图像处理系统、内窥镜系统、图像处理方法以及存储介质
By generating region-of-interest detectors for both ordinary and special light, and using an observation method classifier to select appropriate detectors, the problem of decreased detection accuracy caused by changes in observation methods in endoscopic diagnosis is solved, achieving high-precision lesion detection.
CN114901119BActive Publication Date: 2026-07-17OLYMPUS CORPORATION(JP)
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- OLYMPUS CORPORATION(JP)
- Filing Date
- 2020-01-09
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
In endoscopic diagnosis, existing technologies have failed to effectively maintain the accuracy of lesion detection when the observation method changes, especially when switching from ordinary light observation to special light observation, the detection accuracy decreases.
Method used
By generating region-of-interest detectors based on ordinary and special light, and using an observation method classifier to estimate the observation method of the image, a suitable detector is selected for detection processing.
Benefits of technology
Even with changes in observation methods, it can still maintain high precision in lesion detection, thus improving the accuracy of detection and treatment.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN114901119B_ABST
Abstract
图像处理系统(200)包括：图像取得部(210)，其取得处理对象图像；以及处理部(220)，其进行输出在处理对象图像中检测出关注区域的结果即检测结果的处理。在处理部(220)中，基于观察方法分类器，进行将拍摄处理对象图像时的观察方法分类为第一观察方法或第二观察方法的分类处理，基于观察方法分类器的分类结果，进行选择第一关注区域检测器或第二关注区域检测器的选择处理，并基于所选择的关注区域检测器输出检测结果。
Need to check novelty before this filing date? Find Prior Art