Jeffries-Matusita Distance Research Articles

Wetland mapping based on feature optimization and Google Earth Engine in Zhalong Nature Reserve

Wetlands have very important and irreplaceable ecological service functions in protecting biodiversity and maintaining ecological balance. Due to the unique hydrological characteristics of wetlands, large-scale wetland mapping methods are usually time-consuming, labor intensive, and expensive. With the rise of big data and cloud computing platforms, there are opportunities for long time series and large-scale spatial data processing that improve large-scale wetland mapping. In this study, the Sentinel-1, Sentinel-2, and digital elevation model derived from the Google Earth Engine were used to extract the wetland extent in the Zhalong Nature Reserve based on the Jeffries–Matusita (JM) distance for feature optimization and random forest. Research shows the following: (1) The JM distance for each feature showed that the optical features had the highest separation in all five features and (2) the classification after feature optimization performed best. User accuracy (UA) and producer accuracy (PA) for wetland reached 89.64% and 83.00%, respectively, and overall accuracy reached 90.76%. After feature optimization, PA and UA of wetland were each increased 4.26% and 1.07%, respectively, and the number of features was reduced from 46 to 28. The method used in this study is highly accurate in wetland mapping and can effectively avoid the problem of data redundancy.

Identification of soybean based on Sentinel-1/2 SAR and MSI imagery under a complex planting structure

Soybeans, as the major global oil seed crops, are among the large-scale agricultural products imported into China. In this way, the accurate identification of their growing areas is the basis for agricultural decision-making and planting structure adjustment, which are of great significance to the national food security. Currently, there is insufficient research on the identification of soybean, using remote sensing, under complex agricultural planting structures. In this paper, Guoyang County, a typical soybean-producing region in the north of Anhui Province, China, was selected as the study area, and then the multi-temporal Sentinel-1/2 (S-1/2) microwave and optical multispectral data were integrated to obtain the spatial distribution of the planting areas, via a stepwise hierarchical extraction strategy. Considering the Jeffries-Matusita (JM) distance, August 18 (i.e., the early pod setting stage) was correspondingly determined as more appropriate for soybean extraction, compared with other time phases. For this purpose, a set of rules were established to eliminate the non-cropland pixels, and the total distribution of the field vegetation was derived accordingly. In total, 30 candidate features (viz. 10 spectral bands, 11 vegetation indices, 4 texture features, and 5 microwave polarization features) were selected, and their importance was evaluated based on the ReliefF algorithm. With the support of the typical cover type samples, the weight assessment method for the ReliefF feature was further bundled with three machine-learning (ML) algorithms, namely, the random forest (RF), the back-propagation neural network (BPNN), and the support vector machine (SVM), to single out the optimum subset of the features for soybean identification. The results finally revealed that the ReliefF-RF model outperformed the other two. With nine optimum features, Kappa coefficient also reached up to 0.77–0.85. Furthermore, the extraction accuracy produced by the optimum subset of the features was significantly greater than the original 10 spectral bands (Kappa coefficient of 0.75–0.84), albeit slightly lower than the effect of the total 30 features (Kappa coefficient of 0.78–0.85).

An Analysis of the Optimal Features for Sentinel-1 Oil Spill Datasets Based on an Improved J–M/K-Means Algorithm

With the rapid development of world shipping, oil spill accidents such as tanker collisions, illegal sewage discharges, and oil pipeline ruptures occur frequently. As the SAR system expands from single polarization to multipolarization, the Polarmetric Synthetic Aperture Radar (Pol-SAR) system has been widely used in marine oil spill detection. However, in the studies of the oil spill extraction in SAR images, there are some problems that limit large-scale oil spill detection work. As a transition from single-polarized to full-polarized, the dual-polarized system carries some polarization information and can be obtained in large quantities for free, which has become a major breakthrough in solving the problem of large-scale oil spill detection. In order to optimize the multisource features that can be extracted from dual-polarized SAR images, greatly improve the utilization rate of dual-polarized SAR oil spill images under the premise of reducing workload, and ensure the accuracy of marine oil spill extraction, this paper adopts the metric of inter-class separability, the Jeffries–Matusita distance, which improves on the traditional K-means algorithm by focusing on the noise sensitivity defect of the K-means algorithm; the artificial influence of J–M distance in measuring the separability between classes improves the algorithm in three aspects: sample selection, distance calculation, and data evaluation. Finally, using the inter-sample J–M distance of multisource features, the overall accuracy of image segmentation, the F1-score, and the results of correlation analysis between features, three advantageous features and three subdominant features are selected that can be used for marine oil spill detection.

Identification of vegetation under natural gas leakage by spectral index based on feature selection

ABSTRACT The leakage of natural gas storage has a significant impact on economy, personal safety and natural environment. When the leakage is slight, the effect of direct detection is not ideal. Hyperspectral remote sensing can detect it indirectly through the spectral changes of surface vegetation. In this study, wheat, bean and grass were used as surface experimental objects to analyse the variation characteristics of canopy spectrum and physiological and biochemical parameters of vegetation under natural gas leakage. The results showed that with the increase of natural gas concentration in the soil, the spectral reflectance of vegetation increased significantly in the visible region, and decreased significantly in the near infrared region. The natural gas identification index (NGII) was constructed according the optimal weight index screened by Relief-F algorithm. The quantitative test by Jeffries-Matusita (JM) distance showed that NGII can identify (JM > 1.8) stressed vegetation under natural gas leakage in a short time. This study can provide technical reference for detecting leakage of underground natural gas storage.

Identification of Crop Type Based on C-AENN Using Time Series Sentinel-1A SAR Data

Crop type identification is the initial stage and an important part of the agricultural monitoring system. It is well known that synthetic aperture radar (SAR) Sentinel-1A imagery provides a reliable data source for crop type identification. However, a single-temporal SAR image does not contain enough features, and the unique physical characteristics of radar images are relatively lacking, which limits its potential in crop mapping. In addition, current methods may not be applicable for time-series SAR data. To address the above issues, a new crop type identification method was proposed. Specifically, a farmland mask was firstly generated by the object Markov random field (OMRF) model to remove the interference of non-farmland factors. Then, the features of the standard backscatter coefficient, Sigma-naught (σ0), and the normalized backscatter coefficient by the incident angle, Gamma-naught (γ0), were extracted for each type of crop, and the optimal feature combination was found from time-series SAR images by means of Jeffries-Matusita (J-M) distance analysis. Finally, to make efficient utilization of optimal multi-temporal feature combination, a new network, the convolutional-autoencoder neural network (C-AENN), was developed for the crop type identification task. In order to prove the effectiveness of the method, several classical machine learning methods such as support vector machine (SVM), random forest (RF), etc., and deep learning methods such as one dimensional convolutional neural network (1D-CNN) and stacked auto-encoder (SAE), etc., were used for comparison. In terms of quantitative assessment, the proposed method achieved the highest accuracy, with a macro-F1 score of 0.9825, an overall accuracy (OA) score of 0.9794, and a Kappa coefficient (Kappa) score of 0.9705. In terms of qualitative assessment, four typical regions were chosen for intuitive comparison with the sample maps, and the identification result covering the study area was compared with a contemporaneous optical image, which indicated the high accuracy of the proposed method. In short, this study enables the effective identification of crop types, which demonstrates the importance of multi-temporal radar images in feature combination and the necessity of deep learning networks to extract complex features.

An effective feature subset selection approach based on Jeffries-Matusita distance for multiclass problems

Jeffries-Matusita (JM) distance, a transformation of the Bhattacharyya distance, is a widely used measure of the spectral separability distance between the two class density functions and is generally used as a class separability measure. It can be considered to have good potential to be used for evaluation of the effectiveness of a feature in discriminating two classes. The capability of JM distance as a ranking based feature selection technique for binary classification problems has been verified in some research works as well as in our earlier work. It was found by our simulation experiments with benchmark data sets that JM distance works equally well compared to other popular feature ranking methods based on mutual information, information gain or Relief. Extension of JM distance measure for feature ranking in multiclass problems has also been reported in the literature. But all of them are basically rank based approaches which deliver the ranking of the features and do not automatically produce the final optimal feature subset. In this work, a novel heuristic approach for finding out the optimum feature subset from JM distance based ranked feature lists for multiclass problems have been developed without explicitly using any specific search technique. The proposed approach integrates the extension of JM measure for multiclass problems and the selection of the final optimal feature subset in a unified process. The performance of the proposed algorithm has been evaluated by simulation experiments with benchmark data sets in comparison with two other previously developed multiclass JM distance measures (weighted average JM distance and another multiclass extension equivalent to Bhattacharyya bound) and some other popular filter based feature ranking algorithms. It is found that the proposed algorithm performs better in terms of classification accuracy, F-measure, AUC with a reduced set of features and computational cost.

The Accuracy of Winter Wheat Identification at Different Growth Stages Using Remote Sensing

The aim of this study was to explore the differences in the accuracy of winter wheat identification using remote sensing data at different growth stages using the same methods. Part of northern Henan Province, China was taken as the study area, and the winter wheat growth cycle was divided into five periods (seeding-tillering, overwintering, reviving, jointing-heading, and flowering-maturing) based on monitoring data obtained from agrometeorological stations. With the help of the Google Earth Engine (GEE) platform, the separability between winter wheat and other land cover types was analyzed and compared using the Jeffries-Matusita (J-M) distance method. Spectral features, vegetation index, water index, building index, texture features, and terrain features were generated from Sentinel-2 remote sensing images at different growth periods, and then were used to establish a random forest classification and extraction model. A deep U-Net semantic segmentation model based on the red, green, blue, and near-infrared bands of Sentinel-2 imagery was also established. By combining models with field data, the identification of winter wheat was carried out and the difference between the accuracy of the identification in the five growth periods was analyzed. The experimental results show that, using the random forest classification method, the best separability between winter wheat and the other land cover types was achieved during the jointing-heading period: the overall identification accuracy for the winter wheat was then highest at 96.90% and the kappa coefficient was 0.96. Using the deep-learning classification method, it was also found that the semantic segmentation accuracy of winter wheat and the model performance were best during the jointing-heading period: a precision, recall, F1 score, accuracy, and IoU of 0.94, 0.93, 0.93, and 0.88, respectively, were achieved for this period. Based on municipal statistical data for winter wheat, the accuracy of the extraction of the winter wheat area using the two methods was 96.72% and 88.44%, respectively. Both methods show that the jointing-heading period is the best period for identifying winter wheat using remote sensing and that the identification made during this period is reliable. The results of this study provide a scientific basis for accurately obtaining the area planted with winter wheat and for further studies into winter wheat growth monitoring and yield estimation.

ASI: An artificial surface Index for Landsat 8 imagery

Remotely sensing the spatial extent of artificial surfaces is essential for understanding human footprints and evaluating anthropogenic impacts on the global environment and climate. However, since the spectral signatures of artificial surfaces are similar to other bright natural surfaces, it is still a difficult task to separate artificial surfaces from other natural surfaces. Therefore, this study proposed a new Artificial Surface Index (ASI) to extract artificial surfaces based on the multispectral Landsat 8 imagery. We examined the performance of ASI in eight study areas with a variety of landscapes, including desert, coastal, inland urban and rural areas dominated by different climates and land cover categories. We also compared ASI with seven state-of-the-art built-up area or impervious surface indices that are developed for similar purposes. The qualitative and quantitative evaluations demonstrate that ASI has advantages in suppressing non-artificial surfaces with similar spectral signatures to artificial surfaces, such as deserts, bare soil, and harvested croplands. Furthermore, ASI improves the separability between artificial and non-artificial surfaces by 2% to 75% (measured by Jeffries–Matusita Distance), 13% to 164% (measured by Transformed Divergence), and 3% to 131% (measured by Spectral Discrimination Index) compared to the other seven indices across the eight study areas. The newly proposed ASI index has the potential to support a variety of human-activity-related studies, such as change detection of artificial surfaces, urban planning, land resource management at local, continental, and even global scales.

Mapping invasive strawberry guava (Psidium cattleianum) in tropical forests of Mauritius with Sentinel-2 and machine learning

ABSTRACT Invasive flora alien species (IAS) pose a serious threat to biodiversity in tropical forests. This study evaluated different classification approaches for remote sensing of the target IAS, strawberry guava (Psidium cattleianum), using open-access satellite imagery to aid research and understanding of IAS impacts in tropical forests. This is likely the first study using freely available open-access imagery to map strawberry guava, and the first mapping of invasive species in Mauritius using remote sensing. Ground-based observations in Black River Gorges National Park (BRGNP), Mauritius, resulted in 4670 ‘ground truth’ samples across three land cover classes representing different levels of strawberry guava canopy cover; 70% of these ground truth data were used for training and model tuning, and 30% were held-out for testing. Classification was performed using Sentinel-2 MSI images and Google Earth Engine. Computation of Jeffries-Matusita distance supported the addition of Grey Level Co-Occurrence Matrix texture measures, and spectral indices (ARI1, and ReNDVI) for classification, since these features increased separability of strawberry guava cover classes substantially over using spectral bands alone. Higher than 80% overall and individual class accuracies were achieved with both Random Forest (RF) and Support Vector Machine (SVM) classification of strawberry guava in BRGNP. RF is recommended for similar future applications for its higher overall accuracy (97.60% ± .20% at 95% confidence), lower variation, and prediction of more homogenous class regions. This study found that strawberry guava canopy cover in BRGNP was highest in central regions and correlated to steeper slopes, with little overall change from 2016 to 2020.

Identifying plants under natural gas micro-leakage stress using hyperspectral remote sensing

Natural gas is an important clean energy source. The demand for, and consumption of, natural gas have been increasing in recent years. Slight natural gas leakage can occur during transportation, which can have a negative impact on the environment, economy, and safety. However, it is relatively difficult to directly detect natural gas microleakage. Hyperspectral remote sensing technology is useful for analyzing the spectral characteristics of vegetation near leakage areas, thereby indirectly obtaining leakage information. In this study, a field experiment was designed to simulate natural gas leakage from an underground pipeline and gas stress on three plant species. The canopy spectral reflectance of the vegetation throughout the growth period of the plants was collected and analyzed. Variational mode decomposition was then used to decompose the spectra. Based on the stress distance (SD) and intrinsic mode functions, it was found that the second intrinsic mode function, with a decomposition scale of 32, was sensitive to gas stress. According to the results of SD, the bands (616 and 829 nm) sensitive to natural gas stress for the three plant species were extracted, and the variational mode decomposition index (VMDI) was constructed. The Jeffries–Matusita distance (JMD) was used to quantitatively evaluate the VMDI index and three indices were used to evaluate the ability to recognize stress. It was found that the index proposed in this study could identify stressed wheat and grass one week earlier than other indices and could better identify stressed vegetation throughout the phenological cycle (JMD > 1.8). The results show that the proposed index can be used as a reliable method to identify natural gas-stressed plants, and that hyperspectral technology is promising for detecting the location of natural gas leaks from underground pipelines.

Identifying the Spectral Signatures of Invasive and Native Plant Species in Two Protected Areas of Pakistan through Field Spectroscopy

Globally, biological invasions are considered as one of the major contributing factors for the loss of indigenous biological diversity. Hyperspectral remote sensing plays an important role in the detection and mapping of invasive plant species. The main objective of this study was to discriminate invasive plant species from adjacent native species using a ground-based hyperspectral sensor in two protected areas, Lehri Reserve Forest and Jindi Reserve Forest in Punjab, Pakistan. Field spectral measurements were collected using an ASD FieldSpec handheld2TM spectroradiometer (325–1075 nm) and the discrimination between native and invasive plant species was evaluated statistically using hyperspectral indices as well as leaf wavelength spectra. Finally, spectral separability was calculated using Jeffries Matusita distance index, based on selected wavebands. The results reveal that there were statistically significant differences (p < 0.05) between the different spectral indices of most of the plant species in the forests. However, the red-edge parameters showed the highest potential (p < 0.001) to discriminate different plant species. With leaf spectral signatures, the mean reflectance between all plant species was significantly different (p < 0.05) at 562 (75%) wavelength bands. Among pairwise comparisons, invasive Leucaena leucocephala showed the best discriminating ability, with Dodonaea viscosa having 505 significant wavebands showing variations between them. Jeffries Matusita distance analysis revealed that band combinations of the red-edge region (725, 726 nm) showed the best spectral separability (85%) for all species. Our findings suggest that it is possible to identify and discriminate invasive species through field spectroscopy for their future monitoring and management. However, the upscaling of hyperspectral measurements to airborne and satellite sensors can provide a reliable estimation of invasion through mapping inside the protected areas and can help to conserve biodiversity and environmental ecosystems in the future.

Sensitivity Analysis of Sentinel-1 Backscatter to Oil Palm Plantations at Pluriannual Scale: A Case Study in Gabon, Africa

The present paper focuses on a sensitivity analysis of Sentinel-1 backscattering signatures from oil palm canopies cultivated in Gabon, Africa. We employed one Sentinel-1 image per year during the 2015–2021 period creating two separated time series for both the wet and dry seasons. The first images were almost simultaneously acquired to the initial growth stage of oil palm plants. The VH and VV backscattering signatures were analysed in terms of their corresponding statistics for each date and compared to the ones corresponding to tropical forests. The times series for the wet season showed that, in a time interval of 2–3 years after oil palm plantation, the VV/VH ratio in oil palm parcels increases above the one for forests. Backscattering and VV/VH ratio time series for the dry season exhibit similar patterns as for the wet season but with a more stable behaviour. The separability of oil palm and forest classes was also quantitatively addressed by means of the Jeffries–Matusita distance, which seems to point to the C-band VV/VH ratio as a potential candidate for discrimination between oil palms and natural forests, although further analysis must still be carried out. In addition, issues related to the effect of the number of samples in this particular scenario were also analysed. Overall, the outcomes presented here can contribute to the understanding of the radar signatures from this scenario and to potentially improve the accuracy of mapping techniques for this type of ecosystems by using remote sensing. Nevertheless, further research is still to be done as no classification method was performed due to the lack of the required geocoded reference map. In particular, a statistical assessment of the radar signatures should be carried out to statistically characterise the observed trends.

Identification of the Optimal Season and Spectral Regions for Shrub Cover Estimation in Grasslands.

Woody plant encroachment (WPE), the expansion of native and non-native trees and shrubs into grasslands, is a less studied factor that leads to declines in grassland ecosystem health. With the increasing application of remote sensing in grassland monitoring and measuring, it is still difficult to detect WPE at its early stages when its spectral signals are not strong enough. Even at late stages, woody species have strong vegetation characteristics that are commonly categorized as healthy ecosystems. We focus on how shrub encroachment can be detected through remote sensing by looking at the biophysical and spectral properties of the WPE grassland ecosystem, investigating the appropriate season and wavelengths that identify shrub cover, testing the spectral separability of different shrub cover groups and by revealing the lowest shrub cover that can be detected by remote sensing. Biophysical results indicate spring as the best season to distinguish shrubs in our study area. The earliest shrub encroachment can be identified most likely only when the cover reaches between 10% and 25%. A correlation between wavelength spectra and shrub cover indicated four regions that are statistically significant, which differ by season. Furthermore, spectral separability of shrubs increases with their cover; however, good separation is only possible for pure shrub pixels. From the five separability metrics used, Transformed divergence and Jeffries-Matusita distance have better interpretations. The spectral regions for pure shrub pixel separation are slightly different from those derived by correlation and can be explained by the influences from land cover mixtures along our study transect.

Multiobjective Sine Cosine Algorithm for Remote Sensing Image Spatial-Spectral Clustering.

Remote sensing image data clustering is a tough task, which involves classifying the image without any prior information. Remote sensing image clustering, in essence, belongs to a complex optimization problem, due to the high dimensionality and complexity of remote sensing imagery. Therefore, it can be easily affected by the initial values and trapped in locally optimal solutions. Meanwhile, remote sensing images contain complex and diverse spatial-spectral information, which makes them difficult to model with only a single objective function. Although evolutionary multiobjective optimization methods have been presented for the clustering task, the tradeoff between the global and local search abilities is not well adjusted in the evolutionary process. In this article, in order to address these problems, a multiobjective sine cosine algorithm for remote sensing image data spatial-spectral clustering (MOSCA_SSC) is proposed. In the proposed method, the clustering task is converted into a multiobjective optimization problem, and the Xie-Beni (XB) index and Jeffries-Matusita (Jm) distance combined with the spatial information term (SI_Jm measure) are utilized as the objective functions. In addition, for the first time, the sine cosine algorithm (SCA), which can effectively adjust the local and global search capabilities, is introduced into the framework of multiobjective clustering for continuous optimization. Furthermore, the destination solution in the SCA is automatically selected and updated from the current Pareto front through employing the knee-point-based selection approach. The benefits of the proposed method were demonstrated by clustering experiments with ten UCI datasets and four real remote sensing image datasets.

Can a Remote Sensing Approach with Hyperspectral Data Provide Early Detection and Mapping of Spatial Patterns of Black Bear Bark Stripping in Coast Redwoods?

The prevalence of black bear (Ursus americanus) bark stripping in commercial redwood (Sequoia sempervirens (D. Don) Endl.) timber stands has been increasing in recent years. This stripping is a threat to commercial timber production because of the deleterious effects on redwood tree fitness. This study sought to unveil a remote sensing method to detect these damaged trees early and map their spatial patterns. By developing a timely monitoring method, forest timber companies can manipulate their timber harvesting routines to adapt to the consequences of the problem. We explored the utility of high spatial resolution UAV-collected hyperspectral imagery as a means for early detection of individual trees stripped by black bears. A hyperspectral sensor was used to capture ultra-high spatial and spectral information pertaining to redwood trees with no damage, those that have been recently attacked by bears, and those with old bear damage. This spectral information was assessed using the Jeffries-Matusita (JM) distance to determine regions along the electromagnetic spectrum that are useful for discerning these three-health classes. While we were able to distinguish healthy trees from trees with old damage, we were unable to distinguish healthy trees from recently damaged trees due to the inherent characteristics of redwood tree growth and the subtle spectral changes within individual tree crowns for the time period assessed. The results, however, showed that with further assessment, a time window may be identified that informs damage before trees completely lose value.

Using Machine Learning and Hyperspectral Images to Assess Damages to Corn Plant Caused by Glyphosate and to Evaluate Recoverability

Glyphosate is the most widely used herbicide in crop production due to the widespread adoption of glyphosate-resistant (GR) crops. However, the spray of glyphosate onto non-target crops from ground or aerial applications can cause severe injury to non-GR corn plants. To evaluate the crop damage of the non-GR corn plants from glyphosate and the recoverability of the damaged plants, we used the hyperspectral imaging (HSI) technique in field experiments with different glyphosate application rates. This study investigated the spectral characteristic of corn plants and assessed the corn plant damage from glyphosate. Based on HSI image analysis, a spectral variation pattern was observed at 1 week after treatment (WAT), 2 WAT, and 3 WAT from the glyphosate-treated non-GR corn plants. It was further found that the corn plants treated with glyphosate rates equal to or higher than 0.5X (X = 0.866 kilograms acid equivalents/hectare (kg ae/ha) represents the recommended spray rate for GR corn) would suffer unrecoverable damage. Using the Jeffries–Matusita distance as the spectral sensitivity criterion, three sensitive bands from the measured spectra were selected to create two spectral indices for crop recoverability differentiation in band ratio and normalization forms, respectively. With the two spectral indices, the corn plants recoverable and unrecoverable from damage were classified with an overall accuracy greater than 95%. Then, three machine learning algorithms (k-nearest neighbors, random forest, and support vector machine) were respectively combined with the successive projections algorithm to create models to relate selected feature spectral bands to glyphosate spray rates. The results indicated that the models achieved reasonable accuracy, especially in the group of recoverable plants. This study illustrated the potential of the hyperspectral imaging technique for evaluating crop damage from herbicides and recoverability of the injured plants using different data analysis and machine learning modeling approaches for practical weed management in crop fields.

Incident Angle Dependence of Sentinel-1 Texture Features for Sea Ice Classification

Robust and reliable classification of sea ice types in synthetic aperture radar (SAR) images is needed for various operational and environmental applications. Previous studies have investigated the class-dependent decrease in SAR backscatter intensity with incident angle (IA); others have shown the potential of textural information to improve automated image classification. In this work, we investigate the inclusion of Sentinel-1 (S1) texture features into a Bayesian classifier that accounts for linear per-class variation of its features with IA. We use the S1 extra-wide swath (EW) product in ground-range detected format at medium resolution (GRDM), and we compute seven grey level co-occurrence matrix (GLCM) texture features from the HH and the HV backscatter intensity in the linear and logarithmic domain. While GLCM texture features obtained in the linear domain vary significantly with IA, the features computed from the logarithmic intensity do not depend on IA or reveal only a weak, approximately linear dependency. They can therefore be directly included in the IA-sensitive classifier that assumes a linear variation. The different number of looks in the first sub-swath (EW1) of the product causes a distinct offset in texture at the sub-swath boundary between EW1 and the second sub-swath (EW2). This offset must be considered when using texture in classification; we demonstrate a manual correction for the example of GLCM contrast. Based on the Jeffries–Matusita distance between class histograms, we perform a separability analysis for 57 different GLCM parameter settings. We select a suitable combination of features for the ice classes in our data set and classify several test images using a combination of intensity and texture features. We compare the results to a classifier using only intensity. Particular improvements are achieved for the generalized separation of ice and water, as well as the classification of young ice and multi-year ice.

Classification of Soybean Pubescence from Multispectral Aerial Imagery.

The accurate determination of soybean pubescence is essential for plant breeding programs and cultivar registration. Currently, soybean pubescence is classified visually, which is a labor-intensive and time-consuming activity. Additionally, the three classes of phenotypes (tawny, light tawny, and gray) may be difficult to visually distinguish, especially the light tawny class where misclassification with tawny frequently occurs. The objectives of this study were to solve both the throughput and accuracy issues in the plant breeding workflow, develop a set of indices for distinguishing pubescence classes, and test a machine learning (ML) classification approach. A principal component analysis (PCA) on hyperspectral soybean plot data identified clusters related to pubescence classes, while a Jeffries-Matusita distance analysis indicated that all bands were important for pubescence class separability. Aerial images from 2018, 2019, and 2020 were analyzed in this study. A 60-plot test (2019) of genotypes with known pubescence was used as reference data, while whole-field images from 2018, 2019, and 2020 were used to examine the broad applicability of the classification methodology. Two indices, a red/blue ratio and blue normalized difference vegetation index (blue NDVI), were effective at differentiating tawny and gray pubescence types in high-resolution imagery. A ML approach using a support vector machine (SVM) radial basis function (RBF) classifier was able to differentiate the gray and tawny types (83.1% accuracy and kappa = 0.740 on a pixel basis) on images where reference training data was present. The tested indices and ML model did not generalize across years to imagery that did not contain the reference training panel, indicating limitations of using aerial imagery for pubescence classification in some environmental conditions. High-throughput classification of gray and tawny pubescence types is possible using aerial imagery, but light tawny soybeans remain difficult to classify and may require training data from each field season.

Satellite imagery and spectral matching for improved estimation of calcium carbonate and iron oxide abundance in mine areas

This study attempts grade-wise mapping in a limestone mine in Ariyalur, Southern India, and in the iron ore mines of Noamundi, Eastern India. After noise removal in the Sentinel 2A (multispectral) and EO-1 Hyperion (hyperspectral) image datasets, spectral matching is performed using the Jeffries-Matusita (JM) distance, Spectral Correlation Mapper (SCM), and combined JM-SCM measure. Due to the specific absorption spectra for carbonates (1900 nm, 2000 nm, and 2160 nm) and iron oxide (865 nm), it is possible to identify and map such mineral deposits using the multispectral dataset (Sentinel 2A) and hyperspectral dataset (EO-1 Hyperion) respectively. The grade-wise mapping of carbonate in the Ariyalur mine using the Sentinel 2A dataset by the Jeffries-Matusita (JM) approach and Spectral Correlation Mapper (SCM) yielded R2 values of 0.44 and 0.77 respectively, whereas the combined JM-SCM approach resulted in a higher correlation with an R2 value of 0.87. The grade-wise mapping of iron oxide in Noamundi using the EO-1 Hyperion dataset by the Jeffries-Matusita (JM) approach and the Spectral Correlation Mapper (SCM) approach yielded R2 values of 0.15 and 0.76, respectively, whereas the combined JM-SCM approach resulted in a higher correlation with an R2 value of 0.90. Such an improved performance of the combined approach is primarily due to the simultaneous and effective utilization of band-wise information (by JM) and correlation aspects (by SCM) of the reference and target spectra considered in the matching algorithm. Thus, in this study, the proposed algorithm proved its compatibility and utility in extracting information on mineral abundance distribution for mine areas.

English

The aim of the present study is to test ESA’s Sentinel-2 (S2) satellites (S2A and S2B) for an efficient quantification of land cover (LC) and forest compositions in a tropical environment southwest of Mount Kenya. Furthermore, outcome of the research is used to validate ESA’s S2 prototype LC 20 m map of Africa that was produced in 2016. A decision tree that is based on significant altitudinal ranges was used to discriminate four natural tree compositions that occur within the investigation area. In addition, the classification process was supported by Google Earth images, and land use (LU) data that were provided by the local Kenyan Forest Service (KFS). Final classification products include four LC classes and five subclasses of forest (four natural forest subclasses plus one non-natural forest class). Results of the Jeffries-Matusita (JM) distance test show significant differences in spectral separability between all classes. Furthermore, the study identifies spectral signatures and significant wavelengths for a classification of all LC classes and forest subclasses where wavelengths of SWIR and the red-edge domain show highest importance for the discrimination of tree compositions. Finally, considerable differences can be seen between the utilized multi-temporal classification set (total of 39 bands from three acquisition dates) and ESA’s S2 prototype LC 20 m map of Africa 2016. A visual comparison of ESA’s prototype map within the investigation area indicates an overrepresentation of tree cover areas (as confirmed in previous studies) and also an underrepresentation of water.   Key words: Tropical tree composites, Mt. Kenya, Sentinel-2, ESA S2 LC 20 m map of Africa.