资源简介

摘要： 利用高空间分辨率遥感影像准确提取建筑物信息是在更精细尺度上分析、管理人类活动的重要途径,但建筑物形态的多样性、复杂性使得建筑物提取方法在精度、效率等方面依然面临挑战。基于高分辨率影像中建筑物所呈现的多种形状及结构特点,通过发展形态学建筑物指数并结合具有良好保边去噪性能的引导滤波,提出了基于多结构元素形态学建筑物指数(MMBI)的建筑物提取方法。将之应用于具有不同特征的GeoEye-1影像建筑物提取实验,并与MBI方法进行比较,发现该方法准确率达到88.2%以上,比MBI的准确率提高了约5%,且漏检率下降近6%。结果表明:MMBI方法能够快速、准确地提取出高分辨率影像中的建筑物区域,在建筑物的自动化提取方面具有较好的应用前景。 关键词： 建筑物;  形态学;  高分辨率影像;  MMBI     Abstract: High spatial resolution remote sensing images provide an important and effective way to study and analyze human activities on finer scales.However,the diversity and complexity of buildings make it a challenging task to efficiently and accurately extract building information from high resolution images.This paper proposed a method named Multi-structural element of Morphological Building Index(MMBI) to extract building information.Based on variety shapes and structural characteristics of buildings presented in the high-resolution images,the multi\|structural morphological building index is developed,by combining with guided filter which performs well in denoising and preserving edges.The proposed method was compared with the MBI approach through analyzing their performances in extracting buildings with different characteristics from a GeoEye-1 image.Results show that the MMBI has an accuracy rate above 88.2%,which is 5% higher than that of the MBI.Meanwhile,the omission rate of the MMBI method is 6% lower than the MBI.The results proved that the MMBI method is able to quickly and accurately extracting building information from high spatial resolution remote sensing images and thus has great potential of application in automatic building extraction.

相关资源

1. 水质遥感监测的关键要素叶绿素a的反演算法研究进展
2. 基于无人机遥感技术的森林参数获取研究进展
3. 基于无线传感器网络的森林生态系统观测试验平台构建
4. 基于无人机激光雷达的森林冠层高度分析
5. 多源时序国产卫星影像的森林火灾动态监测
6. 基于无人机遥感技术的台风灾害倒伏绿化树木检测

下载说明

·如果您发现该资源不能下载，请通知管理员.gissky@gmail.com

·为确保下载的资源能正常使用，请使用[WinRAR v3.8]或以上版本解压本站资源,缺省解压密码www.gissky.net ，如果是压缩文件为分卷多文件，请依次下载每一个文件，并按照顺序命名为1.rar,2.rar,3.rar...，然后鼠标右击1.rar解压.

·为了保证您快速的下载速度，我们推荐您使用[网际快车]等专业工具下载.

·站内提供的资源纯属学习交流之用,如侵犯您的版权请与我们联系.