【摘 要】
：
结合PSO算法和加权FCM算法(WFCM)的优势,提出一种新的图像分割算法——基于PSO和加权FCM的图像分割算法(PWF).算法通过PSO进行全局寻优,克服了单纯FCM算法的对初始值敏感及对噪声数据敏感,容易陷入局部最优的缺点,能够找到全局最优的模糊划分,实现模糊聚类图像分割.实验利用了加权二维直方图的WFCM算法,考虑像素间的空间信息,改善了图像分割效果.
论文部分内容阅读
结合PSO算法和加权FCM算法(WFCM)的优势,提出一种新的图像分割算法——基于PSO和加权FCM的图像分割算法(PWF).算法通过PSO进行全局寻优,克服了单纯FCM算法的对初始值敏感及对噪声数据敏感,容易陷入局部最优的缺点,能够找到全局最优的模糊划分,实现模糊聚类图像分割.实验利用了加权二维直方图的WFCM算法,考虑像素间的空间信息,改善了图像分割效果.
其他文献
传统的人控交会对接主要通过观察摄像机产生的十字靶标图像来完成,容易造成人的视觉误差,从而导致手动对接失败。结合图形学理论和虚拟现实技术构建的人控交会对接仿真系统为航天员实时提供逼真的虚拟立体场景,通过各种任务想定对航天员进行飞行任务训练。在实验过程中,航天员通过立体眼镜观察虚拟场景中飞行器的飞行状态,同时操作控制手柄改变飞行器的位置和姿态进行手动对接训练。通过不同的对接实验参数设置,航天员可以反复
An immune kernel clustering network (IKCN) is proposed based on the combination of the artificial immune network and the sup- port vector domain description (SVDD) for the unsupervised image segmentat
本文根据符号学和用户界面概念,阐述了图形用户界面中各元素的符号属性,从符号学的角度探讨了图形用户界面中图符的来源,并将符号的四个特征映射到图形用户界面中,以期为图形用户界面的设计提供另外一个展开的视角。
In order to meet the demands of high efficient and real-time computer assisted diagnosis as well as screening in medical area, to improve the efficacy of parallel medical image processing is of great
An improved image registration method is proposed based on mutual information with hybrid optimizer. Firstly, mutual information measure is combined with morphological gradient information. The essenc
The Rician noise introduced into the diffusion tensor images (DTIs) can bring serious impacts on tensor calculation and fiber tracking. To decrease the effects of the Rician noise, we propose to consi
随着我国国民经济的迅猛发展,仓储物流管理已成为国民经济发展的一个重要组成部分。传统的仓储物流管理方式已经远远不适应当今社会快速发展的需要。RFID电子标签在仓储物流管理中的应用,彻底解决了仓储物流管理中靠手工、台帐管理的难题,打破了仓储物流管理的瓶颈,推动了仓储物流工作的快速发展。
As the fastest integral equation solver to date, the multilevel fast multipole algorithm (MLFMA) has been applied successfully to solve electromagnetic scattering and radiation from 3D electrically la
The definition of rough similarity degree is given based on the axiomatic similarity degree, and the properties of rough similarity degree are listed. Using the properties of rough similarity degree,
A new robust electronic image stabilization system is presented, which involves feature-point, tracking based global motion estimation and Kalman filtering based motion compensation. First, global mot