment the image into low-frequency and high-frequency components, enabling selective enhancement. Despite the additional processing step, the patent optimizes the enhancement algorithms to operate efficiently on each component separately, minimizing the overall processing time while maintaining the benefits of selective enhancement.
Data Source
AI summary
An image processing apparatus is provided. The image processing apparatus includes an extraction unit configured to extract a low-frequency band component and a high-frequency band component from a radiation image formed by passing through a subject, a contrast enhancement processing unit configured to perform a contrast enhancement process to the extracted low-frequency band component, a detail enhancement processing unit configured to perform a detail enhancement process to the extracted high-frequency band component, a synthesis unit configured to synthesize an image resulting from performing the contrast enhancement process and an image resulting from performing the detail enhancement process, and a generation unit configured to reconstruct the synthesized image and generate a diagnosis image.