摘要
为确保土地利用数据的真实性,设计基于多源遥感数据的第三次国土调查内业信息提取方法。以高分二号卫星采集的全色与多光谱影像为基础,采用S-G(Savitzky-Golay)滤波器通过噪声消除、配准、重采样实现全色与多光谱影像融合,利用像元亮度校正改善融合后影像像元亮度饱和现象。根据最小代价切割实现多源遥感影像不同地类边界划分,获取大量遥感影像图块,逐个分析与判断遥感影像图块类型,通过同第二次国土调查数据库相对比提取差异性地类图斑、偏移图斑、在建图斑,实现第三次国土调查内业信息提取。实验结果显示,该方法的遥感影像分割与融合效果好,可有效提升遥感图像清晰度,增强图像细节特征,保证土地利用数据的真实性。
In order to ensure the authenticity of land use data,in this paper,a method was designed to extract the indoor information of the Third Land Survey based on multi-source remote sensing data.Based on the panchromatic and multispectral images collected by gaofen-2 satellite,the S-G filter was applied to realize the fusion of Panchromatic and multispectral images through noise elimination,registration and resampling,and the pixel brightness correction was explored to improve the pixel brightness saturation of the fused image.According to the minimum cost cutting,the boundary of different land types in multi-source remote sensing images as divided,a large number of remote sensing image blocks were obtained,and the types of remote sensing image blocks were analyzed and judged one by one.By comparing with the database of the second land survey,the differences of land type patches,offset patches and map patches under construction were extracted,so as to realize the office information extraction of the third land survey.The experimental results showed that the segmentation and fusion effect of this method was good,which could effectively improve the clarity of remote sensing image,enhance the image detail characteristics,and ensure the authenticity of land use data.
作者
陈宗成
CHEN Zongcheng(Shandong Lunan Institute of Geological Engineering Investigation, Yanzhou Shandong 272100, China;The Second Geological Brigade of Shandong Geological and Mineral Resources Exploration and Development Bureau, Yanzhou Shandong 272100, China;Engineering Technology Research Center for comprehensive treatment of Coal Mining Subsidence Area, Ministry of Natural Resources, Yanzhou Shandong 272100, China)
出处
《北京测绘》
2021年第11期1394-1399,共6页
Beijing Surveying and Mapping
基金
山东省重点研发计划(重大科技创新工程)(2020CXGC011403)。
关键词
多源遥感数据
国土调查
内业信息提取
图像融合
图像分割
图斑
Multi source remote sensing data
land survey
office information extraction
image fusion
image segmentation
image patch
