Motion Vector Frequency Analysis for Intraoperative Site Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in accurately detecting specific living body sites, such as blood vessels, within intraoperative images due to difficulties in distinguishing them from other structures like neural tubes.
Innovation Solution
An image processing device and method that utilize motion vectors detected across different times in intraoperative images to determine frequency patterns, allowing for precise identification of specific body regions by analyzing the frequency of motion vectors and generating frequency maps to differentiate between various body sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature amount (color information, edge information) is used to detect living body sites, then the detection process is simple, but the detection accuracy is insufficient and cannot discriminate between similar structures like blood vessels and neural tubes
Solution Approach 1:
The patent changes the detection parameter from static feature amounts (color, edge) to dynamic motion vectors. By analyzing the frequency characteristics of motion vectors obtained from sequential images, the system can distinguish living body sites based on their motion patterns, thereby improving detection accuracy without significantly increasing system complexity
Solution Approach 2:
The patent introduces dynamic analysis by capturing multiple images over time and computing motion vectors between them. This dynamic approach allows the system to detect living body sites based on their motion characteristics rather than static appearance, resolving the limitation of discriminating between similar-appearing structures
2Measurement precision
If motion vectors from multiple images are analyzed to improve detection accuracy, then detection precision improves, but processing time and computational load increase
Solution Approach 1:
The patent performs preliminary actions by capturing multiple images in advance and computing their motion vectors before final detection. By pre-processing the image sequence and extracting motion information upfront, the system prepares data that can be quickly analyzed for frequency characteristics, reducing the time required for actual detection
Solution Approach 2:
The patent applies partial action by focusing computation only on regions where motion vectors indicate potential living body sites. Rather than analyzing the entire image uniformly, the system concentrates processing resources on areas showing characteristic motion patterns, thereby reducing overall processing time while maintaining detection accuracy
Data Source
AI summary
The present disclosure relates to an image processing device, an image processing method, and a surgical system that enable detection of a region including a specific living body site in an intraoperative image on the basis of a frequency of a motion vector of a living body in the intraoperative image. A motion detection unit detects the motion vector of the living body in the intraoperative image using the intraoperative images at different times. The analysis unit obtains the frequency of the motion vector detected by the motion detection unit. A frequency map generation unit detects the region including the specific living body site in the intraoperative image on the basis of the frequency obtained by the analysis unit. The present disclosure is applicable to, for example, a CCU and the like of an endoscopic surgical system.