Optical Tomography Real-Time Video Stabilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Probe-type optical tomography apparatuses face challenges in producing clear observation images due to noise, uneven scanning, and relative displacement between the subject and the probe, affecting image visibility during real-time video capture.
Innovation Solution
The apparatus employs a measurement unit that divides light into measurement and reference light, using a processor to acquire and synthesize multiple tomographic images in real-time, correcting for misalignment and distortion to generate a stable video image by averaging and aligning sequentially captured images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple tomographic images are acquired and averaged to reduce noise, then image quality improves, but real-time video capability deteriorates due to processing time requirements
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing motion vectors between sequential frames before actual image synthesis. Motion compensation is prepared in advance based on detected displacement patterns, allowing rapid synthesis during real-time video generation without compromising either image quality or frame rate
Solution Approach 2:
The system dynamically adjusts the number of images used for averaging and the degree of motion compensation based on real-time conditions. When motion is detected between frames, the system adaptively modifies synthesis parameters to maintain both real-time performance and image quality, making the processing flexible rather than static
2Measurement precision
If motion compensation is applied to correct relative displacement, then image visibility improves, but processing complexity increases
Solution Approach 1:
The patent introduces motion vectors as an intermediary element that simplifies the motion compensation process. Instead of directly complex image alignment, the system first detects simple translational motion vectors between frames and uses these vectors to guide the synthesis process, reducing overall processing complexity while maintaining visibility
Solution Approach 2:
The patent replaces complex mechanical image alignment mechanisms with computational methods. Motion compensation is achieved through digital signal processing and algorithmic approaches rather than physical adjustment mechanisms, reducing mechanical complexity while improving precision
3Productivity
If scanning speed is increased to improve real-time performance, then frame rate improves, but uneven scanning and distortion increase
Solution Approach 1:
The patent implements feedback by continuously monitoring the actual scanning position and speed during acquisition. Detected deviations from uniform scanning are used to generate correction data that is applied during image synthesis, allowing high scanning speeds while maintaining scanning uniformity through real-time compensation
Solution Approach 2:
The system changes processing parameters dynamically based on detected scanning conditions. When uneven scanning is detected, the synthesis algorithm adjusts parameters such as interpolation factors and alignment corrections to compensate for the non-uniformity, maintaining image quality despite variable scanning speeds
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
This approach results in a satisfactory real-time observation image with reduced noise and improved visibility by correcting for camera shake and uneven scanning, enhancing the quality of the tomographic images captured.
Implementation Method 1
divides light output from a light source into measurement light and reference light, repeatedly scans a subject with the measurement light, and causes a detector to detect interference between reflected light of the measurement light and the reference light
Data Source
AI summary
An optical tomography apparatus has a measurement unit that divides light output from a light source into measurement light and reference light, that repeatedly scans a subject with the measurement light, and that causes a detector to detect interference between reflected light of the measurement light and the reference light, and a processor. The processor performs an acquisition process of are temporally sequentially acquiring tomographic images at the same transverse position in the subject based on a signal output from the detector. The processor performs a video image generation process of generating a synthetic image on a real time basis by synthesizing a new tomographic image frequently and one or more past images, and causing a monitor to display a real time video image formed from the synthetic image.