Minimum Noise Fraction Transformation Research Articles

A Novel Method Based on GPU for Real-Time Anomaly Detection in Airborne Push-Broom Hyperspectral Sensors

The airborne hyperspectral remote sensing systems (AHRSSs) acquire images with high spectral resolution, high spatial resolution, and high temporal dimension. While the AHRSS captures more detailed information from the terrain objects, the computational complexity of data processing is greatly increased. As an important application technology in the hyperspectral domain, anomaly detection (AD) processing must be real-time and high-precision in many cases, such as post-disaster rescue, military battlefield search, and natural disaster detection. In this paper, the real-time AD technology for the push-broom AHRSS is studied, the mathematical model is established, and a novel implementation framework is proposed. Firstly, the optimized kernel minimum noise fraction (OP-KMNF) transformation is employed to extract informative and discriminative features between the background and anomalies. Secondly, the Nyström method is introduced to reduce the computational complexity of OP-KMNF transformation by decomposing and extrapolating the sub-kernel matrix to estimate the eigenvector of the entire kernel matrix. Thirdly, the extracted features are transferred to hard disks for data storage. Then, taking the extracted features as input data, the background separation model-based CEM anomaly detector (BSM-CEMAD) is imported to detect anomalies. Finally, graphics processing unit (GPU) parallel computing is utilized in the Nyström-based OP-KMNF (NOP-KMNF) transformation and the BSM-CEMAD to improve the execution efficiency, and the real-time AD for the push-broom AHRSS could be realized. To test the feasibility of the implementation framework proposed in this paper, the experiment is carried out with the Airborne Multi-Modular Imaging Spectrometer (AMMIS) developed by the Shanghai Institute of Technical Physics as the data acquisition platform. The experimental results show that the proposed method outperforms many other state-of-the-art AD methods in anomalies detection and background suppression. Moreover, under the condition that the downlink data could retain most of the hyperspectral data information, the proposed method achieves real-time detection of pixel-level anomalies, with the initial delay not exceeding 1 s, the false alarm rate (FAR) less than 5%, and the true positive rate (TPR) close to 98%.

Mapping hydrothermal alterations and lineaments associated with epithermal and massive sulphides deposits of Tifraouine (northwest Algerian coast): Use of Landsat 8 OLI data and remote sensing

Within the context of studying mineralizations related to the Miocene volcanic rocks from the northern part of Algeria, Landsat 8 OLI data and remote sensing techniques are carried out to characterize and map hydrothermal alteration outcrops in the Miocene volcanic rocks of Tifraouine on the northwestern Algerian coast, where several indications of mineralization are known. Short-wave infrared (SWIR) bands allowed for the identification of clay alterations, while the visible and near-infrared bands help in the detection of iron alterations. Red-green-blue color combination images of 6/7, 4/2 and 5/6 ratios were produced. Principal component analysis (PCA) was computed to characterise altered zones and extract lineaments. Matched filtering technique (MF) allows direct detection after applying minimum noise fraction (MNF) transformation and pixel purity index to the spectra in order to obtain a distribution map of the target mineral species in studied area. The remote sensing results combined with the structural geological data determine a relationship between mineralization and structural features and show that hydrothermal alterations are clearly controlled by tectonic. They are highly concentrated in the calderas zone and extend towards the Bouzedjar area which allowed checking the map performed according to the field studies. This will contribute to preliminary studies related to mineral exploration and enhance the mining potential of Tifraouine. The acquisition of supplementary data would allow us to extend this study in coming works.

Utilization of ASTER data in lithological and lineament mapping of the southern flank of the Central High Atlas in Morocco

Geological mapping undoubtedly plays an important role in several studies and remote sensing data are of great significance in geological mapping, particularly in poorly mapped areas situated in inaccessible regions. In the present study, Principal Component Analysis (PCA), Band Rationing (BR) and Minimum Noise Fraction (MNF) algorithms are applied to map lithological units and extract lineaments in the Amezri-Amassine area, by using multispectral ASTER image and global digital elevation model (GDEM) data for the first time. Following preprocessing of ASTER images, advanced image algorithms such as PCA, BR and MNF analyses are applied to the 9ASTER bands. Validation of the resultant maps has relied on matching lithological boundaries and faults in the study area and on the basis of pre-existing geological maps. In addition to the PCA image, a new band-ratio image, 4/6–5/8–4/5, as adopted in the present work, provides high accuracy in discriminating lithological units. The MNF transformation reveals improvement over previous enhancement techniques, in detailing most rock units in the area. Hence, results derived from the enhancement techniques show a good correlation with the existing litho-structural map of the study area. In addition, the present results have allowed to update this map by identifying new lithological units and structural lineaments. Consequently, the methodology followed here has provided satisfactory results and has demonstrated the high potential of multispectral ASTER data for improving lithological discrimination and lineament extraction.

An integrated approach for rapid exploration of carbonatites and related mineral resources

AbstractCarbonatites are proven significant repositories of several critical and strategic elements such as rare earth elements, niobium, thorium, and uranium. Owing to their economic significance, mapping of carbonatites and associated mineral deposits has occupied prominent place in mineral resource exploration programs. In this study an integrated approach was developed to map carbonatite and related mineral deposits in the Loe‐Shilman, Northwest Himalaya of Pakistan, using remotely sensed advance space‐borne thermal emission and reflection radiometer (ASTER) multispectral data and visible near infrared and short‐wave infrared (VNIR‐SWIR) spectral characteristics of minerals in these deposits. Several image enhancement techniques, including band ratio (i.e., B4/B3), principal component and minimum noise fraction transformation (PC6 and MNF5, respectively) helped in highlighting the targeted rocks. The results demonstrate the suitability of ASTER data for discriminating carbonatite related mineral deposits from other surrounding lithologies. Results obtained from these methods were validated through field observations in the area and further confirmed through petrographic and chemical analyses of collected specimens. Field data have also served as training data to perform a supervised classification, allowing further improvement of the mapping results. Moreover, the obtained results from the techniques used for exploring carbonatites and related mineral deposits were stacked together for comparison with each other, to check their sensitivities, and assess their efficiency and accuracy. Generally, all these methods successfully highlighted carbonatites and related mineral deposits; however, when used integratively they exhibit higher degree of accuracy, and has proven to be relatively rapid and cost‐effective.

Reflectance spectra and AVIRIS-NG airborne hyperspectral data analysis for mapping ultramafic rocks in igneous terrain

The layered Sittampundi Anorthosite Complex is covered by mafic and ultramafic rocks including anorthosite, gabbro, pyroxenite and other igneous rocks. The ultramafic terrain has frequently undergone metamorphism. In the present study, laboratory spectral measurements were carried out from mafic, ultramafic and felsic rocks in the 350–2500 nm spectral range to characterise their diagnostic spectral features and for further utilisation for rock-type mapping. In 2016, the Sittampundi complex was covered by an AVIRIS-NG airborne survey jointly conducted by the Space Application Centre (SAC-ISRO) and Jet Propulsion Laboratory (NASA). The level-2 AVIRIS-NG data was obtained from SAC and used to interpret various rock types. ENVI 5.3 software was used for digital image processing of the AVIRIS-NG airborne hyperspectral data. The continuum-removed spectra of major rock types including anorthosite, meta-anorthosite, gabbro, meta-gabbro, pyroxenite, pegmatite, granite, gneiss and migmatite were critically analysed and their diagnostic absorption features correlated with chemistry and mineralogy. The AVIRIS-NG data analyses include bad band removal, minimum noise fraction transformation (MNF) and band combination. Out of various band combinations, the MNF composite images B456, B546 and B561 provided an enhanced output for the delineation of various rock types in the ultramafic terrain.

Detection and Relative Quantification of Neodymium in Sillai Patti Carbonatite Using Decision Tree Classification of the Hyperspectral Data.

A recent increase in the importance of Rare Earth Elements (REEs), proportional to advancements in modern technology, green energy, and defense, has urged researchers to look for more sophisticated and efficient exploration methods for their host rocks, such as carbonatites. Hyperspectral remote sensing has long been recognized as having great potential to identify the REEs based on their sharp and distinctive absorption features in the visible near-infrared (VNIR) and shortwave infrared (SWIR) electromagnetic spectral profiles. For instance, neodymium (Nd), one of the most abundant Light Rare Earth Elements (LREEs), has among the most distinctive absorption features of REEs in the VNIR part of the electromagnetic spectrum. Centered at ~580, ~745, ~810, and ~870 nm in the VNIR, the positions of these absorption features have been proved to be independent of the mineralogy that hosts Nd, and the features can be observed in samples as low in Nd as 1000 ppm. In this study, a neodymium index (NI) is proposed based on the 810 nm absorption feature and tested on the hyperspectral images of the Sillai Patai carbonatite samples to identify Nd pixels and to decipher the relative concentration of Nd in the samples based on the depth of the absorption feature. A preliminary spectral study of the carbonatite samples was carried out using a spectroradiometer to determine the presence of Nd in the samples. Only two of the absorption features of Nd, centered at ~745 and ~810 nm, are prominent in the Nd-rich samples. The other absorption features are either weak or suppressed by the featureless spectra of the associated minerals. Similar absorption features are found in the VNIR and SWIR images of the rock samples captured by the laboratory-based hyperspectral cameras that are processed through Minimum Noise Fraction (MNF) and Fast Fourier Transform (FFT) to filter the signal and noise from the reflectance data. An RGB false-color composite of continuum-removed VNIR reflectance bands covering wavelengths of 587.5, 747.91, and 810.25 nm efficiently displayed the spatial distribution of Nd-rich hotspots in the hyperspectral image. The depth of the 810 nm absorption feature, which corresponds to the concentration of Nd in a pixel, is comparatively greater in these zones and is quantified using the proposed NI such that the deeper the absorption feature, the higher the NI. To quantify the Nd-rich pixels in the continuum-removed VNIR images, different threshold values of NI are introduced into a decision tree classifier which generates the number of pixels in each class. The strength of the proposed NI coupled with the decision tree classifier is further supported by the accuracy assessment of the classified images generating the Kappa coefficient of 0.82. Comparing the results of the remote sensing data obtained in this study with some of the previously published studies suggests that the Sillai Patti carbonatite is rich in Nd and associated REEs, with some parts of the samples as high in Nd concentration as 1000 ppm.

SURFACE HANDWRITING ENHANCEMENT OF ARTIFACTS BASED ON MANIFOLD LEARNING AND MIXED PIXEL DECOMPOSITION

Abstract. Written information on the surface of cultural relics can record important historical events. Due to the influence of natural and human factors, the surface of cultural relics fades and the words are difficult to identify. Take advantage of the hyperspectral data image and spectral unity and wide spectral range, a cultural relics surface handwriting enhancement method based on manifold learning and mixed pixel decomposition was proposed. First, the minimum noise fraction (MNF) transformation was carried out on the hyperspectral image, and then the top 10 bands were selected for inverse MNF transformation to reduce noise of the hyperspectral image. Then, the reconstructed image was dimensionally reduced by locally linear embedding (LLE) to obtain a gray image with the maximum amount of information. At the same time, the spectral features of the handwriting and background area in the reconstructed image were analysed. The automatic target generation process (ATGP) was adopted for endmember extraction on the reconstructed image to identify the endmember of handwriting. The abundance map of handwriting area was obtained by the fully constrained least squares (FCLS). Finally, the gray image and the abundance map of the handwriting region were weighted together to obtain the handwriting enhanced image. The true color image was synthesized from the reconstructed image, Then the true color image and the handwriting enhancement image were fused to obtain the handwritting fusion image. The hyperspectral image of a faded text in Shuozhou City, Shanxi Province, China, was used as an example for verification, and the results showed that the method can effectively enhance the text on the surface of the artifacts.

A Novel Method for Fast Kernel Minimum Noise Fraction Transformation in Hyperspectral Image Dimensionality Reduction

Feature extraction, aiming to simplify and optimize data features, is a typical hyperspectral image dimensionality reduction technique. As a kernel-based method, kernel minimum noise fraction (KMNF) transformation is excellent at handling the nonlinear features within HSIs. It adopts the kernel function to ensure data linear separability by transforming the original data to a higher feature space, following which a linear analysis can be performed in this space. However, KMNF transformation has the problem of high computational complexity and low execution efficiency. It is not suitable for the processing of large-scale datasets. In terms of this problem, this paper proposes a graphics processing unit (GPU) and Nyström method-based algorithm for Fast KMNF transformation (GNKMNF). First, the Nyström method estimates the eigenvector of the entire kernel matrix in KMNF transformation by the decomposition and extrapolation of the sub-kernel matrix to reduce the computational complexity. Then, the sample size in the Nyström method is determined utilizing a proportional gradient selection strategy. Finally, GPU parallel computing is employed to further improve the execution efficiency. Experimental results show that compared with KMNF transformation, improvements of up to 1.94% and 2.04% are achieved by GNKMNF in overall classification accuracy and Kappa, respectively. Moreover, with a data size of 64 × 64 × 250, the execution efficiency of GNKMNF speeds up by about 80×. The outcome demonstrates the significant performance of GNKMNF in feature extraction and execution efficiency.

Exploring mineralization zones using remote sensing and aeromagnetic data, West Allaqi Area, Eastern-Desert, Egypt

Wadi-Allaqi district, in the southern part of the Eastern Desert of Egypt, is a historic well-known gold resource that has been exploited since the Pharaonic era. The study area represents the western segment of Wadi-Allaqi, about 220 km southeast Aswan city, Upper Egypt. It covers approximately 4702 km2. Exploring new potential mineralization zones was conducted by integrating the Landsat-8 and aeromagnetic data to map different surfaces of alteration zones and trace shallow and deep mineralized bodies. The corrected Landsat-Operational Land Imager (OLI) bands were processed using the Band Ratio (BR), Crosta, selective for the Minimum Noise Fraction (MNF) transformation, and False-Color Composition (FCC) techniques were used to detect the hydrothermal alteration zones and recognize the different geological units. The enhanced eigenimage (MNF1) was processed to delineate the fractures and weakness zones over the west Allaqi-Heini Suture (AHS). The shallow residual mineralized sources have been detected and mapped using the Tilt Derivative (TDR), First Derivative (FVD) and high pass filter techniques of aeromagnetic data. The depths to the mineralized magnetic sources, and hence the roots of gold and other minerals, were obtained using the power spectrum and Euler techniques. Accordingly, new gold and mineralization potential zones were successfully delineated over the study area's ophiolitic, metavolcanics, and granitic rocks. Lineaments deduced from aeromagnetic and remote sensing data represent the pathways of the ascending hydrothermal fluids. The main trends of the dominant structural systems were detected through the statistical analyses of these lineaments.

Coupling Relationship Analysis of Gold Content Using Gaofen-5 (GF-5) Satellite Hyperspectral Remote Sensing Data: A Potential Method in Chahuazhai Gold Mining Area, Qiubei County, SW China

The gold (Au) geochemical anomaly is an important indicator of gold mineralization. While the traditional field geochemical exploration method is time-consuming and expensive, the hyperspectral remote sensing technique serves as a robust technique for the delineation and mapping of hydrothermally altered and weathered mineral deposits. Nonetheless, mineralization element anomaly detection was still seldomly used in previous hyperspectral remote sensing applications in mineralization. This study explored the coupling relationship between Gaofen-5 (GF-5) hyperspectral data and Au geochemical anomalies through several models. The Au geochemical anomalies in the Chahuazhai mining area, Qiubei County, Yunnan Province, SW China, was studied in detail. First, several noise reduction methods including radiometric calibration, Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes (FLAASH), Savitzky–Golay filter, and endmember choosing methods including Minimum Noise Fraction (MNF) transformation, matched filtering, and Fast Fourier Transform (FFT) transformation were applied to the Gaofen-5 (GF-5) hyperspectral data processing. The Spectrum-Area (S-A) method was introduced to build an FFT filter to highlight the spectral abnormal characteristics associated with Au geochemical anomaly information. Specifically, the Matched Filtering (MF) technique was applied to the dataset to find the Au geochemical anomaly abundances of endmembers with innovative large-sample learning. Then, Multiple Linear Regression (MLR), Partial Least Squares (PLS) regression, a Back Propagation (BP) network, and Geographically Weighted Regression (GWR) were used to reveal the coupling relationship between the spectra of the processed hyperspectral data and the Au geochemical anomalies. The results show that the GWR analysis has a much higher coefficient of determination, which implies that the Au geochemical anomalies and the spectral information are highly related to spatial locations. GWR works especially well for showing the regional Au geochemical anomaly trend and simulating the Au concentrated areas. The GWR model with application of the S-A method is applicable to the detection of Au geochemical anomalies, which could provide a potential method for Au deposit exploration using GF-5 hyperspectral data.

Unsupervised Deep Learning for Landslide Detection from Multispectral Sentinel-2 Imagery

This paper proposes a new approach based on an unsupervised deep learning (DL) model for landslide detection. Recently, supervised DL models using convolutional neural networks (CNN) have been widely studied for landslide detection. Even though these models provide robust performance and reliable results, they depend highly on a large labeled dataset for their training step. As an alternative, in this paper, we developed an unsupervised learning model by employing a convolutional auto-encoder (CAE) to deal with the problem of limited labeled data for training. The CAE was used to learn and extract the abstract and high-level features without using training data. To assess the performance of the proposed approach, we used Sentinel-2 imagery and a digital elevation model (DEM) to map landslides in three different case studies in India, China, and Taiwan. Using minimum noise fraction (MNF) transformation, we reduced the multispectral dimension to three features containing more than 80% of scene information. Next, these features were stacked with slope data and NDVI as inputs to the CAE model. The Huber reconstruction loss was used to evaluate the inputs. We achieved reconstruction losses ranging from 0.10 to 0.147 for the MNF features, slope, and NDVI stack for all three study areas. The mini-batch K-means clustering method was used to cluster the features into two to five classes. To evaluate the impact of deep features on landslide detection, we first clustered a stack of MNF features, slope, and NDVI, then the same ones plus with the deep features. For all cases, clustering based on deep features provided the highest precision, recall, F1-score, and mean intersection over the union in landslide detection.

On the Limit of Detection in Infrared Spectroscopic Imaging.

Infrared (IR) spectroscopic imaging instruments' performance can be characterized and optimized by an analysis of their limit of detection (LOD). Here we report a systematic analysis of the LOD for Fourier transform IR (FT-IR) and discrete frequency IR (DFIR) imaging spectrometers. In addition to traditional measurements of sample and blank data, we propose a decision theory perspective to pose the determination of LOD as a binary classification problem under different assumptions of noise uniformity and correlation. We also examine three spectral analysis approaches, namely, absorbance at a single frequency, average of absorbance over selected frequencies and total spectral distance - to suit instruments that acquire discrete or contiguous spectral bandwidths. The analysis is validated by refining the fabrication of a bovine serum albumin protein microarray to provide eight uniform spots from ∼2.8 nL of solution for each concentration over a wide range (0.05-10 mg/mL). Using scanning parameters that are typical for each instrument, we estimate a LOD of 0.16 mg/mL and 0.12 mg/mL for widefield and line scanning FT-IR imaging systems, respectively, using the spectral distance approach, and 0.22 mg/mL and 0.15 mg/mL using an optimal set of discrete frequencies. As expected, averaging and the use of post-processing techniques such as minimum noise fraction transformation results in LODs as low as ∼0.075 mg/mL that correspond to a spotted protein mass of ∼112 fg/pixel. We emphasize that these measurements were conducted at typical imaging parameters for each instrument and can be improved using the usual trading rules of IR spectroscopy. This systematic analysis and methodology for determining the LOD can allow for quantitative measures of confidence in imaging an analyte's concentration and a basis for further improving IR imaging technology.

Unlocking the potential of hyperspectral and LiDAR for above-ground biomass (AGB) and tree species classification in tropical forests

Tree species classification using a combination of airborne hyperspectral and Light Detection and Ranging (LiDAR) can provide valuable and effective methods for forest management, such as planning and monitoring purposes. However, only a few studies have applied tree species classification using both combinations in the tropical forest. The research takes a comparative classification approach to examine several classifiers using airborne hyperspectral in the tropical forest. In addition, Object-Based Image Analysis (OBIA) method was applied on hyperspectral data to extract the crown of individual tree species for classification and estimation purposes. Minimum Noise Fraction Transform (MNF) was applied to reduce the data dimensionality and different training samples from the various species used in this study. The result shows that Support Vector Machine (SVM) and Random Forest (RF) achieved the highest overall accuracy above 50% compared to other classifiers in the tropical forest. Besides, LiDAR data was also used to estimate individual trees' height for all species in the study area. The multiple coefficients of determination (R2) test result between LiDAR and field observation data in eight years gaps is 0.754. Therefore, Above-Ground Biomass (AGB) and carbon stock will estimate using a combination of LiDAR, hyperspectral, and field observation data for individual tree species. This method has proven to provide the required information for forest planning and generation in a short time, especially in tree species identification, AGB, and carbon stock estimation in tropical forests.

Mixed Noise Estimation Model for Optimized Kernel Minimum Noise Fraction Transformation in Hyperspectral Image Dimensionality Reduction

Dimensionality reduction (DR) is of great significance for simplifying and optimizing hyperspectral image (HSI) features. As a widely used DR method, kernel minimum noise fraction (KMNF) transformation preserves the high-order structures of the original data perfectly. However, the conventional KMNF noise estimation (KMNF-NE) uses the local regression residual of neighbourhood pixels, which depends heavily on spatial information. Due to the limited spatial resolution, there are many mixed pixels in HSI, making KMNF-NE unreliable for noise estimation and leading to poor performance in KMNF for classification on HSIs with low spatial resolution. In order to overcome this problem, a mixed noise estimation model (MNEM) is proposed in this paper for optimized KMNF (OP-KMNF). The MNEM adopts the sequential and linear combination of the Gaussian prior denoising model, median filter, and Sobel operator to estimate noise. It retains more details and edge features, making it more suitable for noise estimation in KMNF. Experiments using several HSI datasets with different spatial and spectral resolutions are conducted. The results show that, compared with some other DR methods, the improvement of OP-KMNF in average classification accuracy is up to 4%. To improve the efficiency, the OP-KMNF was implemented on graphics processing units (GPU) and sped up by about 60× compared to the central processing unit (CPU) implementation. The outcome demonstrates the significant performance of OP-KMNF in terms of classification ability and execution efficiency.

An Efficient Multi-Sensor Remote Sensing Image Clustering in Urban Areas via Boosted Convolutional Autoencoder (BCAE)

High-resolution urban image clustering has remained a challenging task. This is mainly because its performance strongly depends on the discrimination power of features. Recently, several studies focused on unsupervised learning methods by autoencoders to learn and extract more efficient features for clustering purposes. This paper proposes a Boosted Convolutional AutoEncoder (BCAE) method based on feature learning for efficient urban image clustering. The proposed method was applied to multi-sensor remote-sensing images through a multistep workflow. The optical data were first preprocessed by applying a Minimum Noise Fraction (MNF) transformation. Then, these MNF features, in addition to the normalized Digital Surface Model (nDSM) and vegetation indexes such as Normalized Difference Vegetation Index (NDVI) and Excess Green (ExG(2)), were used as the inputs of the BCAE model. Next, our proposed convolutional autoencoder was trained to automatically encode upgraded features and boost the hand-crafted features for producing more clustering-friendly ones. Then, we employed the Mini Batch K-Means algorithm to cluster deep features. Finally, the comparative feature sets were manually designed in three modes to prove the efficiency of the proposed method in extracting compelling features. Experiments on three datasets show the efficiency of BCAE for feature learning. According to the experimental results, by applying the proposed method, the ultimate features become more suitable for clustering, and spatial correlation among the pixels in the feature learning process is also considered.

Superpixel-Based Minimum Noise Fraction Feature Extraction for Classification of Hyperspectral Images

In this paper, a novel feature extraction technique called SuperMNF is proposed, which is an extension of the minimum noise fraction (MNF) transformation. In SuperMNF, each superpixel has its own transformation matrix and MNF transformation is performed on each superpixel individually. The basic idea behind the SuperMNF is that each superpixel contains its specific signal and noise covariance matrices which are different from the adjacent superpixels. The extracted features, owning spatial-spectral content and provided in the lower dimension, are classified by maximum likelihood classifier and support vector machines. Experiments that are conducted on two real hyperspectral images, named Indian Pines and Pavia University, demonstrate the efficiency of SuperMNF since it yielded more promising results than some other feature extraction methods (MNF, PCA, SuperPCA, KPCA, and MMP).

AVIRIS-NG hyperspectral data analysis for pre- and post-MNF transformation using per-pixel classification algorithms

The hyperspectral image such as AVIRIS-NG provides a lot of spectral information and fine resolution that identify and discriminate similar objects based on their spectral reflectance. This study reveals a comparative analysis of multiple classification algorithms for AVIRIS-NG data, before and after minimum noise fraction (MNF) transformation for land use land cover classification. Applied distance methods are entirely based upon the availability of a complete set of endmembers for the data and used to classify the pixels in terms of endmembers. Applied classifiers yield more accurate results especially in terms of the overall accuracy after dimensionality reduction. As a result, minimum distance achieved highest 97.76% overall accuracy with a Kappa coefficient of 0.97, whereas Mahalanobis distance yields more precise results before MNF with 95.34% overall accuracy and a Kappa coefficient of 0.94. This study also indicates the importance of data dimensionality reduction for hyperspectral imagery.

Synergistic Use of Multi-Temporal RADARSAT-2 and VENµS Data for Crop Classification Based on 1D Convolutional Neural Network

Annual crop inventory information is important for many agriculture applications and government statistics. The synergistic use of multi-temporal polarimetric synthetic aperture radar (SAR) and available multispectral remote sensing data can reduce the temporal gaps and provide the spectral and polarimetric information of the crops, which is effective for crop classification in areas with frequent cloud interference. The main objectives of this study are to develop a deep learning model to map agricultural areas using multi-temporal full polarimetric SAR and multi-spectral remote sensing data, and to evaluate the influence of different input features on the performance of deep learning methods in crop classification. In this study, a one-dimensional convolutional neural network (Conv1D) was proposed and tested on multi-temporal RADARSAT-2 and VENµS data for crop classification. Compared with the Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN) and non-deep learning methods including XGBoost, Random Forest (RF), and Support Vector Machina (SVM), the Conv1D performed the best when the multi-temporal RADARSAT-2 data (Pauli decomposition or coherency matrix) and VENµS multispectral data were fused by the Minimum Noise Fraction (MNF) transformation. The Pauli decomposition and coherency matrix gave similar overall accuracy (OA) for Conv1D when fused with the VENµS data by the MNF transformation (OA = 96.65 ± 1.03% and 96.72 ± 0.77%). The MNF transformation improved the OA and F-score for most classes when Conv1D was used. The results reveal that the coherency matrix has a great potential in crop classification and the MNF transformation of multi-temporal RADARSAT-2 and VENµS data can enhance the performance of Conv1D.

Comparison of Support Vector Machine and Random Forest Algorithms for Invasive and Expansive Species Classification Using Airborne Hyperspectral Data

Invasive and expansive plant species are considered a threat to natural biodiversity because of their high adaptability and low habitat requirements. Species investigated in this research, including Solidago spp., Calamagrostis epigejos, and Rubus spp., are successfully displacing native vegetation and claiming new areas, which in turn severely decreases natural ecosystem richness, as they rapidly encroach on protected areas (e.g., Natura 2000 habitats). Because of the damage caused, the European Union (EU) has committed all its member countries to monitor biodiversity. In this paper we compared two machine learning algorithms, Support Vector Machine (SVM) and Random Forest (RF), to identify Solidago spp., Calamagrostis epigejos, and Rubus spp. on HySpex hyperspectral aerial images. SVM and RF are reliable and well-known classifiers that achieve satisfactory results in the literature. Data sets containing 30, 50, 100, 200, and 300 pixels per class in the training data set were used to train SVM and RF classifiers. The classifications were performed on 430-spectral bands and on the most informative 30 bands extracted using the Minimum Noise Fraction (MNF) transformation. As a result, maps of the spatial distribution of analyzed species were achieved; high accuracies were observed for all data sets and classifiers (an average F1 score above 0.78). The highest accuracies were obtained using 30 MNF bands and 300 sample pixels per class in the training data set (average F1 score > 0.9). Lower training data set sample sizes resulted in decreased average F1 scores, up to 13 percentage points in the case of 30-pixel samples per class.

Noise reduction and destriping using local spatial statistics and quadratic regression from Hyperion images

Hyperion images from Earth Observing-1 (EO-1) are being used in natural resources assessment and management. The evaluation and verification of Hyperion images for the above applications are validating the EO-1 mission. However, the presence of random and striping noises in Hyperion images affect the accuracy of the results. Therefore, reduction of random noise and stripes from Hyperion images becomes indispensable for the evaluation of the results in natural resources assessment and in optimum use of the data. Thus, a collective approach for correcting pixels with no-data values and removing random noise and stripes from Hyperion radiance images is developed. In the developed method, first, no-data valued pixels are identified and corrected using a local median filter. Minimum noise fraction transformation is then used to reduce random noise from noise-dominated bands. Further, spatial statistical techniques are used to reduce random noise from the rest of the bands. Finally, a local quadratic regression by a least squares method is used to correct bad columns and global stripes, and a local-spatial-statistics-based algorithm is used to detect and correct local stripes. The effectiveness and efficiency of the algorithm is demonstrated by application to two Hyperion images: one from the Udaipur area, western India, and another from the Lulea area, northern Sweden. The results show that the algorithm reduces random and striping noise without introducing unwanted effects and alterations in the original normal data values in the images.