Spectral Angle Mapper Algorithm Research Articles

Mapping of Asbestos Cement Roofs and Their Weathering Status Using Hyperspectral Aerial Images

The aims of this study were: (i) the mapping of asbestos cement roofs in an urban area; and (ii) the development of a spectral index related to the roof weathering status. Aerial images were collected through the Multispectral Infrared and Visible Imaging Spectrometer (MIVIS) sensor, which acquires data in 102 channels from the visible to the thermal infrared spectral range. An image based supervised classification was performed using the Spectral Angle Mapper (SAM) algorithm. The SAM was trained through a set of pixels selected on roofs of different materials. The map showed an average producer’s accuracy (PA) of 86% and a user’s accuracy (UA) of 89% for the asbestos cement class. A novel spectral index, the “Index of Surface Deterioration” (ISD), was defined based on measurements collected with a portable spectroradiometer on asbestos cement roofs that were characterized by different weathering statuses. The ISD was then calculated on the MIVIS images, allowing the distinction of two weathering classes (i.e., high and low). The asbestos cement map was handled in a Geographic Information System (GIS) in order to supply the municipalities with the cadastral references of each property having an asbestos cement roof. This tool can be purposed for municipalities as an aid to prioritize asbestos removal, based on roof weathering status.

Supervised Classification Approaches to Analyze Hyperspectral Dataset

In this paper, Spectral Angle Mapper (SAM) and Spectral Information Divergence (SID) classification approaches were used to classify hyperspectral image of Georgia, USA, using Environment of Visualizing Images (ENVI). It is a software application used to process and analyze geospatial imagery. Spatial, spectral subset and atmospheric correction have been performed for SAM and SID algorithms. Results showed that classification accuracy using the SAM approach was 72.67%, and SID classification accuracy was 73.12%. Whereas, the accuracy of SID approach is better than SAM approach. Consequently, the two approaches (SID and SAM) have proven to be accurately converged in classification of hyperspectral image of Georgia, USA.

Comparison of support vector machine, artificial neural network, and spectral angle mapper algorithms for crop classification using LISS IV data

The Resourcesat-2 is a highly suitable satellite for crop classification studies with its improved features and capabilities. Data from one of its sensors, the linear imaging and self-scanning (LISS IV), which has a spatial resolution of 5.8 m, was used to compare the relative accuracies achieved by support vector machine (SVM), artificial neural network (ANN), and spectral angle mapper (SAM) algorithms for the classification of various crops and non-crop covering a part of Varanasi district, Uttar Pradesh, India. The separability analysis was performed using a transformed divergence (TD) method between categories to assess the quality of training samples. The outcome of the present study indicates better performance of SVM and ANN algorithms in comparison to SAM for the classification using LISS IV sensor data. The overall accuracies obtained by SVM and ANN were 93.45% and 92.32%, respectively, whereas the lower accuracy of 74.99% was achieved using the SAM algorithm through error matrix analysis. Results derived from SVM, ANN, and SAM classification algorithms were validated with the ground truth information acquired by the field visit on the same day of satellite data acquisition.

Retrieval of soil physicochemical properties towards assessing salt-affected soils using Hyperspectral Data

The relationship between soil salinity parameters and their influence on soil spectral characteristics were analyzed using both satellite data (Hyperion) and reflectance data of soil samples collected from parts of Ahmedabad district of Gujarat, India. The soil spectral reflectance curves were assessed using absorption feature parameters by DISPEC software to identify suitable spectral band for salinity characterization. The Hyperion data of the study area were processed and classified into different classes by spectral angle mapper algorithm using spectral library generated from soil spectra. The results showed that among all the observed soil parameters Electrical Conductivity, Exchangeable Sodium Percentage, Cation Exchange Capacity and Mg++ predictions can be made accurately based on partial least square regression models developed from selected wavelengths. Out of the total study area moderately saline-sodic, severely saline-sodic, severely saline and slightly saline soils occupy 23.5, 12.6, 10.9 and 0.04%, respectively.

Mapping Agewise Discrimination of Arecanut Crop Water Requirement Using Hyperspectral Remote Sensing

Arecanut is one of the major plantation crops in India. The crop is grown extensively in many regions of the country due to its high commercial value. A part of Channageri taluk of Davanagere district, Karnataka, known as “the land of arecanut” is chosen for the study. The majority of the study area consists of arecanut plantations of various age groups. Growth stage is one of the major factors that govern the crop water requirement (CWR), which in turn influences irrigation planning and scheduling. Accurate estimation of crop area, area under different age groups and area under stress are extremely important to estimate CWR. Traditional methods of area estimation are costly and time consuming. With the advancement in hyperspectral remote sensing, within crop variations can be discriminated with enhanced reliability and accuracy. Using ASD FieldSpec Pro® handheld spectroradiometer spectral signatures of different age groups and stressed arecanut crops are collected to build spectral library. Using the spectral library and Hyperion imagery, arecanut crops of various age groups and under stress (crown choke disease) are mapped by Spectral Angle Mapper (SAM) algorithm with 73.68% accuracy. NDVI based approach is adopted to determine crop coefficients, and Hargreaves & Samani equation to compute reference crop evapotranspiration to assess CWR. In the study area 784.80 hectares of land is under arecanut crop cultivation with daily irrigation requirement estimated to be 28056.09 m3. From the analysis, the average CWR is observed to vary from 19 to 23 litres/day for juvenile to adult arecanut crop. Another interesting observation is that though 16.9% of the existing crop area is stressed due to crown choke disease, the water consumption of it is equivalent to that of healthy crops. Immediate response to address this issue will lead to enhanced plant life and thereby yield.

Jeffries Matusita-Spectral Angle Mapper (JM-SAM) spectral matching for species level mapping at Bhitarkanika, Muthupet and Pichavaram mangroves

Abstract. This paper proposes a novel hyperspectral matching algorithm by integrating the stochastic Jeffries-Matusita measure (JM) and the deterministic Spectral Angle Mapper (SAM), to accurately map the species and the associated landcover types of the mangroves of east coast of India using hyperspectral satellite images. The JM-SAM algorithm signifies the combination of a qualitative distance measure (JM) and a quantitative angle measure (SAM). The spectral capabilities of both the measures are orthogonally projected using the tangent and sine functions to result in the combined algorithm. The developed JM-SAM algorithm is implemented to discriminate the mangrove species and the landcover classes of Pichavaram (Tamil Nadu), Muthupet (Tamil Nadu) and Bhitarkanika (Odisha) mangrove forests along the Eastern Indian coast using the Hyperion image dat asets that contain 242 bands. The developed algorithm is extended in a supervised framework for accurate classification of the Hyperion image. The pixel-level matching performance of the developed algorithm is assessed by the Relative Spectral Discriminatory Probability (RSDPB) and Relative Spectral Discriminatory Entropy (RSDE) measures. From the values of RSDPB and RSDE, it is inferred that hybrid JM-SAM matching measure results in improved discriminability of the mangrove species and the associated landcover types than the individual SAM and JM algorithms. This performance is reflected in the classification accuracies of species and landcover map of Pichavaram mangrove ecosystem. Thus, the JM-SAM (TAN) matching algorithm yielded an accuracy better than SAM and JM measures at an average difference of 13.49 %, 7.21 % respectively, followed by JM-SAM (SIN) at 12.06%, 5.78% respectively. Similarly, in the case of Muthupet, JM-SAM (TAN) yielded an increased accuracy than SAM and JM measures at an average difference of 12.5 %, 9.72 % respectively, followed by JM-SAM (SIN) at 8.34 %, 5.55% respectively. For Bhitarkanika, the combined JM-SAM (TAN) and (SIN) measures improved the performance of individual SAM by (16.1 %, 15%) and of JM by (10.3%, 9.2%) respectively.

Remote Sensing of Submerged Aquatic Vegetation in a Shallow Non-Turbid River Using an Unmanned Aerial Vehicle

A passive method for remote sensing of the nuisance green algae Cladophora glomerata in rivers is presented using an unmanned aerial vehicle (UAV). Included are methods for UAV operation, lens distortion correction, image georeferencing, and spectral analysis to support algal cover mapping. Eighteen aerial photography missions were conducted over the summer of 2013 using an off-the-shelf UAV and three-band, wide-angle, red, green, and blue (RGB) digital camera sensor. Images were post-processed, mosaicked, and georeferenced so automated classification and mapping could be completed. An adaptive cosine estimator (ACE) and spectral angle mapper (SAM) algorithm were used to complete the algal identification. Digital analysis of optical imagery correctly identified filamentous algae and background coverage 90% and 92% of the time, and tau coefficients were 0.82 and 0.84 for ACE and SAM, respectively. Thereafter, algal cover was characterized for a one-kilometer channel segment during each of the 18 UAV flights. Percent cover ranged from <5% to >50%, and increased immediately after vernal freshet, peaked in midsummer, and declined in the fall. Results indicate that optical remote sensing with UAV holds promise for completing spatially precise, and multi-temporal measurements of algae or submerged aquatic vegetation in shallow rivers with low turbidity and good optical transmission.

Hyperspectral analysis of clay minerals

Abstract. A study was carried out by collecting soil samples from parts of Gwalior and Shivpuri district, Madhya Pradesh in order to assess the dominant clay mineral of these soils using hyperspectral data, as 0.4 to 2.5 μm spectral range provides abundant and unique information about many important earth-surface minerals. Understanding the spectral response along with the soil chemical properties can provide important clues for retrieval of mineralogical soil properties. The soil samples were collected based on stratified random sampling approach and dominant clay minerals were identified through XRD analysis. The absorption feature parameters like depth, width, area and asymmetry of the absorption peaks were derived from spectral profile of soil samples through DISPEC tool. The derived absorption feature parameters were used as inputs for modelling the dominant soil clay mineral present in the unknown samples using Random forest approach which resulted in kappa accuracy of 0.795. Besides, an attempt was made to classify the Hyperion data using Spectral Angle Mapper (SAM) algorithm with an overall accuracy of 68.43 %. Results showed that kaolinite was the dominant mineral present in the soils followed by montmorillonite in the study area.

Mapping Coral Reef Benthos, Substrates, and Bathymetry, Using Compact Airborne Spectrographic Imager (CASI) Data

This study used a reef-up approach to map coral reef benthos, substrates and bathymetry, with high spatial resolution hyperspectral image data. It investigated a physics-based inversion method for mapping coral reef benthos and substrates using readily available software: Hydrolight and ENVI. Compact Airborne Spectrographic Imager (CASI) data were acquired over Heron Reef in July 2002. The spectral reflectance of coral reef benthos and substrate types were measured in-situ, and using the HydroLight 4.2 radiative transfer model a spectral reflectance library of subsurface reflectance was simulated using water column depths from 0.5–10.0 m at 0.5 m intervals. Using the Spectral Angle Mapper algorithm, sediment, benthic micro-algae, algal turf, crustose coralline algae, macro-algae, and live coral were mapped with an overall accuracy of 65% to a depth of around 8.0 m; in waters deeper than 8.0 m the match between the classified image and field validation data was poor. Qualitative validation of the maps showed accurate mapping of areas dominated by sediment, benthic micro-algae, algal turf, live coral, and macro-algae. A bathymetric map was produced for water column depths 0.5–10.0 m, at 0.5 m intervals, and showed high correspondence with in-situ sonar data (R2 value of 0.93).

Methyl green and nitrotetrazolium blue chloride co-expression in colon tissue: A hyperspectral microscopic imaging analysis

Abstract Histological observation of dual-stained colon sections is usually performed by visual observation under a light microscope, or by viewing on a computer screen with the assistance of image processing software in both research and clinical settings. These traditional methods are usually not sufficient to reliably differentiate spatially overlapping chromogens generated by different dyes. Hyperspectral microscopic imaging technology offers a solution for these constraints as the hyperspectral microscopic images contain information that allows differentiation between spatially co-located chromogens with similar but different spectra. In this paper, a hyperspectral microscopic imaging (HMI) system is used to identify methyl green and nitrotetrazolium blue chloride in dual-stained colon sections. Hyperspectral microscopic images are captured and the normalized score algorithm is proposed to identify the stains and generate the co-expression results. Experimental results show that the proposed normalized score algorithm can generate more accurate co-localization results than the spectral angle mapper algorithm. The hyperspectral microscopic imaging technology can enhance the visualization of dual-stained colon sections, improve the contrast and legibility of each stain using their spectral signatures, which is helpful for pathologist performing histological analyses.

Jeffries Matusita based mixed-measure for improved spectral matching in hyperspectral image analysis

This paper proposes a novel hyperspectral matching technique by integrating the Jeffries-Matusita measure (JM) and the Spectral Angle Mapper (SAM) algorithm. The deterministic Spectral Angle Mapper and stochastic Jeffries-Matusita measure are orthogonally projected using the sine and tangent functions to increase their spectral ability. The developed JM-SAM algorithm is implemented in effectively discriminating the landcover classes and cover types in the hyperspectral images acquired by PROBA/CHRIS and EO-1 Hyperion sensors. The reference spectra for different land-cover classes were derived from each of these images. The performance of the proposed measure is compared with the performance of the individual SAM and JM approaches. From the values of the relative spectral discriminatory probability (RSDPB) and relative discriminatory entropy value (RSDE), it is inferred that the hybrid JM-SAM approach results in a high spectral discriminability than the SAM and JM measures. Besides, the use of the improved JM-SAM algorithm for supervised classification of the images results in 92.9% and 91.47% accuracy compared to 73.13%, 79.41%, and 85.69% of minimum-distance, SAM and JM measures.It is also inferred that the increased spectral discriminability of JM-SAM measure is contributed by the JM distance. Further, it is seen that the proposed JM-SAM measure is compatible with varying spectral resolutions of PROBA/CHRIS (62 bands) and Hyperion (242 bands).

The reconnaissance of mineral resources through aster data-based image processing, interpreting and ground inspection in the Jiafushaersu area, West Junggar, China

Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data covering the Jiafushaersu area in Xinjiang were used for mapping lithology and hydrothermal alteration. The study area situates at a potential mineralization zone in relation to small hypabyssal granodiorite or quartz monzonite intrusions along the margin of granitoid batholiths of Darbut foot wall. The false colour composition of bands 521 and the first three principal component analyses (PCA1, PCA2, PCA3) in RGB identify the lithological units and discriminate the small intrusions very well from the adjacent granitoid batholiths. PCA and spectral angle mapper (SAM) algorithm were employed to discriminate alteration minerals. The results indicate that the hydroxyl-bearing or ferric and less commonly carbonate types show good correlation with the quartz monzonite porphyry and aplite. Field verification led to finding of the Jiafushaersu molybdenum mineralization. The lithological and geochemical features imply that the molybdenum mineralization is close to the porphyry type. This study further verified that the foot wall of the Darbut suture could have served as a more important metallogenic district for the porphyry copper and molybdenum deposits. It is concluded that the ASTER data-based methods can be used as a powerful tool for small intrusion-type mineral resources targeting.

Detection and mapping vegetation cover based on the Spectral Angle Mapper algorithm using NOAA AVHRR data

Satellite data, taken from the National Oceanic and Atmospheric Administration (NOAA) have been proposed and used for the detection and the cartography of vegetation cover in North Africa. The data used were acquired at the Analysis and Application of Radiation Laboratory (LAAR) from the Advanced Very High Resolution Radiometer (AVHRR) sensor of 1km spatial resolution. The Spectral Angle Mapper Algorithm (SAM) is used for the classification of many studies using high resolution satellite data. In the present paper, we propose to apply the SAM algorithm to the moderate resolution of the NOAA AVHRR sensor data for classifying the vegetation cover. This study allows also exploiting other classification methods for the low resolution. First, the normalized difference vegetation index (NDVI) is extracted from two channels 1 and 2 of the AVHRR sensor. In order to obtain an initial density representation of vegetal formation distribution, a methodology, based on the combination between the threshold method and the decision tree, is used. This combination is carried out due to the lack of accurate data related to the thresholds that delimit each class. In a second time, and based on spectral behavior, a vegetation cover map is developed using SAM algorithm. Finally, with the use of low resolution satellite images (NOAA AVHRR) and with only two channels, it is possible to identify the most dominant species in North Africa such as: forests of the Liege oaks, other forests, cereal’s cultivation, steppes and bar soil.

Compositional diversity of near-, far-side transitory zone around Naonobu, Webb and Sinus Successus craters: Inferences from Chandrayaan-1 Moon Mineralogy Mapper (M3) data

This study investigated the potential of Moon Mineralogy Mapper (M3) data for studying compositional variation in the near-, far-side transition zone of the lunar surface. For this purpose, the radiance values of the M3 data were corrected for illumination and emission related effects and converted to apparent reflectance. Dimensionality of the calibrated reflectance image cube was reduced using Independent Component Analysis (ICA) and endmembers were extracted by using Pixel Purity Index (PPI) algorithm. The selected endmembers were linearly unmixed and resolved for mineralogy using United States Geological Survey (USGS) library spectra of minerals. These mineralogically resolved endmembers were used to map the compositional variability within, and outside craters using Spectral Angle Mapper (SAM) algorithm. Cross validation for certain litho types was attempted using band ratios like Optical Maturity (OMAT), Color Ratio Composite and Integrated Band Depth ratio (IBD). The identified lithologies for highland and basin areas match well with published works and strongly support depth related magmatic differentiation. Prevalence of pigeonite-basalt, pigeonite-norite and pyroxenite in crater peaks and floors are unique to the investigated area and are attributed to local, lateral compositional variability in magma composition due to pressure, temperature, and rate of cooling.

Observation of Atherosclerotic Plaque Phantoms through Saline or Blood Layers by Near-Infrared Hyperspectral Imaging

We observed atherosclerotic plaque phantoms using a novel near-infrared (NIR) hyperspectral imaging (HSI) technique. Data were obtained through saline or blood layers to simulate an angioscopic environment for the phantom. For the study, we developed a NIR-HSI system with an NIR supercontinuum light source and mercury-cadmium-telluride camera. Apparent spectral absorbance was obtained at wavelengths of 1150 - 2400 nm. Hyperspectral images of lipid were constructed using a spectral angle mapper algorithm. Bovine fat covered with saline or blood was observed using hyperspectral images at a wavelength around 1200 nm. Our results show that NIR-HSI is a promising angioscopic technique with the potential to identify lipid-rich plaques without clamping and saline injection.

Using Hyperspectral Data to Identify Crops in a Cultivated Agricultural Landscape - A Case Study of Taita Hills, Kenya

Recent advances in hyperspectral remote sensing techniques and technologies allow us to more accurately identify larger range of crop species from airborne measurements. This study employs hyperspectral AISA Eagle VNIR imagery acquired with 9 nm spectral and 0.6 m spatial resolutions over a spectral range of 400 nm to 1000 nm. The area of study is the Taita hills in Kenya. Various crops are grown in this region basically for food and as an economic activity. The crops addressed are: maize, bananas, avocados, and sugarcane and mango trees. The main objectives of this study were to study what crop species can be distinguished from the cultivated population crops in the agricultural landscape and what feature space discriminates most effectively the spectral signatures of different species. Spectral Angle Mapper (SAM) algorithm together with some dissimilarity concepts was applied in this work. The spectral signatures for crops were collected using accurate field plot maps. Accuracy assessment was done using independent training vector data. We achieved an overall accuracy of 77% with a kappa value of 0.67. Various crops in different locations were identified and shown.

Characterization of the top sediment layer in coastal intertidal mudflats from medium-to-coarse resolution satellite imagery and field measurements

Intertidal sediments are critically important in controlling intertidal mudflat microphytobenthic primary productivity and the functioning of intertidal ecosystems. This paper demonstrates the possibility of deriving different intertidal sediment properties from coarse-to-medium resolution remote sensing imagery. Supervised and image based classification methods were used to map different substrate types based on the Spectral Angle Mapper (SAM) algorithm. The algorithm characterized different sediment properties from remote sensing data based on field collected and image-extracted endmembers. The results demonstrate that, different substrate types can be derived from coarse-to-medium resolution images using SAM algorithm. Supervised and image-based classification methods performed well in deriving intertidal sediment properties. From the results, sand sediments cover a wide area in extent than clay whereas Normalized Difference Vegetation Index (NDVI) validation results indicate that, clay sediments have higher NDVI values as compared to sand sediments. We conclude that, intertidal sediment properties can be successfully derived from coarse-to-medium resolution satellite imagery. Key words: Endmember, microphytobenthos, spectral signature, substrates, trios ramses, wadden sea.

A Novel Classification Method for Multispectral Imaging Combined with Portable Raman Spectroscopy for the Analysis of a Painting by Vincent Van Gogh

In this work, a novel combination of portable micro-Raman spectroscopy and semi-automatic methods of data treatment are proposed for the classification and mapping of visible multispectral imaging data for the analysis of a painting on paper by Vincent Van Gogh. Analysis of multispectral imaging data with the sequential maximum-angle convex cone (SMACC) and spectral angle mapper (SAM) algorithms differentiated the surface into areas on the basis of the presence of pigment mixtures. Complementary analytical information was obtained through portable Raman spectroscopy was performed on a few selected points of the painting, allowing for the determination of Van Gogh's palette and the mapping of pigment mixtures on the painting's surface; the number of mixtures employed is varied and at least two different blues are present. The results obtained were integrated with the information from prior ultraviolet (UV)-induced luminescence analysis performed on the same painting to better understand the materials used by the artist. The mathematical treatment of multispectral data using the proposed methods could be extended to the analysis of other painted surfaces.

Comparison of ASTER thermal data sets in lithological mapping at a volcano-sedimentary basin: a case study from southeastern Iran

The thermal bands of ASTER level 1B (L1B) and ASTER 05 (AST_05) data sets were used to map lithological units of an area situated in the southeastern Iranian volcano-sedimentary complex, Kerman Province. The dominant lithological units exposed at the area include felsic and mafic igneous rocks and carbonate masses. The thermal bands of L1B were atmospherically corrected and converted into emissivity using the Reference Channel, Emissivity Normalization, and Alpha Residual algorithms. The Spectral Angle Mapper (SAM) algorithm was implemented on the calibrated L1B and AST_05 data sets. The reference spectra were derived from the image using a spectral endmember selection procedure including minimum noise fraction (MNF), pixel purity index (PPI), and n-dimensional visualizer. Results were compared to the geological map of the study area to determine the accuracy of lithological mapping. The SAM algorithm showed reliable results for discriminating felsic, intermediate to mafic, carbonate, and quartz-sandstone units. AST_05, Reference Channel, and Emissivity Normalization showed approximately similar results with overall accuracy 83, 80.2 and 79.20, respectively. The Alpha Residual data set is less effective in identification of lithological units, as evidenced by a low overall classification accuracy of 50.5. It is concluded that AST_05, Reference Channel, and Emissivity Normalization data sets are suitable for lithological mapping by ASTER data.

Spectral delineation of albite zone using ASTER data in Khetri Copper Belt

The Khetri Copper Belt is well known because of its economically viable Cu deposits. In this region, there is a linear narrow zone following a major crustal fracture. Along this zone, due to hydrothermal activity and subsequent alteration, a new rock, albitite has formed. The study presents an account of the spectral analysis of Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) dataset for albitite mapping in the Khetri Copper Belt. Spectral angle mapper algorithm is applied on the ASTER data taking the ground sample reflectance spectra (VNIR–SWIR) and image derived spectra (TIR) as reference spectra. The albitite-classified areas are analyzed spatially and statistically to determine the extent of albitite zone in the study area. This research is to contribute to the better understanding of the spatial extent of the albitite zone which will assist in determining the potential for the mineralization in the area. The remote sensing approach of utilizing multichannel data offers the capability to resolve the key geological features of this geologically dynamic area.