Land cover classification of finer resolution remote sensing data integrating temporal features from time series coarser resolution data

Land cover classification of finer resolution remote sensing data is always difficult to acquire high-frequency time series data which contains temporal features for improving classification accuracy. This paper proposed a method of land cover classification with finer resolution remote sensing data integrating temporal features extracted from time series coarser resolution data. The coarser resolution vegetation index data is first fused with finer resolution data to obtain time series finer resolution data. Temporal features are extracted from the fused data and added to improve classification accuracy. The result indicates that temporal features extracted from coarser resolution data have significant effect on improving classification accuracy of finer resolution data, especially for vegetation types. The overall classification accuracy is significantly improved approximately 4% from 90.4% to 94.6% and 89.0% to 93.7% for using Landsat 8 and Landsat 5 data, respectively. The user and producer accuracies for all land cover types have been improved.     

Characterization of aboveground biomass in an unmanaged boreal forest using Landsat temporal segmentation metrics

Canada is dominated by forested ecosystems which are subject to various inventory and management practices, with more northern boreal forests subject to neither. Our objectives were to measure the capacity of temporal trajectory metrics for estimating selected forest attributes in a northern Canadian boreal forest context using Landsat imagery and investigate the importance of different types of temporal trajectory metrics. Results indicated that Wetness was the best Tasseled Cap (TC) component for aboveground biomass estimation (R² = 50%, RMSE% = 56%), and the combination of simple and complex metrics from all TC components produced the highest R² (62%) and lowest RMSE% (49%). Using a similar combination of variables, other forest attributes were estimated equally reliably with lower RMSE% values. The most important temporal trajectory metrics were simple and described TC component values at each point of change in the temporal trajectory, however the most important variables overall were environmental variables.     

基于遥感影像的黄河内蒙古段冰-水分类研究

及时获取凌汛期河冰和水体的空间分布特征,对于预测冰凌灾害、提高防凌信息化管理水平有重要意义。遥感技术是当前获取河冰和水体空间分布的最主要手段之一。但是,黄河水体有大量悬浮泥沙,这给基于遥感技术的高精度冰-水分类带来了挑战。以黄河内蒙古段为例,基于Landsat 8 OLI遥感影像数据,在利用归一化积雪指数（NDSI）及河道矢量数据排除无关地物的基础上,对比了近红外波段反射率值、归一化差异水体指数（NDWI）、归一化积雪指数（NDSI）、改进的归一化积雪指数（MNDSI）以及归一化差异未封冻水体指数（NDUWI）在黄河内蒙古段典型河道河冰、水体分类中的表现,计算各指标总体分类精度及Kappa系数并进行阈值稳定性分析。结果表明：在利用NDSI和高清历史影像排除河道外无关地物后,NDUWI在各子段影像中的总体分类精度和Kappa系数均达到90.00%及0.90以上,其河冰、水体最优区分阈值大体分布于阈值中值附近。研究结果可为凌汛期黄河冰凌监测方法的选取以及冰上爆破位置的拟定提供依据。     

An ensemble pansharpening approach for finer-scale mapping of sugarcane with Landsat 8 imagery

We tested the effects of three fast pansharpening methods – Intensity-Hue-Saturation (IHS), Brovey Transform (BT), and Additive Wavelet Transform (AWT) – on sugarcane classification in a Landsat 8 image (bands 1–7), and proposed two ensemble pansharpening approaches (band stacking and band averaging) which combine the pixel-level information of multiple pansharpened images for classification. To test the proposed ensemble pansharpening approaches, we classified “sugarcane” and “other” land cover in the unsharpened Landsat multispectral image, the individual pansharpened images, and the band-stacked and band-averaged ensemble images using Support Vector Machines (SVM), and assessed the classification accuracy of each image. Of the individual pansharpened images, the AWT image achieved higher classification accuracy than the unsharpened image, while the IHS and BT images did not. The band-stacked ensemble images achieved higher classification accuracies than the unsharpened and individual pansharpened images, with the IHS-BT-AWT band-stacked image producing the most accurate classification result, followed by the IHS-BT band-stacked image. The ensemble images containing averaged pixel values from multiple pansharpened images achieved lower classification accuracies than the band-stacked ensemble images, but most still had higher accuracies than the unsharpened and individual pansharpened results. Our results indicate that ensemble pansharpening approaches have the potential to increase classification accuracy, at least for relatively simple classification tasks. Based on the results of the study, we recommend further investigation of ensemble pansharpening for image analysis (e.g. classification and regression tasks) in agricultural and non-agricultural environments.     

基于 Landsat 8的城市热岛效应研究初探--以厦门市为例

以厦门市为研究区域,以2013年的Landsat 8为研究数据,研究亮度温度反演和地表温度反演算法,通过ENVI/IDL软件编程实现,并且利用厦门市气象局实测温度验证地表反演结果的有效性。同时,还分析了当前厦门市热岛效应空间分布情况,以期为厦门市城市规划和建设旅行城市等工作提供重要参考意义。     

北京地区Landsat 8 OLI高空间分辨率气溶胶光学厚度反演

卫星气溶胶光学厚度(AOD)反演中,传统暗目标方法在反射率较低的水体、浓密植被覆盖区域取得了较好效果,在反射率较高且结构复杂的高反射地表上空目前多采用深蓝算法,但存在空间分辨率较低,对细节分布描述性较差等问题。为解决这一问题,本文首先以5年(2008年—2012年)长时间序列MODIS地表反射率产品为基础,采用最小值合成法建立500 m分辨率逐月地表反射率产品数据集,然后利用地物波谱库中典型地物波谱数据,分析建立MODIS与Landsat 8 OLI传感器蓝光波段反射率转换模型,最后北京地区AERONET地基观测数据确定了气溶胶光学物理参数,并反演获取了北京地区上空500 m分辨率的AOD分布。为验证反演算法的精度,分别将反演结果同AERONET及MODIS/Terra气溶胶产品(MOD04)进行交叉对比,同时利用相关系数R,均方根误差RMSE,平均绝对误差MAE以及MODIS AOD产品预期误差EE共4个指标进行衡量。结果表明:算法反演获取的AOD与AERONET观测值具有较高的一致性,各指标分别为R=0.963,RMSE=0.156,MAE=0.097,EE=85.3%,稍优于MOD04产品(R=0.962,RMSE=0.158,MAE=0.101,EE=75.8%),并且有效的对比点数也高于MOD04。通过与地基观测相比,卫星遥感获取的高分辨率城市地区AOD精度可作为定量评估城市空气质量的有效依据。     

遥感在大兴安岭森林覆盖区地质矿产调查中的应用——以黑龙江洛古河1:5万区域地质矿产调查工作为例

大兴安岭地区森林覆盖严重, 气候严寒, 交通极为不便, 野外有效工作时间短, 给区域地质矿产调查工作增加了难度, 急需遥感手段提高成果质量和效率. 在黑龙江大兴安岭洛古河等4幅1:5万区域地质矿产调查工作中, 利用SPOT7、Landsat7/8、ASTER等多种遥感数据开展地质矿产解译, 进行遥感影像分区, 建立地层、构造和侵入岩解译标志, 提取羟基和铁染蚀变异常, 结合水系沉积物测量成果划分了成矿有利区, 有效降低了地质矿产调查强度, 提高了调查效率, 增强了调查质量. 表明遥感技术在大兴安岭高植被覆盖区地质矿产调查过程中能够取得较好效果.     

Recent trends on glacier area retreat over the group of Nevados Caullaraju-Pastoruri (Cordillera Blanca,Peru) using Landsat imagery

The Cordillera Blanca, located in the central zone of the Andes Mountains in Peru, has shown a retreat in its glaciers. This paper presents a trend analysis of the glacier area over the groups of Nevados Caullaraju-Pastoruri from 1975 to 2010 using Landsat-5 Thematic Mapper (TM) imagery. In the case of the Nevados Pastoruri/Tuco, the study period was extended back to 1957 by using an aerial photograph taken that year. The extent of clean glacier ice was estimated using Normalized Difference Snow Index (NDSI) thresholds. Moreover, the estimation of debris-covered glacier ice was retrieved by means of a decision tree classification method using NDSI, Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST). Area estimations derived from Landsat imagery were compared to the glacier ground-truth data in 1975 and 2010. Results show a statistically significant (p < 0.05) decreasing trend over the whole study area. Total glacier area decreased at a rate of 4.5 km² per decade from 1975 to 2010, with a total loss of 22.5 km² (58%). Lower decreasing rates were found for the period 1987–2010: 3.5 km² per decade with a total loss of 7.7 km² (32.5%). In the case of the Nevados Pastoruri/Tuco, decreasing rates of clean ice extent were constant for the periods 1957–2010, 1975–2010 and 1987–2010, with values close to 1.4 km² per decade and a total loss between 1957 and 2010 estimated at about 5 km² (54%). This work shows an evident area decrease in the Caullaraju-Pastoruri tropical glaciers, which needs to be included in a future hydrological scenario of local adaptability and water management.     

Landsat 8数据地表温度反演算法对比

随着卫星遥感技术的发展,利用遥感反演地表温度的方法不断出现,如劈窗法、双角度法和单通道算法等。Landsat系列卫星的遥感数据是地表温度反演的重要数据之一。本文选择无锡周边区域为研究区,利用Landsat 8卫星遥感数据,对两种劈窗算法(Juan C.Jiménez-Muoz劈窗算法和Offer Rozenstein劈窗算法)和两种单窗算法(Juan C.Jiménez-Muoz单通道算法和覃志豪单窗算法)的地表温度反演精度进行了对比和敏感性分析。采用太湖16个浮标站的实测数据来验证了4种算法的反演精度。结果表明:两种劈窗算法的精度较高且较为接近,误差为0.7 K左右;覃志豪单窗算法和Juan C.Jiménez-Muoz单通道算法精度较低,误差分别为1.3 K和1.4 K左右。Juan C.Jiménez-Muoz劈窗算法对参数的敏感性最低,Juan C.Jiménez-Muoz单通道算法次之,覃志豪单窗算法和Offer Rozenstein劈窗算法敏感性相对最高。其中Juan C.Jiménez-Muoz单通道算法只适用于一定的水汽含量范围,有一定的局限性。     

资源一号02C与Landsat8影像融合方法对比分析

针对以往关于资源一号02C和Landsat8卫星影像数据融合的研究不足的问题,该文利用前者在空间分辨率上高于后者、后者具有前者所不具有的光谱信息这一特性,选取主成分变换法、比值变换法、色彩变换法、高通滤波法和超分辨率贝叶斯法5种融合方法,分别对两种数据本身及数据间进行融合,并利用定性与定量的方法对融合结果进行评价,得出:资源一号02C星全色波段与多光谱波段数据融合结果中高通滤波法与超分辨率贝叶斯法效果较好,Landsat8OLI全色波段与多光谱数据融合结果中高通滤波法效果最好,资源一号02C星全色波段与Landsat8OLI多光谱数据融合结果中高通滤波法效果最好。