文档介绍：基于图像处理的森林火灾检测系统的技术研究
摘要
图像预处理的主要方法有:灰度化、二值化、均值滤波、中值滤波、高斯滤波。
本研究还将各向异性扩散方法应用于医学图像的去噪处理。采用基于各向异性扩散的偏微分方程,其初始值为输入图像,转化为差分格式迭代求解滤波结果。以在去除噪声的同时,保持重要的边缘、局部细节。在此基础上提出八向各向异性扩散和边缘增强各向异性扩散,取得了满意的结果。
K-均值聚类是一个快速的图像分割算法。大量的实验表明K-均值聚类是一个快速收敛的算法。
经典的C-均值聚类算法是将图像分割成C类的常用方法。它是一种非监督模糊聚类后再标定的过程,适合灰度图像中存在着模糊和不确定性的特点。但是这种算法存在不足,如聚类数无法自动确定,运算的开销太大。
医学器官虚拟三维重构技术是现代外科计算机化,信息化发展的重要方向,是计算机科学、数学、机械学、外科及生物医学工程学等多学科高技术的结晶。本文把各种图像预处理运用到三维医学重构上,使重构效果更好。
关键词:森林火灾检测系统;图像处理;图像分割;
Color Image Segmentation and an’s 3D-Reconstruction
Abstract
In this paper,firstly introduce image pre-processing,for example,gray-value,two-value, gaussian filters, anisotropic diffusion.
To remove the speckle noise in medical images by using anisotropic diffusion on anisotropic diffusion,a partial differential equation,of which the initial data was the input images,was transformed into differential forms and solved with diffusion can remove the speckle noise effedtively and has great potential in filtering medical images. It can preserve edges and local detail.
K-means is a fast segmentation process. Extensive experiments show it is fast and generic.
The classical C-means clustering algorithm is a well-knowm clustering method to partition an image into homogeneous is a procedure of the label following an un supervised fuzzy clustering that fuzzy c-means(FCM) algorithm is applied for intensity image segmentation,and it suits for the uncertain and ambiguous characteristic in intensity ,there are some deficiencies in the algorithm,for example,the number of clustering can not be determined automatically and the operational const,for large data sets ,is too high;
an’s dummy 3D-reconstruction is an important direction of the development in modern surgery’puterization and informationize field. It’s bination of the latest techniques puter science , mathematics , mechanics , surgery and biological engineering. In this paper, image pre-processing is appled