论文部分内容阅读
胶囊内窥镜技术是胃肠道内窥镜检查的一个重大突破,但其临床反馈尚有一些问题待解决,其中一个问题就是定位跟踪。在各种可能的定位技术中,磁体定位技术具有不用供电且体积小等特点,应用于胶囊内窥镜的定位问题是非常合适的。文章针对磁体定位技术中的磁偶极子定位模型,提出一种新的算法——混合人工蜂群算法,并将该算法与标准人工蜂群算法、随机复合形算法、Levenberg-Marquart(LM)算法进行对比。实验结果表明,混合人工蜂群算法在算法定位精度、算法稳定性和抗噪能力等方面,比其他三种算法具有更优越的性能。
Capsule endoscopy is a major breakthrough in gastrointestinal endoscopy, but its clinical feedback, there are still some problems to be solved, one of the problems is positioning and tracking. Among the various possible positioning technologies, the magnet positioning technology has characteristics of no power supply and small volume, and the positioning problem applied to the capsule endoscope is very suitable. Aiming at the magnetic dipole localization model in magnet positioning technology, a new hybrid algorithm named hybrid artificial bee colony algorithm is proposed. The algorithm is compared with standard artificial bee colony algorithm, stochastic complex algorithm, Levenberg-Marquart (LM) Algorithm for comparison. The experimental results show that the hybrid artificial bee colony algorithm has superior performance over the other three algorithms in terms of algorithm location accuracy, algorithm stability and noise immunity.