Search

This article introduces a special issue on hyperspectral remote sensing, which involves advanced satellite sensors that provide data in three categories: hyperspatial, superspectral, and hyperspectral. These data types offer higher resolutions but also present challenges in handling and analysis. The special issue aims to explore the characteristics and limitations of hyperspectral data and their opportunities for advancing remote sensing science. The selected papers in the issue cover various applications and use advanced techniques to provide valuable knowledge in the field. The document discusses three research papers related to remote sensing and hyperspectral data, focusing on mapping building shadows, identifying asphalt pavement aging, and separating wheat crops from similar crops. It emphasizes the potential of hyperspectral data in advancing remote sensing science while acknowledging the challenges associated with data analysis and calibration. [Extracted from the article]

Remote sensing applications require high-resolution images to obtain precise information about the Earth’s surface. Multispectral images have high spatial resolution but low spectral resolution. Hyperspectral images have high spectral resolution but low spatial resolution. This study proposes a residual learning and attention-based parallel network based on residual network and channel attention. The network performs image fusion of a high spatial resolution multispectral image and a low spatial resolution hyperspectral image. The network training and fusion experiments are conducted on four public benchmark data sets to show the effectiveness of the proposed model. The fusion performance is compared with classical signal processing–based image fusion techniques. Four image metrics are used for the quantitative evaluation of the fused images. The proposed network improved fusion ability by reducing the root mean square error and relative dimensionless global error in synthesis and increased the peak signal-to-noise ratio when compared to other state-of-the-art models. [ABSTRACT FROM AUTHOR]

Debris flow susceptibility evaluation plays a crucial role in the prevention and control of debris flow disasters. Therefore, this paper proposes a convolutional neural network model named multi-level feature extraction network (MFENet). First, a dual-channel CNN architecture incorporating the Embedding Channel Attention mechanism is used to extract shallow features from both digital elevation model images and multispectral images. Subsequently, channel shuffle and feature concatenation are applied to the features from the two channels to obtain fused feature sets. Following this, a deep feature extraction is performed on the fused feature sets using a residual module improved by maximum pooling. Finally, the susceptibility index of gullies to debris flows is calculated based on the similarity scores. Experimental results demonstrate that the model exhibits favorable classification performance, with an accuracy of 73.45%. Furthermore, the percentage of debris flow valleys in high and very high susceptibility zones reaches 93.97%. [ABSTRACT FROM AUTHOR]

It was difficult to accurately obtain crop planting structure by using the spectral information of high spatial resolution and low spatial resolution multispectral images of panchromatic images at the same time. In this paper, we propose a method of planting structure extraction based on indices and an improved U-Net semantic segmentation network. Based on the original band of Landsat-8, we used an image fusion algorithm to highlight the characteristics of vegetation, water, and soil respectively by three indices added, and the improved U-Net network was used to classify the type of planting structure. The results showed that the overall accuracy of classification was more than 91.6%, and the accuracy of crops was up to 93.8%. Automated water extraction index in image fusion effectively improved the classification accuracy. This method could extract a variety of information about planting structures automatically and accurately. It provided theoretical support for adjusting and optimizing regional planting structures. [ABSTRACT FROM AUTHOR]

This paper introduces a satellite-based cross-calibration (SCC) method for spectral reflectance estimation of unmanned aircraft system (UAS) multispectral imagery. The SCC method provides a low-cost and feasible solution to convert high-resolution UAS images in digital numbers (DN) to reflectance when satellite data is available. The proposed method is evaluated using a multispectral data set, including orthorectified KHawk UAS DN imagery and Landsat 8 Operational Land Imager Level-2 surface reflectance (SR) data over a forest/grassland area. The estimated UAS reflectance images are compared with the National Ecological Observatory Network's imaging spectrometer (NIS) SR data for validation. The UAS reflectance showed high similarities with the NIS data for the near-infrared and red bands with Pearson's r values being 97 and 95.74, and root-mean-square errors being 0.0239 and 0.0096 over a 32-subplot hayfield. [ABSTRACT FROM AUTHOR]

Multispectral satellite imagery is the primary data source for monitoring land cover change and characterizing land cover globally. However, the consistency of land cover monitoring is limited by the spatial and temporal resolutions of the acquired satellite images. The public availability of daily highresolution images is still scarce. This paper aims to fill this gap by proposing a novel spatiotemporal fusion method to enhance daily low spatial resolution land cover mapping using a weakly supervised deep convolutional neural network. We merge Sentinel images and moderate resolution imaging spectroradiometer (MODIS)-derived thematic land cover maps under the application background of massive remote sensing data and the large spatial resolution gaps between MODIS data and Sentinel images. The neural network training was conducted on the public data set SEN12MS, while the validation and testing used ground truth data from the 2020 IEEE Geoscience and Remote Sensing Society data fusion contest. The proposed data fusion method shows that the synthesized land cover map has significantly higher spatial resolution than the corresponding MODIS-derived land cover map. The ensemble approach can be implemented for generating highresolution time series of satellite images by fusing fine images from Sentinel-1 and -2 and daily coarse images from MODIS. [ABSTRACT FROM AUTHOR]

This article presents a novel strategy for improving the wellestablished component substitution-based multispectral image fusion methods, because the fused results obtained by component substitution methods tend to exhibit significant spectral distortion. The main cause of spectral distortion is analyzed and discussed based on the component substitution method's general model. An improved scheme is derived from the sensitivity imaging model to refine the approximate spatial detail and obtain one that is almost ideal. The experimental results on two data sets show that when it has been integrated into the Gram-Schmidt method and the generalized intensity-hue-saturation method, the proposed scheme allows the production of fused images of the same spatial sharpness as standard implementations but with significantly increased spectral quality. Quantitative scores and visual inspection at full resolution and spatially reduced resolution confirm the superiority of the improved methods over the conventional algorithms. [ABSTRACT FROM AUTHOR]

By 2008, OSM data was available for downloading to GPS receivers, and its popularity further increased. [Extracted from the article]

In this paper, we propose a machine learning-based oil spill segmentation using aerial images. In detail, a novel deep neural network-based object segmentation, named Deep Believe Active Contours (DBAC), is introduced, where a pre-trained deep belief neural network is utilized to guide the moments of active contours. Results show that (1) Unsupervised pre-trained deep neural network can efficiently control the evolution of active contour segmentation of oil spill regions; and (2) When applying the proposed DBAC algorithm on the test data from an oil spill image database, it produced a recall rate of 66% and a precision rate of 60%, which outperformed the state-of-the-art methods in the range of 4% ~ 18% and 1% ~ 10%, respectively. Moreover, DBAC produced a better Hausdorff distance (an amount of 13.34) compared to the competing methods. These results show the promises of DBAC for the task of oil spill segmentation in ocean environment. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging has been successfully utilized to locate clandestine graves. This study applied a Genetic Programming technique called Brain Programming (BP) for automating the design of Hyperspectral Visual Attention Models (H-VAM.), which is proposed as a new method for the detection of buried remains. Four graves were simulated and monitored during six months by taking in situ spectral measurements of the ground. Two experiments were implemented using Kappa and weighted Kappa coefficients as classification accuracy measures for guiding the BP search of the best H-VAM. Experimental results demonstrate that the proposed BP method improves classification accuracy compared to a previous approach. A better detection performance was observed for the image acquired after three months from burial. Moreover, results suggest that the use of spectral bands that respond to vegetation and water content of the plants and provide evidence that the number of buried bodies plays a crucial role on a successful detection. [ABSTRACT FROM AUTHOR]

The estimation of land cover fractions from remote sensing images is a frequently used indicator of the environmental quality. This paper focuses on the quantification of land cover fractions in an urban area of Berlin, Germany, using simulated hyperspectral EnMAP data with a spatial resolution of 30 m × 30 m. We use constrained sparse representation, where each pixel with unknown surface characteristics is expressed by a weighted linear combination of elementary spectra with known land cover class. We automatically determine the elementary spectra from image reference data using archetypal analysis by simplex volume maximization, and combine it with reversible jump Markov chain Monte Carlo method. In our experiments, the estimation of the automatically derived elementary spectra is compared to the estimation obtained by a manually designed spectral library by means of reconstruction error, mean absolute error of the fraction estimates, sum of fractions, R2, and the number of used elementary spectra. The experiments show that a collection of archetypes can be an adequate and efficient alternative to the manually designed spectral library with respect to the mentioned criteria. [ABSTRACT FROM AUTHOR]

Object-based image analysis and machine learning classification procedures, after field calibration and photogrammetric processing of consumer-grade unmanned aerial vehicle (UAV) digital camera data, were implemented to classify tree species in a conifer forest in the Great Lakes/St Lawrence Lowlands Ecoregion, Ontario, Canada. A red-green-blue (RGB) digital camera yielded approximately 72 percent classification accuracy for three commercial tree species and one conifer shrub. Accuracy improved approximately 15 percent, to 87 percent overall, with higher radiometric quality data acquired separately using a digital camera that included near infrared observations (at a lower spatial resolution). Interpretation of the point cloud, spectral, texture and object (tree crown) classification Variable Importance (VI) selected by a machine learning algorithm suggested a good correspondence with the traditional aerial photointerpretation cues used in the development of well-established large-scale photography northern conifer elimination keys, which use three-dimensional crown shape, spectral response (tone), texture derivatives to quantify branching characteristics, and crown size, development and outline features. These results suggest that commonly available consumer-grade UAV-based digital cameras can be used with object-based image analysis to obtain acceptable conifer species classification accuracy to support operational forest inventory applications. [ABSTRACT FROM AUTHOR]

State transportation agencies regularly collect data on pavement surface distresses. These data are used to assess overall pavement conditions and to make maintenance and repair decisions. Routinely-acquired and publically-available high spatial resolution (hsr) multispectral digital aerial photography provides a potential method for collecting distress information that can supplement or replace currently-used technologies. Principal component analysis and linear least squares regression models were used to evaluate the potential of using HSR multispectral digital aerial photographs to estimate pavement surface overall distress conditions. Various models were developed using HSR multispectral digital aerial photographs of different spatial resolution (6-inch, 12-inch, and 24-inch) and reference pavement surface distress data collected manually at multiple sample sites using standard protocols. The results show that the spectral response of HSR multispectral digital aerial photographs correlate strongly with reference distress rates at all tested spatial resolutions, but the 6-inch aerial photos exhibit the strongest correlation (R 2 > 0.95), even when using only half of the sample sites (R 2 > 0.92). These results indicate that straightforward analysis of HSR multispectral digital aerial photographs, routinely acquired by most municipalities and states, can permit assessment of pavement surface distress conditions as well as current manual evaluation protocols. [ABSTRACT FROM AUTHOR]

Multispectral images of very high spatial resolution and vector data from geospatial databases, such as cadastral maps, are among the cost-effective and broadly available geodata in urban environments. Therefore, we aim to address building change detection based on pre-existing building footprint information and a single very high resolution multispectral image. An object-based classification methodology was developed that employs advanced scale-space filtering, unsupervised clustering, and knowledge-based classification. The developed framework effectively integrates prior vector data and multispectral observations, through incorporating the prior knowledge into the training process and defining the proper object-based classification rules. The methodology successfully identified important building changes, which were validated by employing the vector information of a building geodatabase and a QuickBird image acquired in 2003 and 2007, respectively, over urban regions in the city of Thessaloniki, Greece. The performed quantitative and qualitative evaluation indicated that the proposed analysis framework can detect the new buildings with high accuracy rates and, to a lesser degree, their exact shape and size. [ABSTRACT FROM AUTHOR]

The ZiYuan-3 satellite allows users to obtain triplet along-track stereo and four-band multispectral images within a single pass. Its basic sensor-corrected products are generated by reimaging with a virtual CCD array in which inner and exterior distortions are resolved and the introduced errors are negligible. In this study, three zY-3 datasets (Dengfeng, Anping, and Taihang) were used to evaluate the accuracy of bundle adjustment using triplet stereo images. For both standard and long-strip scenes, the RMS errors of plane and elevation were 2.6 m and 1.6 m, respectively, when ground control points (GCPs) were deployed around the corners. The digital surface model (DSM) of the Dengfeng dataset was compared with a reference DEM (1:2000); the RMS errors produced varied with topography, with values of 3.9 m, 6.4 m, and 6.8 m for plain, hilly, and steep mountainous areas, respectively. [ABSTRACT FROM AUTHOR]

Semi-Supervised Learning (SSL) method has recently caught much attention in the fields of machine learning and computer vision owing to its superiority in classifying abundant unlabelled samples using a few labeled samples. The goal of this paper is to provide an experimental efficiency comparison between graph based SSL algorithms and traditional supervised learning algorithms (e.g., support vector machines) for multispectra] image classification. This research shows that SSL algorithms generally outperform supervised learning algorithms in both classification accuracy and anti-noise ability. In the experiments carried out on two data sets (hyperspectra] image and Landsat image), the mean overall accuracies (OAs) of supervised learning algorithms are 15 percent and 86 percent, while the mean OAs of SSL algorithms are 26 percent and 99 percent. To overcome the polynomial complexity of SSL algorithms, we also developed a linearcomplexity algorithm by employing m ultivaria te Taylor Series Expansion (TSE) and Woodbury Formula. [ABSTRACT FROM AUTHOR]

A rapidly deployable aerial multispectral sensor utilizing four channels in the visible-near-m and one channel in the thermal IR was developed along with processing software to identify oil-on-water and map its spatial extents and thickness distribution patterns. Following validation over natural oil seeps and at Bureau of Safety and Environmental Enforcement's (BSEE'S) Ohmsett test tank, the system was utilized operationally on a near-daily basis for three months during the Deepwater Horizon (MC-252) spill in the Gulf of Mexico in summer 2010. Digital, Gis-compatible analyses were produced and disseminated following each flight mission. The analysis products were utilized for a multitude of response activities including daily offshore oil recovery planning, oil trajectory modeling, dispersant application effect documentation, beached oil mapping and documentation of the relative oil amount along the spill's offshore perimeter. The system's prime limitation was its relatively narrow imaging footprint and low sun angle requirement to minimize sunglint, both of which limited the total area that could be imaged each day. This paper discusses the system's various applications as well as limitations that were encountered during its use in the Deepwater Horizon incident. [ABSTRACT FROM AUTHOR]

An attempt to delineate impervious surfaces in the City of Scottsbluff, Nebraska, was made using multispectral high spatial resolution imagery and lidar data. An ISODATA classification was performed and results aggregated into two parent classes, impervious and pervious. The ISODATA classification yielded an overall accuracy of 91.0 percent with a Kappa of 82.0 percent. A Knowledge Based Expert System (KBE5) set of rules was designed incorporating the imagery classification with lidar data to derive two models, Cover Height and Cover Slope, to provide critical information not available from multispectral imagery. The rules were applied to the initial ISODATA classification to improve the classification accuracy to an overall accuracy of 94.0 percent with a Kappa of 87.9 percent. In this study, it was shown that lidar holds promise for improving the accuracy of impervious surface measurement, as well as the potential identification and measurement of other significant planimetric features such as buildings and trees. [ABSTRACT FROM AUTHOR]

A new algorithm based on the fusion of edge and multispectral information is proposed for the pixel-wise classification Of very high-resolution (VHR) remotely sensed imagery. It integrates the multispectral, spatial and structural information existing in the image. The edge feature is first extracted using an improved multispectral edge detection method, which takes into account the original multispectral bands, the linear NDVI, and the independent spectral components extracted by independent component analysis (ICA). Direction-lines are then defined using the edge and multi-spectral information. Two effective spatial measures are calculated based on the direction-lines in order to describe the contextual information and strengthen the multispectral feature space. Then, the support vector machine (SVM) is employed to classify the hybrid structural-multispectral feature set. In experiments, the proposed spatial measures were compared with the pixel shape index (PSI) and the gray level co-occurrence matrix (GLCM). The experimental results show that the proposed algorithm performs well in terms of classification accuracies and visual interpretation. [ABSTRACT FROM AUTHOR]

While the spatial resolution of remotely sensed data has improved, multispectral imagery is still not sufficient for urban classification. Problems include the difficulty in discriminating between trees and grass, the misclassification of buildings due to diverse roof compositions and shadow effects, and the misclassification of cars on roads. Recently, lidar (light detection and ranging) data have been integrated with remotely sensed data to obtain better classification results. In this study, we first conducted maximum likelihood classification (MLC) experiments, a traditional pixel-based classification method, to identify features suitable for urban classification using lidar data and aerial imagery. The addition of lidar height data improved the overall accuracy by up to 28 and 18 percent, respectively, compared to cases with only red-green-blue (RGB) and multispectral imagery. To further improve classification, we propose a knowledge-based classification system (KBCS) that includes a three-level height, "asphalt road, vegetation, and non-vegetation" (A-V-N) classification rule-based scheme and knowledge-based correction (KBC). The proposed KBCS improved overall accuracy by 12 and 7 percent compared to maximum likelihood and object-based classification, respectively. [ABSTRACT FROM AUTHOR]

The term "image fusion" covers multiple techniques used to combine the geometric detail of a high-resolution panchromatic image and the color information of a low resolution multispectral image to produce a final image with the highest possible spatial information content while still preserving good spectral information quality. During the last twenty years, many methods such as Principal Component Analysis (PcA), Multiplicative Transform, Brovey Transform, and IHS Transform have been developed producing good quality fused images. Despite the quite good visual results, many research papers have reported the limitations of the above fusion techniques. The most significant problem is color distortion. Another common problem is that the fusion quality often depends upon the operator's fusion experience and upon the data set being fused. In this study, we compare the efficiency of nine fusion techniques and more specifically the efficiency of IHS, Modified IHS, PCA, Pansharp, Wavelet, LMM (Local Mean Matching), LMVM (Local Mean and Variance Matching), Brovey, and Multiplicative fusion techniques for the fusion of Quick Bird data. The suitability of these fusion techniques for various applications depends on the spectral and spatial quality of the fused images. In order to quantitatively measure the quality of the fused images, we have made the following controls, First, we have examined the visual qualitative result. Then, we examined the correlation between the original multispectral and the fused images and all the statistical parameters of the histograms of the various frequency bands. Finally, we performed an unsupervised classification, and we compared the resulting images. All the fusion techniques improve the resolution and the visual result. The resampling method practically has no effect on the final visual result. The LMVM, the LMM, the Pansharp, and the Wavelet merging technique do not change the statistical parameters of the original images. The Modified IHS provokes minor changes to the statistical parameters than the classical IHS or than the PCA. After all the controls, the LMVM, the LMM, the Pansharp, and the Modified IHS algorithm seem to gather the more advantages in fusion panchromatic and multispectral data. [ABSTRACT FROM AUTHOR]

Anomaly detection in hyperspectral imagery is gaining increasing interest. However, most covariance matrix-based detectors are applied directly to all the hyperspectral data bands without considering the spectral variation and their possible different contribution to detection. Besides, the limited sample number as compared with the high dimensionality of the data can lead to the imprecise estimation of the covariance matrix and even the singularity problem. In this paper, a band-subset fuzzy integral fusion (BS-FI) detection method is presented to solve these problems. The complete set of hyperspectral data bands is first partitioned into several lower dimensional band-subsets, whose detection results are obtained separately and finally merged by a fuzzy integral fusion method. We adopt a non-parametric fuzzy support function which can utilize more statistical information and avoid the model discrepancy that might be brought in by a fixed distribution model. In addition, the fuzzy density is assigned by the ratio between the target signal and noise, which is the key to the target detection problem through an adaptive eigenvalue-based approach. In the experiments on real OMIS-I hyperspectral imagery, the proposed method outperforms the Rx detector both on the complete set of bands and on each band-subset, and other band-subset fusion detectors. [ABSTRACT FROM AUTHOR]

We compared the capability of the Earth Observing-1 (EQ-1) Hyperion hyperspectral (HS) data with that of the EQ-1 Advanced Land Imager (ALI) multispectral (MS) data for discriminating different land-use and land-cover classes in Fremont, California. We designed a classification scheme of two levels with level I including general classes and level II including more specific classes. Classification shows that the HS data does not produce better results than the MS data when we directly applied a Mahalanobis distance (MD) classifier. We tested a number of feature reduction and extraction algorithms for the HS image. These algorithms include principal component analysis (PCA), segmented PCA (SEGPCA), linear discriminant analysis (LDA), segmented LDA (SEGLDA), penalized discriminant analysis (PDA) and segmented PDA (SEGPDA). Feature reductions were all followed by an MD classifier for image classification. With SEGPDA, SEGLDA, PDA, and LDA, similar accuracies were achieved while a segmentation-based approach we proposed (SEGPDA or SEGLDA) greatly improved computation efficiency. They all outperformed SEGPCA and PCA by 4 to 5 percent (level II) and 1 to 3 percent (level I) in classification accuracy. For level II classification, overall accuracies obtained by using the features extracted from the HS image were 2 to 3 percent greater than those obtained with the MS image. For various vegetation class and impervious land use categories, the HS data consistently produced better results than the MS data. For level I classification, the HS image generated a thematic map that is <0.01 greater in kappa coefficient comparing to the MS image. When we collapsed the level II classification map to a level I map, 5 percent (HS) to 7 percent (MS) improvements were achieved. [ABSTRACT FROM AUTHOR]

In this paper, we present a new approach combining canonical correlation analysis and contextual Bayes decision for change detection in bi-temporal multispectral remotely sensed images. Canonical correlation analysis in the form of Minimum Noise Fraction/Multivariate Alteration Detection transformation was first applied to two multispectral images to effectively yield a difference image, followed by a contextual Bayes decision procedure using automatic thresholding and Markov random field modeling techniques to identify areas where changes may have actually occurred from the difference image. An experiment of monitoring land reclamations in Hong Kong from Landsat TM/ETM+ images was conducted, and the results demonstrated effectiveness of our approach. [ABSTRACT FROM AUTHOR]

This paper puts forward a pixel-based and a region-based method for detecting land-use changes using a multi-temporal pair of satellite KVR panchromatic and aerial Daedalus multispectral images. These methods, based on fuzzy logic, provide a confidence level of their post-classification change detection results. They were applied to support an early stage of mine clearance in Croatia, where mines laid during the latest conflict (1991 to 1995) cause disruption in the land-use evolution. They were used for locating the agricultural land that was abandoned during the war and is therefore considered as a mine contamination indicator. The mere location of the abandoned agricultural land is of limited use for supporting mine clearance. On the other hand, the integration of the location and spatial confidence level can help the mine experts in their decision making by evaluating both unlikelihood and likelihood of a plot of land to be abandoned. [ABSTRACT FROM AUTHOR]

The objective of this research focused on the utility of QuickBird multispectral data for identifying and classifying tropical-marine benthic habitats after applying atmospheric and water-column corrections for an area around Roatan Island, Honduras. Atmospheric (Rayleigh and aerosol path radiance) and water column corrections (water depth and water column attenuation) were applied to the imagery, making it an effective method for mapping benthic habitats. Water depth for each pixel was calculated based on a linear model by regressing transformed radiance over known homogenous benthos against measured depths. Water column correction was achieved by deriving absorption and backscattering coefficients for each band of the image using a 50 × 50 window of clear water pixels. Corrections for water path radiance and water column attenuation of the bottom reflected radiance were made for the entire scene, allowing the bottom albedo to be determined for shallow coastal areas. An image of the bottom (i.e., an albedo image), minus the water column, was produced. Albedos were ≤8 percent for seagrass benthos, approximately 8 to 18 percent for coral areas, and 218 percent for sand dominated areas. An unsupervised classification algorithm was applied to the bottom albedo image, generating a classified map of benthic habitats. Accuracy assessment based on 383 reference points revealed an overall accuracy of 81 percent, with an overall Kappa value of 0.774. [ABSTRACT FROM AUTHOR]

This paper analyses the potential of a high-resolution airborne remote sensing system, the Digital Multi-Spectral Imagery (DMSI), for detecting canola growth variability within a field to help farmers for future incorporation of the system into site-specific crop management approaches for agriculture. Transect sampling within a canola field of a broad acre agricultural property in the South West of Western Australia was conducted synchronous with the capture of one-meter spatial resolution DMSL Four individual bands (blue, green, red, and NIR) and five image transformations namely the Normalized Difference Vegetation Index (NDVI), Normalized Difference Vegetation Index - Green (NDVI-green), Soil Adjusted Vegetation Index (SAVI), Photosynthetic Vigor Ratio (PVR) and Plant Pigment Ratio (PPR) of DMSI were investigated. Canola density was correlated with the four individual bands and five image transformations, while the LAI was correlated with the four individual bands. The NDVI-green, red and near-infrared bands of DMSI produced the best correlations with the density of canola, whereas the LAI had significant (α = 0.05) negative correlations with the blue (-0.93) and red (-0.89) DMSI bands, and a significant positive correlation were found with the near- infrared band (0.82). [ABSTRACT FROM AUTHOR]

The capabilities of high-resolution, multispectral remote sensing imagery to map important stream features is investigated. Eighty centimeter spatial resolution CASI imagery was acquired in eight spectral bands over Tofino Creek on the west coast of Vancouver Island, British Columbia. A spectral angle mapping algorithm was used to classify stream habitat including hydraulic habitat, substrate material, and woody debris. Subclasses were attempted in terms of streambed material and water depth, but results were not reliable. A classification of deep water, moderate depth water, shallow water, sand, gravel and cobble, and woody debris in sunlit conditions, however, proved accurate (80 percent on average). Individual logs and piles of woody debris were consistently detected. Silty substrate in a tidal fiats zone was also classified, but results indicated that different substrate material beneath the water may require separate classes and can result in problematic water depth classification. Patterns of general classes were reasonably represented within shadowed areas cast by isolated trees or groups of trees. However, problems do arise within lengthy shadowed stretches. Some boundaries of stream features with surrounding forest and between some zones of sand, gravel, and cobble were also misclassified. High-resolution, multispectral imagery in four or more bands combined with good geometric correction, image mosaicking, and appropriate automatic classification techniques offer a viable tool for stream mapping to meet a variety of issues and applications. In the future, a powerful suite of stream information may be compiled from multispectral classification combined with high-resolution thermal and lidar data. [ABSTRACT FROM AUTHOR]

Four distinctly different classifiers were used to analyze multi-spectral data. Which of these classifiers is most suitable for a specific study area is not always clear. This paper provides a comparison of minimum-distance classifier (MDC), maximum-likelihood classifier (MLC), extraction and classification of homogeneous objects (ECHO), and decision-tree classifier based on linear spectral mixture analysis (DTC-LSMA). Each of the classifiers used both Landsat Thematic Mapper data and identical field-based training sample datasets in a western Brazilian Amazon study area. Seven land-cover classes— mature forest, advanced secondary succession, initial secondary succession, pasture lands, agricultural lands, bare lands, and water—were classified. Classification results indicate that the DTC-LSMA and ECHO classifiers were more accurate than were the MDC and MLC. The overall accuracy of the DTC-LSMA approach was 86 percent with a 0.82 kappa coefficient and ECHO had an accuracy of 83 percent with a 0.79 kappa coefficient. The accuracy of the other classifiers ranged from 77 to 80 percent with kappa coefficients from 0.72 to 0.75. [ABSTRACT FROM AUTHOR]

The brushtail possum flYichosurus Vulpecula, is a wildlife vector for tuberculosis (TB) caused by Mycobacterium bovis in New Zealand. Supervised automatic classification of a SPOT3 multi spectral image was used to generate a vegetation map. which was used together with slope data to model the risk of TB-infected possums being present in habitat patches. The vegetation data were also used to identify habitat patterns which, together with other geographic variables, were incorporated into logistic regression models to identify predictors of possum TB risk of farms. The impact of the predicted possum TB risk data on the cost-effectiveness of vector control programs at both individual farm and larger regional control areas is discussed, plus issues associated with the uptake of the models by operational managers. [ABSTRACT FROM AUTHOR]

An unmixing method was used to estimate the area of alder trees from multitemporal Thematic Mapper (TM) imagery.of the Kushiro mire, an important wetland area in Japan. Because the spectral radiance (endmember) of alder coverage within a pixel must be precisely known before the unmixing method is applied, we first collected ground truth data for the canopy area of alder using field survey methods, and aerial photographs. The alder endmember was calculated by linking TM imagery and the ground truth data. Alder coverage was estimated more precisely than it was with classification by the maximum-likelihood method. Furthermore, the alder area could be evaluated within a pixel, which cannot be classified as alder by the maximumlikelihood method. [ABSTRACT FROM AUTHOR]

Radiometric fidelity of calibrated airborne video imagery is important if the imagery is to be used for quantitative analysis of target surfaces. A four-channel Multispectral Airborne Video System (MAVS) is used at Charles Sturt University (CSU) for a range of environmental and agricultural monitoring applications. The radiometric distortions in the MAVS imagery could occur because of lens characteristics in the form of vignetting effects and optical aberrations. This paper details the control experiments conducted to detect and quantify such distortions. The vignetting in particular is shown to be creating a non- uniform brightness level across the MAVS imagery. The paper then develops, efficient procedures to correct the vignetting in the MAVS imagery by producing relevant correction coefficients, templates, and equations. The accuracy of these de-vignetting (correction) procedures is shown to be comparable to the ac- curacy of similar corrections reported elsewhere. An investi- gation into the effect of spectrum related vignetting on the MAVS imagery due to spectral filter characteristics found negligible distortions that did not warrant corrections. This particular type of vignetting is usually caused by the wavelength shifts in the band-pass window of a filter at large incident angles. [ABSTRACT FROM AUTHOR]

A collection of spectral indices, derived from a range of remote sensing imagery spatial resolutions, are compared to on-farm measurements of maize chlorophyll content and yield at two trial farms in central Malawi to evaluate what spatial resolutions are most effective for relating multispectral images with crop status. Single and multiple linear regressions were tested for spatial resolutions ranging from 7 cm to 20 m using a small unmanned aerial system (sUAS) and satellite imagery from Planet, SPOT 6, Pléiades, and Sentinel-2. Results suggest that imagery with spatial resolutions nearer the maize plant scale (i.e., 14-27 cm) are most effective for relating spectral signals with crop health on smallholder farms in Malawi. Consistent with other studies, green-band indices were more strongly correlated with maize chlorophyll content and yield than conventional red-band indices, and multivariable models often outperformed single variable models. [ABSTRACT FROM AUTHOR]

The traditional spatial-spectral classification methods applied to hyperspectral remote sensing imagery are conducted by combining the spatial information vector and the spectral information vector in a separate manner, which may cause information loss and concatenation deficiency between the spatial and spectral information. In addition, the traditional morphological-based spatial-spectral classification methods require the design of handcrafted features according to experience, which is far from automatic and lacks generalization ability. To automatically represent the spatial-spectral features around the central pixel within a spatial neighborhood window, a novel spatial-spectral feature classification method based on the unsupervised convolutional sparse auto-encoder (UCSAE) with a window-in-window strategy is proposed in this study. The UCSAE algorithm features a unique spatial-spectral feature extraction approach which is executed in two stages. The first stage represents the spatial-spectral features within a spatial neighborhood window on the basis of spatial-spectral feature extraction of sub-windows with a sparse auto-encoder (SAE). The second stage exploits the spatial-spectral feature representation with a convolution mechanism for the larger outer windows. The UCSAE algorithm was validated by two widely used hyperspectral imagery datasets (the Pavia University dataset and the Washington DC Mall dataset) obtaining accuracies of 90.03 percent and 96.88 percent, respectively, which are better results than those obtained by the traditional hyperspectral spatial-spectral classification approaches. [ABSTRACT FROM AUTHOR]

This research deals with the automatic generation of 2D land cover maps of urban areas using very high resolution multispectral aerial imagery. The appropriate selection of classifier and attributes is important to achieve high thematic accuracies. In this paper, new attributes are generated to increase the discriminative power of auxiliary information provided by remote sensing images. The generated attributes are derived from the vegetation index and elevation information using morphological attribute profiles. The extended experimental evaluation and comparison of attribute profile-based mapping solutions is conducted to derive the optimal combinations of attributes required for classification and to understand the genericity of attributes on a range of classifiers, i.e., various combinations of attributes and classifiers. Experimental results with two high resolution images show that the proposed attributes derived on auxiliary information outperform the existing attribute profiles computed on original image and its principal components. [ABSTRACT FROM AUTHOR]