基于光流算法的医学柔式内窥镜系统  

Optical Flow Algorithm Based on the Medical Flexible Endoscope System

作　　者：刘桥[1] 李晗[1] 何凌杰 

机构地区：[1]长沙理工大学电气与信息工程学院,湖南长沙410114

出　　处：《电子科技》2015年第4期61-64,共4页Electronic Science and Technology

摘　　要：手动操纵的柔式内窥镜在介入内腔管道时,要求医生必须有熟练的操作技巧和技术。针对手动操纵柔式内窥镜所带来的弊端,设计了一种基于LK光流算法的机器人柔式内窥镜系统。利用内窥镜图像控制内窥镜指向内腔管道的中心线,从而可提高内窥镜的智能化、简化对操作医师技术的依赖、减少医疗事故的发生。以常规胃镜检查为例,对常用的图像灰度算法进行了分析比较,实验结果表明,基于LK光流算法得到的内腔管道中心路径更为精确。Manually steering the tip of a flexible endoscope to navigate through an endoluminal path relies on the physician＇s dexterity and experience. According to the drawbacks of manually steering the flexible endoscope, we propose a robot flexible endoscope system based on the LK optical flow algorithm. Using the endoscopic images to control the flexible endoscope directed towards the center path of the lumen can improve endoscopic intelligence, simplify the technique required of the operating physician, and reduce the incidence of medical malpractice. Taking the example of conventional gastroscopy, the commonly used image intensity algorithms are analyzed and compared. Experimental results show that the LK optical flow algorithm based on the center path of the lumen is more precise.

关 键 词：柔式内窥镜图像 光流算法 图像灰度值算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]