An official website of the United States government
Vegetation classifications on large geographic scales are necessary to inform conservation decisions and monitor keystone, invasive, and endangered species. These classifications are often effectively achieved by applying models to imaging spectroscopy, a type of remote sensing data, but such undertakings are often limited in spatial extent. Here we provide accurate, high-resolution spatial data on the keystone species Metrosideros polymorpha, a highly polymorphic tree species distributed across bioclimatic zones and environmental gradients on Hawai’i Island using airborne imaging spectroscopy and LiDAR. We compare two tree species classification techniques, the support vector machine (SVM) and spectral mixture analysis (SMA), to assess their ability to map M. polymorpha over 28,000 square kilometers where differences in topography, background vegetation, sun angle relative to the aircraft, and day of data collection, among others, challenge accurate classification. To capture spatial variability in model performance, we applied Gaussian process classification (GPC) to estimate the spatial probability density of M. polymorpha occurrence using only training sample locations. We found that while SVM and SMA models exhibit similar raw score accuracy over the test set (96.0% and 93.4%, respectively), SVM better reproduces the spatial distribution of M. polymorpha than SMA. We developed a final 2 m × 2 m M. polymorpha presence dataset and a 30 m × 30 m M. polymorpha density dataset using SVM classifications that have been made publicly available for use in conservation applications. Accurate, large-scale species classifications are achievable, but metrics for model performance assessments must account for spatial variation of model accuracy.
more »
« less
Evaluation of spectral collection strategies for identification of Dalbergia spp. using handheld laser‐induced breakdown spectroscopy
Abstract The illegal timber trade has significant impact on the survival of endangered tropical hardwood species likeDalbergiaspp. (rosewood), a world‐wide protected genus from the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). Due to increased threat toDalbergiaspp., and lack of action to reduce threats, port of entry analysis methods are required to identifyDalbergiaspp. Handheld laser‐induced breakdown spectroscopy (LIBS) has been shown to be capable of identifying species and establishing provenance ofDalbergiaspp. and other tropical hardwoods, but analysis methods for this work have yet to be investigated in detail. The present work investigates five well‐known algorithms—partial least squares discriminant analysis (PLS‐DA), classification and regression trees (CART),k‐nearest neighbor (k‐NN), random forest (RF), and support vector machine (SVM)—two training/test set sampling regimes, and data collection at two signal‐to‐noise (S/N) ratios to assess the potential for handheld LIBS analyses. Additionally, imbalanced classes are addressed. For this application, SVM and RF yield near identical results (though RF takes nearly 100 longer to compute), while the S/N ratio has a significant effect on model success assuming all else is equal. It was found that forming a training set with replicate low S/N analyses can perform as well as higher precision training sets for true prediction, even if the predicted samples have low signal to noise! This work confirms handheld LIBS analyzers can provide a viable method for classification of hardwood species, even within the same genus.
more »
« less
- Award ID(s):
- 2003839
- PAR ID:
- 10412154
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- Journal of Chemometrics
- Volume:
- 38
- Issue:
- 5
- ISSN:
- 0886-9383
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Purpose. To investigate the relationship between spatial parotid dose and the risk of xerostomia in patients undergoing head-and-neck cancer radiotherapy, using machine learning (ML) methods.Methods. Prior to conducting voxel-based ML analysis of the spatial dose, two steps were taken: (1) The parotid dose was standardized through deformable image registration to a reference patient; (2) Bilateral parotid doses were regrouped into contralateral and ipsilateral portions depending on their proximity to the gross tumor target. Individual dose voxels were input into six commonly used ML models, which were tuned with ten-fold cross validation: random forest (RF), ridge regression (RR), support vector machine (SVM), extra trees (ET), k-nearest neighbor (kNN), and naïve Bayes (NB). Binary endpoints from 240 patients were used for model training and validation: 0 (N = 119) for xerostomia grades 0 or 1, and 1 (N = 121) for grades 2 or higher. Model performance was evaluated using multiple metrics, including accuracy, F1score, areas under the receiver operating characteristics curves (auROC), and area under the precision–recall curves (auPRC). Dose voxel importance was assessed to identify local dose patterns associated with xerostomia risk.Results. Four models, including RF, SVM, ET, and NB, yielded average auROCs and auPRCs greater than 0.60 from ten-fold cross-validation on the training data, except for a lower auROC from NB. The first three models, along with kNN, demonstrated higher accuracy and F1scores. A bootstrapping analysis confirmed test uncertainty. Voxel importance analysis from kNN indicated that the posterior portion of the ipsilateral gland was more predictive of xerostomia, but no clear patterns were identified from the other models.Conclusion. Voxel doses as predictors of xerostomia were confirmed with some ML classifiers, but no clear regional patterns could be established among these classifiers, except kNN. Further research with a larger patient dataset is needed to identify conclusive patterns.more » « less
-
KiekiePolotow & Brescovit, 2018 is a Neotropical genus of Ctenidae, with most of its species occuring in Central America. In this study, we review the systematics ofKiekieand describe five new species and the unknown females ofK. barrocoloradoPolotow & Brescovit, 2018 andK. garifunaPolotow & Brescovit, 2018, and the unknown male ofK. verbenaPolotow & Brescovit, 2018. In addition, we described the female ofK. montanensewhich was wrongly assigned asK. griswoldiPolotow & Brescovit, 2018 (both species are sympatric). We provided a modified diagnosis for previously described species based on the morphology of the newly discovered species andin situphotographs of living specimens. We inferred a molecular phylogeny using four nuclear (histone H3, 28S rRNA, 18S rRNA and ITS-2) and three mitochondrial genes (cytochrome c oxidase subunit I or COI, 12S rRNA and 16S rRNA) to test the monophyly of the genus and the evolutionary relationships of its species. Lastly, we reconstruct the historical biogeography and map diversity and endemism distributional patterns of the different species. This study increased the number of known species ofKiekiefrom 13 to 18, and we describe a new genus,Eldivowhich is sister lineage ofKiekie. Most of the diversity and endemism of the genusKiekieis located in the montane ecosystems of Costa Rica followed by the lowland rainforest of the Pacific side (Limon Basin).Kiekieoriginated in the North America Tropical region, this genus started diversifying in the Late Miocene and spread to Lower Central America and South America. In that region,Kiekiecolonized independently several times the montane ecosystems corresponding to periods of uplifting of Talamanca and Central Cordilleras.more » « less
-
Generic characteristics of the leafhopper genus Kusala Dworakowska are revised, and a new subgenus Scodela is established. A species checklist of the Kusala is provided and three new species are added: Kusala (Kusala) viraktamathi, K. (Scodela) directa and K. (S.) sinuata spp. nov.. The related genus Diomma Motschulsky is also redescribed and a new species is described: Diomma (Diomma) sangzhiensis sp. nov..more » « less
-
Markopoulos, Panos P.; Ouyang, Bing (Ed.)We consider the problem of unsupervised (blind) evaluation and assessment of the quality of data used for deep neural network (DNN) RF signal classification. When neural networks train on noisy or mislabeled data, they often (over-)fit to the noise measurements and faulty labels, which leads to significant performance degradation. Also, DNNs are vulnerable to adversarial attacks, which can considerably reduce their classification performance, with extremely small perturbations of their input. In this paper, we consider a new method based on L1-norm principal-component analysis (PCA) to improve the quality of labeled wireless data sets that are used for training a convolutional neural network (CNN), and a deep residual network (ResNet) for RF signal classification. Experiments with data generated for eleven classes of digital and analog modulated signals show that L1-norm tensor conformity curation of the data identifies and removes from the training data set inappropriate class instances that appear due to mislabeling and universal black-box adversarial attacks and drastically improves/restores the classification accuracy of the identified deep neural network architectures.more » « less
