基于处理—分裂—合并的彩色图像自动分割系统  

Information Processing and Automatic Segmentation System for Color Images

作　　者：朱学芳[1] 曾娜[1] 薛燕[1] 

机构地区：[1]南京大学信息管理系多媒体信息处理研究所,南京210093

出　　处：《自动化与仪表》2007年第3期12-16,共5页Automation & Instrumentation

摘　　要：叙述一个基于处理—分裂—合并的彩色图像自动分割处理系统。其主要功能为:根据图像的区域特征将彩色图像自动分割成若干块并存储区域的边界特征,实现了地图的面查询,并可实现原图和分割出来的区域图的放大和缩小显示。采用适当的去噪、平滑方法使区域图像分割更精确。本系统采用面向对象的分析设计方法,在Windows平台上,进行编程开发实现。系统对GIS、图像数据库及其检索和图像及文字模式识别的研究有较重要的意义。A color image information processing and segmentation system is primarily described in the paper. Based on the color homogeneity,images can be divided into several regions which have different attributions,with the border information of the regions stored in a database. After segmentation,the map can be queried . Also,the function of zooming out and zooming in the map are realized. Methods of OOP （object-oriented programming）on the Windows platform are adopted in all the system modules and the OWL （object windows library）is used to implement the windows programming. The system is helpful to the study of geographic information system（GIS）,image database and retrieval ,recognition of image including characters, and so on.

关 键 词：彩色图像 分割 噪声去除 图像平滑 分裂 合并 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]