X-ray Image Noise Reduction via Dynamic Filter Application
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional smoothing filters fail to effectively reduce noise in X-ray images of moving objects, such as a heart, leading to motion blur and inadequate noise reduction in intravascular interventional treatments.
Innovation Solution
An X-ray diagnosis apparatus and image processing apparatus that detect feature points in sequential X-ray images, determine the application range of a smoothing filter based on the motion of these points, and apply the filter accordingly to ensure consistent noise reduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a smoothing filter is applied to reduce noise in X-ray images of moving objects, then noise reduction effect is improved, but motion blur occurs due to object movement between frames
Solution Approach 1:
The patent applies dynamics by making the filtering process adaptive to motion. The system dynamically adjusts the filtering strength based on detected motion magnitude between frames. When motion is detected, the filtering is weakened or suspended for affected regions, preventing motion blur. When motion is minimal, stronger filtering is applied to maximize noise reduction. This dynamic adjustment resolves the contradiction between noise reduction and motion blur prevention.
Solution Approach 2:
The patent applies local quality by differentiating filtering application across different regions of the image. Instead of uniform filtering, the system identifies regions with motion (using feature point detection and motion vector calculation) and applies different filtering strengths to different regions. Static regions receive strong noise reduction filtering, while moving regions receive reduced or no filtering to avoid motion blur. This localized approach resolves the contradiction by allowing both noise reduction and motion preservation in their respective appropriate regions.
2Reliability
If a recursive filter is applied to X-ray images radiographed at minimum dose, then noise reduction is improved, but the filter cannot be applied strongly to moving objects
Solution Approach 1:
The system dynamically adapts the filtering application based on motion detection. The recursive filter is applied with variable strength depending on the detected motion magnitude in each region and frame. This dynamic adaptability allows the filter to work effectively on both stationary and moving objects, resolving the contradiction between noise reduction capability and applicability to moving objects.
Solution Approach 2:
The patent changes the filtering parameters (strength, application range) based on motion detection results. By adjusting these parameters dynamically according to the motion state of objects in the image, the system maintains effective noise reduction for stationary regions while preventing motion blur in moving regions, thus making the filter adaptable to both static and dynamic scenarios.
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 solution ensures a consistent noise reduction effect by dynamically adjusting the filter application range, preventing motion blur and enhancing image clarity in real-time X-ray imaging of moving objects.
Implementation Method 1
an image data generating unit that generates X-ray images in a time sequence by detecting an X-ray that is radiated from an X-ray tube and is transmitted through an subject
Data Source
AI summary
When a plurality of X-ray images in a time sequence is stored in an image data storing unit (25), a marker coordinate detecting unit (26a) detects coordinates of a stent marker in each X-ray image, and a motion vector calculating unit (26b) calculates, with coordinates of the stent marker detected in a first frame as reference coordinates, a motion vector of the coordinates of the stent marker detected in each X-ray image of a second and subsequent frames with respect to the reference coordinates. Then, a filter application range determining unit (26c) moves and determines an application range of a smoothing filter in each X-ray image based on the motion vector, and the filtered image generating unit (26d) generates a filtered image by performing a process by the smoothing filter between application ranges determined in a process target image and a reference image.