Optimum pixel size for hyperspectral studies of ecosystem function in southern California chaparral and grassland

Hyperspectral remotely sensed data are useful for studying ecosystem processes and patterns. However, spatial characterization of such remotely sensed images is needed to optimize sampling procedures and address scaling issues. We have investigated spatial scaling in ground-based and airborne hyperspectral data for canopy- to watershed-level ecosystem studies of southern California chaparral and grassland vegetation. Three optical reflectance indices, namely, Normalized Difference Vegetation Index (NDVI), Water Band Index (WBI) and Photochemical Reflectance Index (PRI) were used as indicators of biomass, plant water content and photosynthetic activity, respectively. Two geostatistical procedures, the semivariogram and local variance, were used for the spatial scaling analysis of these indices. The results indicate that a pixel size of 6 m or less would be optimal for studying functional properties of southern California grassland and chaparral ecosystems using hyperspectral remote sensing. These results provide a guide for selecting the spatial resolution of future airborne and satellite-based hyperspectral sensors.     

Combining vegetation indices, constrained ordination and fuzzy classification for mapping semi-natural vegetation units from hyperspectral imagery

Vegetation mapping of plant communities at fine spatial scales is increasingly supported by remote sensing technology. However, combining ecological ground truth information and remote sensing datasets for mapping approaches is complicated by the complexity of ecological datasets. In this study, we present a new approach that uses high spatial resolution hyperspectral datasets to map vegetation units of a semiarid rangeland in Central Namibia. Field vegetation surveys provide the input to the workflow presented in this study. The collected data were classified by hierarchical cluster analysis into seven vegetation units that reflect different ecological states occurring in the study area. Spectral indices covering vegetation and soil characteristics were calculated from hyperspectral remote sensing imagery and used as environmental variables in a constrained ordination by applying redundancy analysis (RDA). The resulting statistical relationships between vegetation data and spectral indices were transferred into images of ordination axes, which were subsequently used in a supervised fuzzy c-means classification approach relying on a k-NN distance metric. Membership images for each vegetation unit as well as a confusion image of the classification result allowed a sound ecological interpretation of the resulting hard classification map. Classification results were validated with two independent reference datasets. For an internal and external validation dataset, overall accuracy reached 98% and 64% with kappa values of 0.98 and 0.53, respectively. Critical steps during the mapping workflow were highlighted and compared with similar mapping approaches.     

Mangrove above-ground carbon stock mapping of multi-resolution passive remote-sensing systems

This research address several issues related to mangrove above-ground carbon stock (AGC) mapping using the integration of remote-sensing images and field data. These issues are (1) remote-sensing image availability for specific mangrove AGC mapping scale and precision, (2) the impact on mangrove AGC modelling due to the difference between images spatial resolution and plot size of field mangrove AGC measurement, which follow the standardized procedure and not specially developed to be integrated with remote-sensing data, and (3) the accuracy of performing mangrove AGC mapping using image at different spatial resolutions using similar field size mangrove AGC data. Four multispectral data sets, namely Worldview-2, Advanced Land Observation System Advanced Visible and Near-Infrared Radiometer-2 (ALOS AVNIR-2), Advanced Spectral and Thermal Radiometer (ASTER) Visible Near-Infrared (VNIR) and Landsat 8 Operational Land Imager (OLI), and a Hyperion hyperspectral image were tested for their performance for mangrove AGC mapping. These images represent various spatial, spectral, and radiometric resolutions of remote-sensing data available to date. The mapping was performed using their original spatial resolution, and for Worldview-2 the mapping was also conducted using 10 m spatial resolution. Image radiometric corrections, vegetation indices, principal component analysis and minimum noise fraction were applied to each image. These were used as input in the empirical modelling of mangrove AGC. The results indicate that (1) it is not possible to perform empirical modelling of mangrove AGC using image with sub-canopy spatial resolution, (2) decreasing the spatial resolution may be beneficial to obtaining a significant correlation with mangrove AGC, and (3) it is possible to perform empirical modelling of mangrove AGC mapping using field data not specially intended to be integrated with remote-sensing data, along with some adjustments. This opens up the possibility of utilizing the available field mangrove AGC data collected by stakeholders, that is, government institutions, NGOs, academics, private sector, to assist mangrove AGC mapping across the nation.     

Non-photosynthetic vegetation biomass estimation in semiarid Canadian mixed grasslands using ground hyperspectral data,Landsat 8 OLI,and Sentinel-2 images

ABSTRACT

Research on quantifying non-photosynthetic vegetation (NPV) with optical remote-sensing approaches has been focusing on optically distinguishing NPV from green vegetation and bare soil. With a very similar spectral response curve to NPV, dry moss is a significant component in semiarid mixed grasslands and plays a large role in NPV estimation. However, limited attention has been paid to this role. We investigated the potential of optical remote sensing to distinguish NPV biomass in semiarid grasslands characterized by NPV, biological soil crust dominated by moss and lichen, and bare soil. First, hyperspectral spectral indices were examined to determine the most useful spectral wavelength regions for NPV biomass estimation. Second, multispectral red-edge indices and shortwave infrared (SWIR) indices were simulated based on Landsat 8 Operational Land Imager (OLI) and Sentinel-2A MultiSpectral Instrument band reflectance, respectively, to determine the most suitable multispectral indices for NPV estimation. The potential multispectral indices were then applied to Landsat 8 OLI images and Sentinel-2A images acquired in early, middle, peak, and early senescence growing seasons to investigate the potential of satellite images for quantifying NPV biomass. Our results indicated that hyperspectral red-edge indices, modified simple ratio, modified red-edge normalized difference vegetation index (mNDVI₇₀₅), and normalized difference vegetation index (NDVI₇₀₅) are better than the SWIR hyperspectral indices, including cellulose absorption index for quantifying NPV biomass. The simulated multispectral red-edge spectral indices (NDVI_red-edge and mNDVI_red-edge) demonstrate good and comparable performance on quantifying NPV biomass with SWIR multispectral indices (normalized difference index [NDI5 and NDI7] and soil-adjusted corn residue index). Nevertheless, the multispectral indices derived from Landsat 8 OLI and Sentinel-2 images have limited potential for NPV biomass estimation.     

Cloud removal for hyperspectral remotely sensed images based on hyperspectral information fusion

ABSTRACT

Hyperspectral remote sensing plays an important role in a wide variety of fields. However, its specific application for land surface analysis has been constrained due to the different shapes of thick, opaque cloud cover. The reconstruction of missing information obscured by clouds in remote-sensing images is an area of active research. However, most of the available cloud-removal methods are not suitable for hyperspectral images, because they lose the spectral information which is very important for hyperspectral analysis. In this article, we developed a new spectral resolution enhancement method for cloud removal (SREM-CR) from hyperspectral images, with the help of an auxiliary cloud-free multispectral image acquired at different times. In the fixed hyperspectral image, spectra of the cloud cover pixels are reconstructed depending on the relationship between the original hyperspectral and multispectral images. The final resulting image has the same spectral resolution as the original hyperspectral image but without clouds. This approach was tested on two experiments, in which the results were compared by visual interpretation and statistical indices. Our method demonstrated good performance.     

Evaluating hyperspectral imaging of wetland vegetation as a tool for detecting estuarine nutrient enrichment

Nutrient enrichment and eutrophication are major concerns in many estuarine and wetland ecosystems, and the need is urgent for fast, efficient, and synoptic ways to detect and monitor nutrients in wetlands and other coastal systems across multiple spatial and temporal scales. We integrated three approaches in a multi-disciplinary evaluation of the potential for using hyperspectral imaging as a tool to assess nutrient enrichment and vegetation responses in tidal wetlands. For hyperspectral imaging to be an effective tool, spectral signatures must vary in ways correlated with water nutrient content either directly, or indirectly via such proxies as vegetation responses to elevated nitrogen. Working in Elkhorn Slough, central California, where intensive farming practices generate considerable runoff of fertilizers and pesticides, we looked first for long- and short-term trends among temporally ephemeral point data for nutrients and other water quality characters collected monthly at 18 water sampling stations since 1988. Second, we assessed responses of the dominant wetland plant, Salicornia virginica (common pickleweed) to two fertilizer regimes in 0.25 m² experimental plots, and measured changes in tissue composition (C, H, N), biomass, and spectral responses at leaf and at canopy scales. Third, we used HyMap hyperspectral imagery (126 bands; 15–19 nm spectral resolution; 2.5 m spatial resolution) for a synoptic assessment of the entire wetland ecosystem of Elkhorn Slough. We mapped monospecific Salicornia patches (～ 56–500 m²) on the ground adjacent to the 18 regular water sampling sites, and then located these patches in the hyperspectral imagery to correlate long-term responses of larger patches to water nutrient regimes. These were used as standards for correlating plant canopy spectral responses with nitrogen variation described by the water sampling program. There were consistent positive relationships between nitrogen levels and plant responses in both the field experiment and the landscape analyses. Two spectral indices, the Photochemical Reflectance Index (PRI) and Derivative Chlorophyll Index (DCI), were correlated significantly with water nutrients. We conclude that hyperspectral imagery can be used to detect nutrient enrichment across three spatial and at least two temporal scales, and suggest that more quantitative information could be extracted with further research and a greater understanding of physiological and physical mechanisms linking water chemistry, plant properties and spectral imaging characteristics.     

The relationships between electrical conductivity of soil and reflectance of canopy,grain, and leaf of rice in northeastern Thailand

Soil salinity is a global environmental problem and the most widespread land degradation problem that reduces crop yields and agricultural productivity. The characteristic of soil salinity is conventionally measured by the electric conductivity (EC) of soil while remote-sensing techniques have been extensively applied to detect the presence of salts indirectly through the vegetation using crop spectral reflectance. This study aims primarily to investigate whether salt stress the rice can be detected by field reflectance or not, and second, to search the significant bands of vegetation indices that can indicate the relationships between the EC of soil and field hyperspectral reflectance of canopy, grain, and leaf of rice, using the normalized difference spectral index (NDSI). Field investigations on various paddy fields in northeastern Thailand were carried out in late November 2010 during the ripening season just before harvest in an attempt to realize the applications of the field hyperspectral technique for monitoring the spread of saline soils and estimation of the effects of soil salinity on rice plants. Jasmine rice and glutinous rice were two different rice species selected for this study. Rice plant investigations were conducted by collecting data on crop length, panicle length, canopy openness, leaf area index, and digital photographs of plant conditions from each site. The statistical analysis revealed that the changes in soil EC were significantly sensitive to the ripening stages of both jasmine rice and glutinous rice planted on different levels of soil salinity. Among reflectance measurements, canopy reflectance was highly correlated with soil EC. However, the estimated accuracies of the relationship between soil EC and reflectance of glutinous rice were relatively lower than those of jasmine rice.     

Detection of stressed oil palms from an airborne sensor using optimized spectral indices

Existing vegetation indices and red-edge techniques have been widely used for the assessment of vegetation status and vegetation health from remote-sensing instruments. This study proposed and applied optimized Airborne Imaging Spectrometer for Applications (AISA) airborne hyperspectral indices in assessing and mapping stressed oil palm trees. Six vegetation indices, four red-edge techniques, a standard supervised classifier and three optimized AISA spectral indices were compared in mapping diseased oil palms using AISA airborne hyperspectral imagery. The optimized AISA spectral indices algorithms used newly defined reflectance values at wavelength locations of 734 nm (near-infrared (NIR)) and 616 nm (red). The selection of these two bands was based on laboratory statistical analysis using field spectroradiometer reflectance data. These two bands were then applied to the AISA airborne hyperspectral imagery using the three optimized algorithms for AISA data. The newly formulated AISA hyperspectral indices were D2 = R ₆₁₆/R ₇₃₄, normalized difference vegetation index a (NDVIa)?=?(R ₇₃₄ – R ₆₁₆)/(R ₇₃₄?+?R ₆₁₆) and transformed vegetation index a (TVIa)?=?((NDVIa?+?0.5)/(abs (NDVIa?+?0.5))?×?[abs (NDVIa?+?0.5)]^1/2. The classification results from the optimized AISA hyperspectral indices were compared with the other techniques and the optimized AISA spectral indices obtained the highest overall accuracy. D2 and NDVIa obtained 86% of overall accuracy followed by TVIa with 84% of overall accuracy.     

Book reviews

Proximal and remote sensing measurements were used to calculate different vegetation indices that were applied as predictors of gross primary production (GPP), total ecosystem respiration (TER), net ecosystem production (NEP) and leaf area index (LAI). Reflectance data and carbon fluxes were collected during the 2005 growing season at a mountain grassland site in the Italian Alps. Significant relationships were found between GPP, TER, NEP, LAI and the most commonly used spectral vegetation indices, the Normalized Difference Vegetation Index (NDVI) and Green‐NDVI. Saturation of the spectral indices was evident in the estimation of both biophysical and ecophysiological parameters. Among the different indices, Green‐NDVI was less affected by saturation on both a spatial and a temporal basis. Therefore, the use of an additional green‐band sensor for spectral measurements at eddy covariance grassland sites is recommended. Concerning the bandwidth for the calculation of the indices, the highest predictive capacities among the sensor simulations included in the analysis were those of the Moderate Resolution Imaging Spectroradiometer (MODIS) and the high‐resolution hyperspectral instrument Hyperion, indicating the advantage of narrow bands for the prediction of plant parameters. Further analyses are, however, required to investigate the relationships between NEP, GPP and vegetation indices retrieved from satellite platforms, using the bands available on MODIS and Hyperion sensors.     

A tutorial on modelling and inference in undirected graphical models for hyperspectral image analysis

ABSTRACT

Undirected graphical models have been successfully used to jointly model the spatial and the spectral dependencies in earth observing hyperspectral images. They produce less noisy, smooth, and spatially coherent land-cover maps and give top accuracies on many datasets. Moreover, they can easily be combined with other state-of-the-art approaches, such as deep learning. This has made them an essential tool for remote-sensing researchers and practitioners. However, graphical models have not been easily accessible to the larger remote-sensing community as they are not discussed in standard remote-sensing textbooks and not included in the popular remote-sensing software and toolboxes. In this tutorial, we provide a theoretical introduction to Markov random fields and conditional random fields-based spatial–spectral classification for land-cover mapping along with a detailed step-by-step practical guide on applying these methods using freely available software. Furthermore, the discussed methods are benchmarked on four public hyperspectral datasets for a fair comparison among themselves and easy comparison with the vast number of methods in literature which use the same datasets. The source code necessary to reproduce all the results in the paper is published on-line to make it easier for the readers to apply these techniques to different remote-sensing problems.     

Multisampling Compressive Video Spectroscopy

The coded aperture snapshot spectral imaging (CASSI) architecture has been employed widely for capturing hyperspectral video. Despite allowing concurrent capture of hyperspectral video, spatial modulation in CASSI sacrifices image resolution significantly while reconstructing spectral projection via sparse sampling. Several multiview alternatives have been proposed to handle this low spatial resolution problem and improve measurement accuracy, for instance, by adding a translation stage for the coded aperture or changing the static coded aperture with a digital micromirror device for dynamic modulation. State‐of‐the‐art solutions enhance spatial resolution significantly but are incapable of capturing video using CASSI. In this paper, we present a novel compressive coded aperture imaging design that increases spatial resolution while capturing 4D hyperspectral video of dynamic scenes. We revise the traditional CASSI design to allow for multiple sampling of the randomness of spatial modulation in a single frame. We demonstrate that our compressive video spectroscopy approach yields enhanced spatial resolution and consistent measurements, compared with the traditional CASSI design.     

Remote sensing of emergent and submerged wetlands: an overview

To plan for wetland protection and responsible coastal development, scientists and managers need to monitor changes in the coastal zone, as the sea level continues to rise and the coastal population keeps expanding. Advances in sensor design and data analysis techniques are now making remote-sensing systems practical and cost-effective for monitoring natural and human-induced coastal changes. Multispectral and hyperspectral imagers, light detection and ranging (lidar), and radar systems are available for mapping coastal marshes, submerged aquatic vegetation, coral reefs, beach profiles, algal blooms, and concentrations of suspended particles and dissolved substances in coastal waters. Since coastal ecosystems have high spatial complexity and temporal variability, they should be observed with high spatial, spectral, and temporal resolutions. New satellites, carrying sensors with fine spatial (0.4–4 m) or spectral (200 narrow bands) resolution, are now more accurately detecting changes in coastal wetland extent, ecosystem health, biological productivity, and habitat quality. Using airborne lidars, one can produce topographic and bathymetric maps, even in moderately turbid coastal waters. Imaging radars are sensitive to soil moisture and inundation and can detect hydrologic features beneath the vegetation canopy. Combining these techniques and using time-series of images enables scientists to study the health of coastal ecosystems and accurately determine long-term trends and short-term changes.     

Downscaling time series of MERIS full resolution data to monitor vegetation seasonal dynamics

Monitoring vegetation dynamics is fundamental for improving Earth system models and for increasing our understanding of the terrestrial carbon cycle and the interactions between biosphere and climate. Medium spatial resolution sensors, like MERIS, exhibit a significant potential to study these dynamics over large areas because of their spatial, spectral and temporal resolution. However, the spatial resolution provided by MERIS (300 m in full resolution mode) is not appropriate to monitor heterogeneous landscapes, where typical length scales of these dynamics rarely reach 300 m. We, therefore, motivate the use of data fusion techniques to downscale medium spatial resolution data (MERIS full resolution, FR) to a Landsat-like spatial resolution (25 m). An unmixing-based data fusion approach was applied to a time series of MERIS FR images acquired over The Netherlands. The selected data fusion approach is based on the linear mixing model and uses a high spatial resolution land use database to produce images having the spectral and temporal resolution as provided by MERIS, but a Landsat-like spatial resolution. A quantitative assessment of the quality of the fused images was done in order to test the validity of the proposed method and to evaluate the radiometric characteristics of the MERIS fused images. The resulting series of fused images was subsequently used to compute two vegetation indices specifically designed for MERIS: the MERIS terrestrial chlorophyll index (MTCI) and the MERIS global vegetation index (MGVI). These indices represent continuous fields of canopy chlorophyll (MTCI) and of the fraction of photosynthetically active radiation absorbed by the canopy (MGVI). Results indicate that the selected data fusion approach can be successfully used to downscale MERIS data and, therefore, to monitor vegetation dynamics at Landsat-like spatial, and MERIS-like spectral and temporal resolution.     

非负矩阵分解融合高光谱和多光谱数据

由于光谱分辨率和空间分辨率的制约以及物理条件的限制,高光谱数据具有很高的光谱分辨率而其空间分辨率却很低。因此,一般高光谱数据的空间分辨率往往低于仅有几个波段的多光谱数据的空间分辨率。高光谱数据和多光谱数据的融合可以得到同时具有高空间分辨率和高光谱分辨率的数据,进而应用于更高空间分辨率下地物的识别和分类。非负矩阵分解(Nonnegative Matrix Factorization)算法用于实现低空间分辨率高光谱数据和高空间分辨率多光谱数据的融合。首先利用顶点成分分析法VCA(Vertex Component Analysis)分解高光谱数据,得到初始的端元波谱矩阵和端元丰度矩阵;然后用非负矩阵分解算法交替地对高光谱数据和多光谱数据进行分解,得到高光谱分辨率的端元波谱矩阵和高空间分辨率的丰度矩阵;最后两个矩阵相乘得到高空间分辨率和高光谱分辨率的融合结果。在每一步非负矩阵分解过程中,数据之间的传感器观测模型用于分解矩阵的初始化。AVIRIS和HJ-1A数据实验结果分析表明:非负矩阵分解算法有效提高了高光谱数据的所有波长范围内波段数据的空间分辨率,而高精度的融合结果可用于地物的目标识别和分类。     

Estimation of Lambert parameter based on leaf-scale hyperspectral images using dichromatic model-based PCA

Low-altitude hyperspectral observation systems are promising sensing tools for acquisition of optical remote-sensing data under the humid subtropical climate in Japan. The system is also capable of acquiring leaf-scale optical information free from atmospheric effect. However, the leaf-scale hyperspectral data are affected by shading and various illumination conditions such that it is difficult to obtain consistent characteristics of the spectral information. The aim of this article is the extraction of Lambert coefficients as an inherent leaf spectral profile. In this work, we propose a dichromatic model-based principal component analysis on hyperspectral data by utilizing leaf-scale hyperspectral data in order to diminish the spectral difference caused by the illumination condition and bidirectional reflectance distribution function. The results show that indices of chlorophyll content based on the estimated Lambert coefficients are consistent with the growth stages of a paddy field, whether the illumination condition is clear sky or overcast.     

波段自适应细节注入的高分五号与Sentinel-2遥感影像空谱融合

目的 针对当前空谱融合方法应用到高光谱图像融合时,出现的空间细节信息提升明显但光谱失真,或者光谱保真度高但空间细节信息提升不足的问题,本文提出一种波段自适应细节注入的高分五号（GF-5）高光谱图像（30 m）与Sentinel-2多光谱图像（10 m）的遥感影像空谱融合方法。方法 首先,为了解决两个多波段图像不便于直接融合的问题,提出一种波段自适应的融合策略,对多光谱图像波谱范围以外的高光谱图像波段,以相关系数为标准将待融合图像进行分组。其次,针对传统Gram-Schmidt （GS）融合方法用平均权重系数模拟低分辨率图像造成的光谱失真问题,使用最小均方误差估计计算线性拟合系数,再将拟合图像作为第1分量进行GS正变换,提升融合图像的光谱保真度。最后,为了能同时注入更多的空间细节信息,通过非下采样轮廓波变换将拟合图像、空间细节信息图像和多光谱图像的空间、光谱信息融入到重构的高空间分辨率图像中,再将其与其他GS分量一起进行逆变换,最终得到10 m分辨率的GF-5融合图像。结果 通过与当前用于高光谱图像空谱融合的典型方法比较,本文方法对于受时相影响较小的城镇区域,在提升空间分辨率的同时有较好的光谱保真度,且不会出现噪点;对于受时相变化影响大的植被密集区域,本文方法融合图像有较好的清晰度和地物细节信息,且没有噪点出现。本文方法的CC （correlation coefficient）、ERGAS （erreur relative globale adimensionnelle de synthèse）和SAM （spectral angle mapper）相比于传统GS方法分别提升8%、26%和28%,表明本文方法的光谱保真度大大提高。结论 本文方法的结果空间上没有噪点且光谱曲线与原始光谱曲线基本保持一致,是一种兼具高空间分辨率和高光谱保真度的高光谱图像融合方法。     

Comparison of spectral and spatial windows for local anomaly detection in hyperspectral imagery

A new way of implementing two local anomaly detectors in a hyperspectral image is presented in this study. Generally, most local anomaly detector implementations are carried out on the spatial windows of images, because the local area of the image scene is more suitable for a single statistical model than for global data. These detectors are applied by using linear projections. However, these detectors are quite improper if the hyperspectral dataset is adopted as the nonlinear manifolds in spectral space. As multivariate data, the hyperspectral image datasets can be considered to be low-dimensional manifolds embedded in the high-dimensional spectral space. In real environments, the nonlinear spectral mixture occurs more frequently, and these manifolds could be nonlinear. In this case, traditional local anomaly detectors are based on linear projections and cannot distinguish weak anomalies from background data. In this article, local linear manifold learning concepts have been adopted, and anomaly detection algorithms have used spectral space windows with respect to the linear projection. Output performance is determined by comparison between the proposed detectors and the classic spatial local detectors accompanied by the hyperspectral remote-sensing images. The result demonstrates that the effectiveness of the proposed algorithms is promising to improve detection of weak anomalies and to decrease false alarms.     

Land-cover classification using multi-temporal GF-1 wide field view data

ABSTRACT

Vegetation is an important land-cover type and its growth characteristics have potential for improving land-cover classification accuracy using remote-sensing data. However, due to lack of suitable remote-sensing data, temporal features are difficult to acquire for high spatial resolution land-cover classification. Several studies have extracted temporal features by fusing time-series Moderate Resolution Imaging Spectroradiometer data and Landsat data. Nevertheless, this method needs assumption of no land-cover change occurring during the period of blended data and the fusion results also present certain errors influencing temporal features extraction. Therefore, time-series high spatial resolution data from a single sensor are ideal for land-cover classification using temporal features. The Chinese GF-1 satellite wide field view (WFV) sensor has realized the ability of acquiring multispectral data with decametric spatial resolution, high temporal resolution and wide coverage, which contain abundant temporal information for improving land-cover classification accuracy. Therefore, it is of important significance to investigate the performance of GF-1 WFV data on land-cover classification. Time-series GF-1 WFV data covering the vegetation growth period were collected and temporal features reflecting the dynamic change characteristics of ground-objects were extracted. Then, Support Vector Machine classifier was used to land-cover classification based on the spectral features and their combination with temporal features. The validation results indicated that temporal features could effectively reflect the growth characteristics of different vegetation and finally improved classification accuracy of approximately 7%, reaching 92.89% with vegetation type identification accuracy greatly improved. The study confirmed that GF-1 WFV data had good performances on land-cover classification, which could provide reliable high spatial resolution land-cover data for related applications.     

Object-based land-cover change detection applied to Brazilian seasonal savannahs using geostatistical features

A new method for remote-sensing land-use/land-cover (LULC) change detection is proposed to eliminate the effects of forest phenology on classification results. This method is insensitive to spectral changes caused by vegetation seasonality and uses an object-based approach to extract geostatistical features from bitemporal Landsat TM (Thematic Mapper) images. We first create image objects by multiresolution segmentation to extract geostatistical features (semivariogram parameters and indices) and spectral information (average values) from NDVI (normalized difference vegetation index), acquired in the wet and dry seasons, as input data to train a Support Vector Machine algorithm. We also used the image difference traditional change-detection method to validate the effectiveness of the proposed method. We used two classes: (1) LULC change class and (2) seasonal change class. Using the most geostatistical features, the change detection results are considerably improved compared with the spectral features and image differencing technique. The highest accuracy was achieved by the sill (σ² overall variability) semivariogram parameter (95%) and the AFM (area first lag–first maximum) semivariogram index (88.33%), which were not affected by vegetation seasonality. The results indicate that the geostatistical context makes possible the use of bitemporal NDVI images to address the challenge of accurately detecting LULC changes in Brazilian seasonal savannahs, disregarding changes caused by phenological differences, without using a dense time series of remote-sensing images. The challenge of extracting accurate semivariogram curves from objects of long and narrow shapes requires further study, along with the relationship between the scale of segmentation and image spatial resolution, including the type of change and the initial land-cover class.     

The effect of spectral and spatial degradation of hyperspectral imagery for the Sclerophyll tree species classification

Hyperspectral imaging can be a useful remote-sensing technology for classifying tree species. Prior to the image classification stage, effective mapping endeavours must first identify the optimal spectral and spatial resolutions for discriminating the species of interest. Such a procedure may contribute to improving the classification accuracy, as well as the image acquisition planning. In this work, we address the effect of degrading the original bandwidth and pixel size of a hyperspectral and hyperspatial image for the classification of Sclerophyll forest tree species. A HySpex-VNIR 1600 airborne-based hyperspectral image with submetric spatial resolution was acquired in December 2009 for a native forest located in the foothills of the Andes of central Chile. The main tree species of this forest were then sampled in the field between January and February 2010. The original image spectral and spatial resolutions (160 bands with a width of 3.7 nm and pixel sizes of 0.3 m) were systematically degraded by resampling using a Gaussian model and a nearest neighbour method, respectively (until reaching 39 bands with a width of 14.8 nm and pixel sizes of 2.4 m). As a result, 12 images with different spectral and spatial resolution combinations were created. Subsequently, these images were noise-reduced using the minimum noise fraction procedure and 12 additional images were created. Statistical class separabilities from the spectral divergence measure and an assessment of classification accuracy of two supervised hyperspectral classifiers (spectral angle mapper (SAM) and spectral information divergence (SID)) were applied for each of the 24 images. The best overall and per-class classification accuracies (>80%) were observed when the SAM classifier was applied on the noise-reduced reflectance image at its original spectral and spatial resolutions. This result indicates that pixels somewhat smaller than the tree canopy diameters were the most appropriate to represent the spatial variability of the tree species of interest. On the other hand, it suggests that noise-reduced bands derived from the full image spectral resolution rendered the best discrimination of the spectral properties of the tree species of interest. Meanwhile, the better performance of SAM over SID may result from the ability of the former to classify tree species regardless of the illumination differences in the image. This technical approach can be particularly useful in native forest environments, where the irregular surface of the uppermost canopy is subject to a differentiated illumination.