Analysis of the landscape change at River Basin scale based on SPOT and TM fusion remote sensing images: a case study of the Weigou River Basin on the Chinese Loess Plateau

Eco-environmental changes which generally manifest as the ecological landscape changes are mainly affected by natural and human factors. Through complex interaction, natural factors and human activities alter the landscape structure and decide the service function of regional ecosystem. With the development of geographical information system (GIS) and satellite remote sensing (RS) techniques, the researches on quantitative analysis of landscape changes have made great strides forward. In this paper, the landscape change dynamics in the Weigou River basin have been investigated by the combined use of high-resolution RS images and GIS techniques. The objectives are to determine the landscape transition rates among landscape types and to quantify changes of various landscape indexes using FRAGSTATS, one of the spatial pattern analysis programs for categorical maps. Under man–machine interactive interpretation method, all of the 10-m resolution SPOT + TM images are classified into six primary types (farmland, forest, grassland, water, construction area and desert) and 12 sub-types based on the interpretation key which was established based on 1947 filed pictures in the Weigou River basin. A field check on the data accuracy shows that the total interpretation accuracy approaches 97.53%. Significant land-use change has taken place in the Weigou River basin over the six years from 1998 to 2004 due to rehabilitation measures. The results show that there has been a notable decrease in farmland mainly due to conversion to forest and grassland, the landscape heterogeneity and evenness has increased, and there is a greater connectivity. The dominance of farmland patch has decreased. And hence the ecological environment has started to develop in a reversing direction. The study of the integration of high-resolution RS images and GIS technique is an effective approach to analyze the landscape changes at river basin scale.     

城市绿地信息提取遥感影像尺度效应分析

绿地信息提取精度与影像的空间尺度有关,又与提取目标的特点有关。目前用于绿地信息提取的高分辨率遥感影像主要有QUICKBIRD、IKONOS和SPOT5,不同尺度影像对绿地信息提取精度的影响还缺少相关的研究。本文利用目视解译方法,研究三种尺度遥感影像提取绿地信息的精度。结果表明,三种影像中,由于QUICK-BIRD影像有更详细的空间结构,更适合于城市绿地零星分布特点,在以7层建筑物以下为主的研究区,绿地信息提取精度最高。     

用SPOT图像进行土地利用调查和动态监测研究

选取常熟市绿塘等5个乡为试验区，用1994年和1999年的SPOT资料作为信息源，用人机交互式解释法进行土地利用现状分类及动态监测，并将1994年调查结果与同年航空像片详查值作对比，进行了精度分析。将1999年SPOT卫片判读结果与1994年航片详查对比，可得到5乡一级土地分类的动态面积和动态率，其中耕地5a间的动态率为－18．3‰，并与常熟市国土局统计值得出的动态面积和动态率作了比较。研究结果表明：SPOT卫片监测县级土地利用是可行的，选择适宜的SPOT卫片时相以及剔除细小地物的影响，可提高解释判读精度。     

SPOT5在矿山监测中的应用

为了快速监控矿山的实际开采情况与矿山的登记现状之间的差异,维护矿产资源管理秩序,打击无证采矿、越界采矿,保护生态环境,促进依法办矿和科学办矿,先利用SPOT5图像进行目标确定,然后野外检查.通过研究发现,利用遥感手段对煤矿、砂石(建筑用材料)、铝土矿等矿山的监测效果明显.     

海岸带地区SPOT卫星影像大气校正方法比较及精度验证

为获得反演海岸带地区地表真实反射率的最佳大气校正方法,分别利用DOS、6S及FLAASH三种模型对SPOT卫星影像进行大气校正,对比分析校正前后影像的典型地物光谱曲线及NDVI指数的变化情况,并对结果进行精度验证。研究表明:三种大气校正模型均能有效去除大气对遥感影像的影响,获取较为准确的地表反射率信息;对比河流、水库、海水、植被、城镇、裸地等六种典型地物校正前后的光谱曲线,与DOS及6S模型相比,FLAASH模型校正结果中各地物反射率差别较大、特征明显,能较好的还原地表真实情况;NDVI指数检验显示,FLAASH模型对地物NDVI响应更强烈,在地物光谱响应特征分析等方面具有更强的优势,是三种模型中最适宜海岸带地区SPOT影像的大气反演方法。研究方法及结论可为同类研究提供参考与借鉴。     

Extracting vegetation phenology metrics in Changbai Mountains using an improved logistic model

Remotely sensing images are now available for monitoring vegetation dynamics over large areas.In this paper,an improved logistic model that combines double logistic model and global function was developed.Using this model with SPOT/NDVI data,three key vegetation phenology metrics,the start of growing season (SOS),the end of growing season (EOS) and the length of growing season (LOS),were extracted and mapped in the Changbai Mountains,and the relationship between the key phenology metrics and elevation were ...     

Precision rectification of high resolution satellite imagery without ephemeris data

The huge capability of high resolution satellite imageries (HRSI), that includes spatial, spectral, temporal and radiometric resolutions as well as stereoscopic vision introduces them as a powerful new source for the Photogrammetry, Remote Sensing and GIS communities. High resolution data increases the need for higher accuracy of data modeling. The satellite orbit, position, attitude angles and interior orientation parameters have to be adjusted in the geometrical model to achieve optimal accuracy with the use of a minimum number of Ground Control Points (GCPs). But most high resolution satellite vendors do not intend to publish their sensor models and ephemeris data. There is consequently a need for a range of alternative, practical approaches for extracting accurate 2D and 3D terrain information from HRSI. The flexibility and good accuracy of the alternative models demonstrated with KFA-1000 and the well-known SPOT level 1A images. A block of eight KFA-1000 space photos in two strips with 60% longitudinal overlap and 15% lateral sidelap and SPOT image with rational function, DLT, 2D projective, polynomials, affine, conformal, multiquadric and finite element methods were used in the test. The test areas cover parts of South and West of Iran. Considering the quality of GCPs, the best result was found with the DLT method with a RMSE of 8.44 m for the KFA-1000 space photos.     

用拟合法进行SPOT影象的近似核线排列

本文提出了一种SPOT影象的近似核线排列的新方法——多项式拟合法,它不需要预先已知SPOT影象的任何参数和地面高程信息。论文对此方法之原理、实际的作业过程与理论分析作了较详细的阐述,并通过实例证明此方法是有效和正确的。     

基于SPOT影像的筏式养殖区提取方法研究

本文以山东省荣成市黑泥湾及其北部海域作为研究区,分析了筏式养殖区在SPOT5假彩色合成影像上的亮度特征;根据纯水体亮度受悬沙分布影响较大的特点,设计了以邻域分析为基础的图像增强及分类方法,并在ArcGIS软件中实现;运用Erdas软件聚类统计、去除分析工具消除噪声,提高精度。该方法可快速监测浅海筏式养殖区空间分布情况,为提高用海管理水平并进行科学用海规划提供服务。