Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Image Classification”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
NASA NTRS2007

Potential of VIIRS Time Series Data for Aiding the USDA Forest Service Early Warning System for Forest Health Threats: A Gypsy Moth Defoliation Case Study

This report details one of three experiments performed during FY 2007 for the NASA RPC (Rapid Prototyping Capability) at Stennis Space Center. This RPC experiment assesses the potential of VIIRS (Visible/Infrared Imager/Radiometer Suite) and MODIS (Moderate Resolution Imaging Spectroradiometer) data for detecting and monitoring forest defoliation from the non-native Eurasian gypsy moth (Lymantria dispar). The intent of the RPC experiment was to assess the degree to which VIIRS data can provide forest disturbance monitoring information as an input to a forest threat EWS (Early Warning System) as compared to the level of information that can be obtained from MODIS data. The USDA Forest Service (USFS) plans to use MODIS products for generating broad-scaled, regional monitoring products as input to an EWS for forest health threat assessment. NASA SSC is helping the USFS to evaluate and integrate currently available satellite remote sensing technologies and data products for the EWS, including the use of MODIS products for regional monitoring of forest disturbance. Gypsy moth defoliation of the mid-Appalachian highland region was selected as a case study. Gypsy moth is one of eight major forest insect threats listed in the Healthy Forest Restoration Act (HFRA) of 2003; the gypsy moth threatens eastern U.S. hardwood forests, which are also a concern highlighted in the HFRA of 2003. This region was selected for the project because extensive gypsy moth defoliation occurred there over multiple years during the MODIS operational period. This RPC experiment is relevant to several nationally important mapping applications, including agricultural efficiency, coastal management, ecological forecasting, disaster management, and carbon management. In this experiment, MODIS data and VIIRS data simulated from MODIS were assessed for their ability to contribute broad, regional geospatial information on gypsy moth defoliation. Landsat and ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) data were used to assess the quality of gypsy moth defoliation mapping products derived from MODIS data and from simulated VIIRS data. The project focused on use of data from MODIS Terra as opposed to MODIS Aqua mainly because only MODIS Terra data was collected during 2000 and 2001-years with comparatively high amounts of gypsy moth defoliation within the study area. The project assessed the quality of VIIRS data simulation products. Hyperion data was employed to assess the quality of MODIS-based VIIRS simulation datasets using image correlation analysis techniques. The ART (Application Research Toolbox) software was used for data simulation. Correlation analysis between MODIS-simulated VIIRS data and Hyperion-simulated VIIRS data for red, NIR (near-infrared), and NDVI (Normalized Difference Vegetation Index) image data products collectively indicate that useful, effective VIIRS simulations can be produced using Hyperion and MODIS data sources. The r(exp 2) for red, NIR, and NDVI products were 0.56, 0.63, and 0.62, respectively, indicating a moderately high correlation between the 2 data sources. Temporal decorrelation from different data acquisition times and image misregistration may have lowered correlation results. The RPC experiment also generated MODIS-based time series data products using the TSPT (Time Series Product Tool) software. Time series of simulated VIIRS NDVI products were produced at approximately 400-meter resolution GSD (Ground Sampling Distance) at nadir for comparison to MODIS NDVI products at either 250- or 500-meter GSD. The project also computed MODIS (MOD02) NDMI (Normalized Difference Moisture Index) products at 500-meter GSD for comparison to NDVI-based products. For each year during 2000-2006, MODIS and VIIRS (simulated from MOD02) time series were computed during the peak gypsy moth defoliation time frame in the study area (approximately June 10 through July 27). Gypsy moth defoliation mapping products from simated VIIRS and MOD02 time series were produced using multiple methods, including image classification and change detection via image differencing. The latter enabled an automated defoliation detection product computed using percent change in maximum NDVI for a peak defoliation period during 2001 compared to maximum NDVI across the entire 2000-2006 time frame. Final gypsy moth defoliation mapping products were assessed for accuracy using randomly sampled locations found on available geospatial reference data (Landsat and ASTER data in conjunction with defoliation map data from the USFS). Extensive gypsy moth defoliation patches were evident on screen displays of multitemporal color composites derived from MODIS data and from simulated VIIRS vegetation index data. Such defoliation was particularly evident for 2001, although widespread denuded forests were also seen for 2000 and 2003. These visualizations were validated using aforementioned reference data. Defoliation patches were visible on displays of MODIS-based NDVI and NDMI data. The viewing of apparent defoliation patches on all of these products necessitated adoption of a specialized temporal data processing method (e.g., maximum NDVI during the peak defoliation time frame). The frequency of cloud cover necessitated this approach. Multitemporal simulated VIIRS and MODIS Terra data both produced effective general classifications of defoliated forest versus other land cover. For 2001, the MOD02-simulated VIIRS 400-meter NDVI classification produced a similar yet slightly lower overall accuracy (87.28 percent with 0.72 Kappa) than the MOD02 250-meter NDVI classification (88.44 percent with 0.75 Kappa). The MOD13 250-meter NDVI classification had a lower overall accuracy (79.13 percent) and a much lower Kappa (0.46). The report discusses accuracy assessment results in much more detail, comparing overall classification and individual class accuracy statistics for simulated VIIRS 400-meter NDVI, MOD02 250-meter NDVI, MOD02-500 meter NDVI, MOD13 250-meter NDVI, and MOD02 500-meter NDMI classifications. Automated defoliation detection products from simulated VIIRS and MOD02 data for 2001 also yielded similar, relatively high overall classification accuracy (85.55 percent for the VIIRS 400-meter NDVI versus 87.28 percent for the MOD02 250-meter NDVI). In contrast, the USFS aerial sketch map of gypsy moth defoliation showed a lower overall classification accuracy at 73.64 percent. The overall classification Kappa values were also similar for the VIIRS (approximately 0.67 Kappa) versus the MOD02 (approximately 0.72 Kappa) automated defoliation detection product, which were much higher than the values exhibited by the USFS sketch map product (overall Kappa of approximately 0.47). The report provides additional details on the accuracy of automated gypsy moth defoliation detection products compared with USFS sketch maps. The results suggest that VIIRS data can be effectively simulated from MODIS data and that VIIRS data will produce gypsy moth defoliation mapping products that are similar to MODIS-based products. The results of the RPC experiment indicate that VIIRS and MODIS data products have good potential for integration into the forest threat EWS. The accuracy assessment was performed only for 2001 because of time constraints and a relative scarcity of cloud-free Landsat and ASTER data for the peak defoliation period of the other years in the 2000-2006 time series. Additional work should be performed to assess the accuracy of gypsy moth defoliation detection products for additional years.The study area (mid-Appalachian highlands) and application (gypsy moth forest defoliation) are not necessarily representative of all forested regions and of all forest threat disturbance agents. Additional work should be performed on other inland and coastal regions as well as for other major forest threats.

Spruce, Joseph P.↗
NASA NTRSDate not supplied

Characterizing Detailed Grain Shape and Size Distribution Properties of Lunar Regolith

Introduction: As the nation prepares to return to the Moon, there is an increasing need for testing tools, instruments, and equipment in simulated environments on Earth to ensure successful operations during lunar missions. Regolith will affect all aspects of future lunar missions, from plume interactions during landing to space suit and tool design [1]. Because of this, it is important to understand the grain shape and size properties of lunar regolith and how those influence regolith behavior in order to prepare for these missions. This knowledge is also vital to create more accurate lunar regolith simulants for testing equipment in a lunar environment. While particle size analyses have been performed on most Apollo soils using simple sieving, shape has only been crudely addressed [2]. This work analyzes 4 lunar regolith samples to provide a better understanding of these size and shape parameters and will provide a more accurate baseline of data to create high-fidelity lunar regolith simulants. Methods: New technologies exist today that are capable of measuring size and shape simultaneously for hundreds of thousands of particles in a single measurement. We conducted a rigorous analysis of the particle size distribution (PSD) as well as the size-dependent 2D and 3D shape parameters of lunar regolith samples of different compositions and maturity levels. This analysis was done using a Microtrac SYNC which provides a unique combination of tri-laser diffraction and Dynamic Image Analysis (DIA). Sample Selection. Four regolith samples were selected for analysis based on maturity level and lunar terrain type: • 10084 – Mature high-Ti mare regolith • 15601 – Immature low-Ti mare regolith • 64501 – Mature highland regolith • 67461 – Immature highland regolith Sample Analysis. After receiving the samples, each sample was imaged with an optical microscope (Figure 1). To obtain 2D and 3D particle size and shape, 0.1 g of each sample was then added to the SYNC for analysis, which outputs more than 30 size and shape parameters for each individual grain as well as the complete size distribution from 0.01-2000 μm by blending laser diffraction and DIA together. For the 0.1 g sample masses requested, ~100,000 grains per sample were captured by DIA. Results: From the PSD analysis, it can be seen that the average particle size of 10084 is ~24.5 µm which is much smaller than the average particle size of the Apollo sample collection (~72 µm) [3], while the average particle size of samples 15601, 64501, and 67461 are larger than the Apollo sample average at ~106 µm, ~103 µm, and ~118.5 µm respectively. The size and shape measurements for the samples output ~30 parameters, four of which were focused on for this study: sphericity, aspect ratio, roundness, and concavity (Figure 3). However, after analyzing the plots it was found that only sphericity and aspect ratio showed differences between the samples. These two parameters are measured on a scale from 0 to 1, with 1 being a perfect sphere with equal dimensions. The results show that the sphericity values are slightly high-er for the mature samples 10084 and 64501 (~0.96) than they are for samples 15601 and 67461 (~0.95) (Figure 4). The aspect ratio values are slightly lower for samples 64501 (~0.7) and 67461 (~0.74) than they are for samples 10084 (~0.76) and 15601 (~0.8) (Figure 5). Discussion: Mature regoliths are those that have been exposed to micrometeorite and solar wind bombardment for long periods of time, breaking up particles and causing them to become more rounded [4]. Therefore, the smaller particle sizes and higher sphericity values for samples 10084 and 64501 are expected due to their higher maturity level compared to 15601 and 67461. However, the aspect ratio values are not dependent on maturity and are instead dependent on terrain type. The lower aspect ratio values for the high-land samples is potentially due to the higher plagioclase content, which occurs in elongated particles and does not break down as easily as the pyroxenes or olivines that are present in the mare. Conclusion and Future Work: The results of this work provide a baseline of high-quality data that will contribute to the creation of high-fidelity lunar simulants and will greatly benefit NASA’s efforts of establishing a human presence on the Moon. Future work includes performing unsupervised image classification on the ~105 particle images per sample in order to identify different classes of grains. These grain classes can then be linked to detailed shape properties, and the relative abundance of each class in the samples can be compared. Acknowledgments: We would like to thank the Extraterrestrial Materials Analysis Group (ExMAG) and the Astromaterials Allocation Review Board (AARB) for allocating lunar regolith samples 10084,9010, 15601,365, 64501,249, and 67461,171 for this work. References: [1] Taylor, L. A. et al. (2005) AIAA #2510. [2] Katagiri et al. (2015) ASCE. [3] Carrier III, W. D. (2005) Lunar Geotech Institute Tech Report. [4] McKay, D. S. et al. (1991) The Lunar Sourcebook, Chapter 7.

S R Deitrick↗
NASA NTRS2012

Advances in Spectral-Spatial Classification of Hyperspectral Images

Recent advances in spectral-spatial classification of hyperspectral images are presented in this paper. Several techniques are investigated for combining both spatial and spectral information. Spatial information is extracted at the object (set of pixels) level rather than at the conventional pixel level. Mathematical morphology is first used to derive the morphological profile of the image, which includes characteristics about the size, orientation, and contrast of the spatial structures present in the image. Then, the morphological neighborhood is defined and used to derive additional features for classification. Classification is performed with support vector machines (SVMs) using the available spectral information and the extracted spatial information. Spatial postprocessing is next investigated to build more homogeneous and spatially consistent thematic maps. To that end, three presegmentation techniques are applied to define regions that are used to regularize the preliminary pixel-wise thematic map. Finally, a multiple-classifier (MC) system is defined to produce relevant markers that are exploited to segment the hyperspectral image with the minimum spanning forest algorithm. Experimental results conducted on three real hyperspectral images with different spatial and spectral resolutions and corresponding to various contexts are presented. They highlight the importance of spectral–spatial strategies for the accurate classification of hyperspectral images and validate the proposed methods.

hyperspectral image↗
NASA NTRS1982

An automatic agricultural zone classification procedure for crop inventory satellite images

A classification procedure for assessing crop areal proportion in multispectral scanner image is discussed. The procedure is into four parts: labeling; classification; proportion estimation; and evaluation. The procedure also has the following characteristics: multitemporal classification; the need for a minimum field information; and verification capability between automatic classification and analyst labeling. The processing steps and the main algorithms involved are discussed. An outlook on the future of this technology is also presented.

Parada, N. D. J.↗
DOE OSTI2023

Deep learning methods for event verification and image re-purposing detection

Systems and methods herein describe accessing an image, generating a resized image, generating an image feature vector by applying an image classification neural network to the resized image, generating analysis of the image by processing the image feature vector using a machine-learning classifier trained to analyze the image feature vector, and based on the analysis, determining an event that is attributed to the image.

Nataraj, Lakshmanan↗
DOE OSTI2023

An extended focused assessment with sonography in trauma ultrasound tissue-mimicking phantom for developing automated diagnostic technologies

Medical imaging-based triage is critical for ensuring medical treatment is timely and prioritized. However, without proper image collection and interpretation, triage decisions can be hard to make. While automation approaches can enhance these triage applications, tissue phantoms must be developed to train and mature these novel technologies. Here, we have developed a tissue phantom modeling the ultrasound views imaged during the enhanced focused assessment with sonography in trauma exam (eFAST). The tissue phantom utilized synthetic clear ballistic gel with carveouts in the abdomen and rib cage corresponding to the various eFAST scan points. Various approaches were taken to simulate proper physiology without injuries present or to mimic pneumothorax, hemothorax, or abdominal hemorrhage at multiple locations in the torso. Multiple ultrasound imaging systems were used to acquire ultrasound scans with or without injury present and were used to train deep learning image classification predictive models. Performance of the artificial intelligent (AI) models trained in this study achieved over 97% accuracy for each eFAST scan site. We used a previously trained AI model for pneumothorax which achieved 74% accuracy in blind predictions for images collected with the novel eFAST tissue phantom. Grad-CAM heat map overlays for the predictions identified that the AI models were tracking the area of interest for each scan point in the tissue phantom. Overall, the eFAST tissue phantom ultrasound scans resembled human images and were successful in training AI models. Tissue phantoms are critical first steps in troubleshooting and developing medical imaging automation technologies for this application that can accelerate the widespread use of ultrasound imaging for emergency triage.

60 APPLIED LIFE SCIENCES↗
NASA NTRS2012

Advances in Spectral-Spatial Classification of Hyperspectral Images

Recent advances in spectral-spatial classification of hyperspectral images are presented in this paper. Several techniques are investigated for combining both spatial and spectral information. Spatial information is extracted at the object (set of pixels) level rather than at the conventional pixel level. Mathematical morphology is first used to derive the morphological profile of the image, which includes characteristics about the size, orientation and contrast of the spatial structures present in the image. Then the morphological neighborhood is defined and used to derive additional features for classification. Classification is performed with support vector machines using the available spectral information and the extracted spatial information. Spatial post-processing is next investigated to build more homogeneous and spatially consistent thematic maps. To that end, three presegmentation techniques are applied to define regions that are used to regularize the preliminary pixel-wise thematic map. Finally, a multiple classifier system is defined to produce relevant markers that are exploited to segment the hyperspectral image with the minimum spanning forest algorithm. Experimental results conducted on three real hyperspectral images with different spatial and spectral resolutions and corresponding to various contexts are presented. They highlight the importance of spectral-spatial strategies for the accurate classification of hyperspectral images and validate the proposed methods.

Fauvel, Mathieu↗
DOE OSTI2022

Nonlocal Kernel Network (NKN): a Stable and Resolution-Independent Deep Neural Network.

Neural operators have recently become popular tools for designing solution maps between function spaces in the form of neural networks. Differently from classical scientific machine learning approaches that learn parameters of a known partial differential equation (PDE) for a single instance of the input parameters at a fixed resolution, neural operators approximate the solution map of a family of PDEs [6, 7]. Despite their success, the uses of neural operators are so far restricted to relatively shallow neural networks and confined to learning hidden governing laws. In this work, we propose a novel nonlocal neural operator, which we refer to as nonlocal kernel network (NKN), that is resolution independent, characterized by deep neural networks, and capable of handling a variety of tasks such as learning governing equations and classifying images. Our NKN stems from the interpretation of the neural network as a discrete nonlocal diffusion reaction equation that, in the limit of infinite layers, is equivalent to a parabolic nonlocal equation, whose stability is analyzed via nonlocal vector calculus. The resemblance with integral forms of neural operators allows NKNs to capture long-range dependencies in the feature space, while the continuous treatment of node-to-node interactions makes NKNs resolution independent. The resemblance with neural ODEs, reinterpreted in a nonlocal sense, and the stable network dynamics between layers allow for generalization of NKN’s optimal parameters from shallow to deep networks. This fact enables the use of shallow-to-deep initialization techniques [8]. Our tests show that NKNs outperform baseline methods in both learning governing equations and image classification tasks and generalize well to different resolutions and depths.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Peri-Net-Pro: the neural processes with quantified uncertainty for crack patterns

Abstract This paper develops a deep learning tool based on neural processes (NPs) called the Peri-Net-Pro, to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified uncertainties. In particular, image classification and regression studies are conducted by means of convolutional neural networks (CNNs) and NPs. First, the amount and quality of the data are enhanced by using peridynamics to theoretically compensate for the problems of the finite element method (FEM) in generating crack pattern images. Second, case studies are conducted with the prototype microelastic brittle (PMB), linear peridynamic solid (LPS), and viscoelastic solid (VES) models obtained by using the peridynamic theory. The case studies are performed to classify the images by using CNNs and determine the suitability of the PMB, LBS, and VES models. Finally, a regression analysis is performed on the crack pattern images with NPs to predict the crack patterns. The regression analysis results confirm that the variance decreases when the number of epochs increases by using the NPs. The training results gradually improve, and the variance ranges decrease to less than 0.035. The main finding of this study is that the NPs enable accurate predictions, even with missing or insufficient training data. The results demonstrate that if the context points are set to the 10th, 100th, 300th, and 784th, the training information is deliberately omitted for the context points of the 10th, 100th, and 300th, and the predictions are different when the context points are significantly lower. However, the comparison of the results of the 100th and 784th context points shows that the predicted results are similar because of the Gaussian processes in the NPs. Therefore, if the NPs are employed for training, the missing information of the training data can be supplemented to predict the results.

Mathematics↗
DOE OSTI2020

Improving the accessibility and transferability of machine learning algorithms for identification of animals in camera trap images: MLWIC2

Motion-activated wildlife cameras (or “camera traps”) are frequently used to remotely and noninvasively observe animals. The vast number of images collected from camera trap projects has prompted some biologists to employ machine learning algorithms to automatically recognize species in these images, or at least filter-out images that do not contain animals. These approaches are often limited by model transferability, as a model trained to recognize species from one location might not work as well for the same species in different locations. Furthermore, these methods often require advanced computational skills, making them inaccessible to many biologists. We used 3 million camera trap images from 18 studies in 10 states across the United States of America to train two deep neural networks, one that recognizes 58 species, the “species model,” and one that determines if an image is empty or if it contains an animal, the “empty-animal model.” Our species model and empty-animal model had accuracies of 96.8% and 97.3%, respectively. Furthermore, the models performed well on some out-of-sample datasets, as the species model had 91% accuracy on species from Canada (accuracy range 36%–91% across all out-of-sample datasets) and the empty-animal model achieved an accuracy of 91%–94% on out-of-sample datasets from different continents. Our software addresses some of the limitations of using machine learning to classify images from camera traps. By including many species from several locations, our species model is potentially applicable to many camera trap studies in North America. We also found that our empty-animal model can facilitate removal of images without animals globally. We provide the trained models in an R package (MLWIC2: Machine Learning for Wildlife Image Classification in R), which contains Shiny Applications that allow scientists with minimal programming experience to use trained models and train new models in six neural network architectures with varying depths.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1984

A comparison of Landsat point and rectangular field training sets for land-use classification

Rectangular training fields of homogeneous spectroreflectance are commonly used in supervised pattern recognition efforts. Trial image classification with manually selected training sets gives irregular and misleading results due to statistical bias. A self-verifying, grid-sampled training point approach is proposed as a more statistically valid feature extraction technique. A systematic pixel sampling network of every ninth row and ninth column efficiently replaced the full image scene with smaller statistical vectors which preserved the necessary characteristics for classification. The composite second- and third-order average classification accuracy of 50.1 percent for 331,776 pixels in the full image substantially agreed with the 51 percent value predicted by the grid-sampled, 4,100-point training set.

Tom, C. H.↗
NASA NTRS1978

Analysis of the classification of US and Canadian intensive test sites using the Image 100 hybrid classification system

The author has identified the following significant results. Labeling of wheat rather than total grains, particularly with only one acquisition, led to significant overestimates in some segments. The Image-100 software and procedures were written to facilitate classification of the LACIE segments but were not designed to record data for later accuracy assessment. A much better evaluation would have been possible if accuracy assessment data had been collected following each satisfactory classification.

Hocutt, W. T.↗
NASA NTRS1991

Neuro-classification of multi-type Landsat Thematic Mapper data

Neural networks have been successful in image classification and have shown potential for classifying remotely sensed data. This paper presents classifications of multitype Landsat Thematic Mapper (TM) data using neural networks. The Landsat TM Image for March 23, 1987 with accompanying ground observation data for a study area In Miami County, Indiana, U.S.A. was utilized to assess recognition of crop residues. Principal components and spectral ratio transformations were performed on the TM data. In addition, a layer of the geographic information system (GIS) for the study site was incorporated to generate GIS-enhanced TM data. This paper discusses (1) the performance of neuro-classification on each type of data, (2) how neural networks recognized each type of data as a new image and (3) comparisons of the results for each type of data obtained using neural networks, maximum likelihood, and minimum distance classifiers.

Zhuang, Xin↗
NASA NTRS1974

Use of feature extraction techniques for the texture and context information in ERTS imagery: Spectral and textural processing of ERTS imagery

The author has identified the following significant results. A procedure was developed to extract cross-band textural features from ERTS MSS imagery. Evolving from a single image texture extraction procedure which uses spatial dependence matrices to measure relative co-occurrence of nearest neighbor grey tones, the cross-band texture procedure uses the distribution of neighboring grey tone N-tuple differences to measure the spatial interrelationships, or co-occurrences, of the grey tone N-tuples present in a texture pattern. In both procedures, texture is characterized in such a way as to be invariant under linear grey tone transformations. However, the cross-band procedure complements the single image procedure by extracting texture information and spectral information contained in ERTS multi-images. Classification experiments show that when used alone, without spectral processing, the cross-band texture procedure extracts more information than the single image texture analysis. Results show an improvement in average correct classification from 86.2% to 88.8% for ERTS image no. 1021-16333 with the cross-band texture procedure. However, when used together with spectral features, the single image texture plus spectral features perform better than the cross-band texture plus spectral features, with an average correct classification of 93.8% and 91.6%, respectively.

Haralick, R. H.↗
NASA NTRS1983

User alternatives in post-processing for raster-to-vector conversion

A number of Landsat-based coniferous forest stratum maps have been created of the Eldorado National Forest in California. These maps were produced in raster image format which is not directly usable by the U.S. Forest Service's vector-based Wildland Resource Information System (WRIS). As a solution, raster-to-vector conversion software has been developed for processing classified images into polygonal data structures. Before conversion, however, the digital classification images must be simplified to remove high spatial variance ('noise', 'speckle') and meet a USFS ten acre minimum requirement. A post-processing (simplification) strategy different from those commonly used in raster image processing may be desired for preparing maps for conversion to vector format, because simplification routines typically permit diagonal connections in the process of reclassifying pixels and forming new polygons. Diagonal connections are often undesirable when converting to vector format because they permit polygons to effectively cross over each other and occupy a common location. Three alternative methodologies are discussed for simplifying raster data for conversion to vector format.

Logan, T. L.↗
NASA NTRS2000

Hyperspectral Imaging Sensors and the Marine Coastal Zone

Hyperspectral imaging sensors greatly expand the potential of remote sensing to assess, map, and monitor marine coastal zones. Each pixel in a hyperspectral image contains an entire spectrum of information. As a result, hyperspectral image data can be processed in two very different ways: by image classification techniques, to produce mapped outputs of features in the image on a regional scale; and by use of spectral analysis of the spectral data embedded within each pixel of the image. The latter is particularly useful in marine coastal zones because of the spectral complexity of suspended as well as benthic features found in these environments. Spectral-based analysis of hyperspectral (AVIRIS) imagery was carried out to investigate a marine coastal zone of South Florida, USA. Florida Bay is a phytoplankton-rich estuary characterized by taxonomically distinct phytoplankton assemblages and extensive seagrass beds. End-member spectra were extracted from AVIRIS image data corresponding to ground-truth sample stations and well-known field sites. Spectral libraries were constructed from the AVIRIS end-member spectra and used to classify images using the Spectral Angle Mapper (SAM) algorithm, a spectral-based approach that compares the spectrum, in each pixel of an image with each spectrum in a spectral library. Using this approach different phytoplankton assemblages containing diatoms, cyanobacteria, and green microalgae, as well as benthic community (seagrasses), were mapped.

Richardson, Laurie L.↗
NASA NTRS2012

Improved Hierarchical Optimization-Based Classification of Hyperspectral Images Using Shape Analysis

A new spectral-spatial method for classification of hyperspectral images is proposed. The HSegClas method is based on the integration of probabilistic classification and shape analysis within the hierarchical step-wise optimization algorithm. First, probabilistic support vector machines classification is applied. Then, at each iteration two neighboring regions with the smallest Dissimilarity Criterion (DC) are merged, and classification probabilities are recomputed. The important contribution of this work consists in estimating a DC between regions as a function of statistical, classification and geometrical (area and rectangularity) features. Experimental results are presented on a 102-band ROSIS image of the Center of Pavia, Italy. The developed approach yields more accurate classification results when compared to previously proposed methods.

Classification↗
DOE OSTI2021

HYPPO: A Surrogate-Based Multi-Level Parallelism Tool for Hyperparameter Optimization

We present a new software, HYPPO, that enables the automatic tuning of hyperparameters of various deep learning (DL) models. Unlike other hyperparameter optimization (HPO) methods, HYPPO uses adaptive surrogate models and directly accounts for uncertainty in model predictions to find accurate and reliable models that make robust predictions. Using asynchronous nested parallelism, we are able to significantly alleviate the computational burden of training complex architectures and quantifying the uncertainty. HYPPO is implemented in Python and can be used with both TensorFlow and PyTorch libraries. We demonstrate various software features on time-series prediction and image classification problems as well as a scientific application in computed tomography image reconstruction. Finally, we show that (1) we can reduce by an order of magnitude the number of evaluations necessary to find the most optimal region in the hyperparameter space and (2) we can reduce by two orders of magnitude the throughput for such HPO process to complete.

adaptation models↗