Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Automatic Data Processing”

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 559 records · Page 31
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

STAIR 2.0: A Generic and Automatic Algorithm to Fuse Modis, Landsat, and Sentinel-2 to Generate 10 m, Daily, and Cloud-/Gap-Free Surface Reflectance Product

Remote sensing datasets with both high spatial and high temporal resolution are critical for monitoring and modeling the dynamics of land surfaces. However, no current satellite sensor could simultaneously achieve both high spatial resolution and high revisiting frequency. Therefore, the integration of different sources of satellite data to produce a fusion product has become a popular solution to address this challenge. Many methods have been proposed to generate synthetic images with rich spatial details and high temporal frequency by combining two types of satellite datasets—usually frequent coarse-resolution images (e.g., MODIS) and sparse fine-resolution images (e.g., Landsat). In this paper, we introduce STAIR 2.0, a new fusion method that extends the previous STAIR fusion framework, to fuse three types of satellite datasets, including MODIS, Landsat, and Sentinel-2. In STAIR 2.0, input images are first processed to impute missing-value pixels that are due to clouds or sensor mechanical issues using a gap-filling algorithm. The multiple refined time series are then integrated stepwisely, from coarse- to fine- and high-resolution, ultimately providing a synthetic daily, high-resolution surface reflectance observations. We applied STAIR 2.0 to generate a 10-m, daily, cloud-/gap-free time series that covers the 2017 growing season of Saunders County, Nebraska. Moreover, the framework is generic and can be extended to integrate more types of satellite data sources, further improving the quality of the fusion product. View Full-Text

47 OTHER INSTRUMENTATION↗
DOE OSTI2015

SU-E-T-138: Automated Chart Review Module Including Cross-Vendor Data Transfer Verification Developed for IHE-RO Initiative

Purpose: To follow the Integrating Healthcare Enterprise - Radiation Oncology (IHE-RO) initiative of proper cross-vendor technology integration, an automated chart checker (ACC) was developed. ACC compares extracted data from an approved patient plan in the Eclipse treatment planning system (TPS) against data existing in the Mosaiq treatment management system (TMS). ACC automatically analyzes these parameters using built-in quality checklists to provide further aid in chart review. Methods: Eclipse TPS data are obtained using Eclipse scripting API (ESAPI) while Mosaiq TMS data are obtained from a radiotherapy-treatment-planning (RTP) file. Using this information, ACC identifies TPS-TMS discrepancies in 18 primary beam parameters including MU, energy, jaw positions, gantry angle, table angle, accessories, and bolus for up to 31 beams. Next, approximately 40 items from traditional quality checklists are evaluated such as prescription consistency, DRR graticule placement, plan approval status, global max dose, and dose tracking coefficients. Parameters were artificially modified to determine if ACC would detect an error in data transfer and to test each component of quality checklists. Results: Using ESAPI scripting and RTP file-processing, ACC was able to properly aggregate data from TPS and TMS for up to 31 beams. Errors were artificially introduced into each plan parameter, and ACC was able to successfully detect all of them within seconds. Next, ACC was able to successfully detect mistakes in the chart by identifying deviations with its quality checklists, within seconds. Conclusion: ACC effectively addresses the potential issue of faulty cross-vendor data transfer, as described by IHE-RO. In addition, ACC was also able to detect deviations from its built-in quality checklists. ACC is already an invaluable tool for efficient and standardized chart review and will continue to improve as its incorporated checklists become more comprehensive.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

SU-E-T-143: Automatic PBS Treatment Planning for Prostate Cancer

Purpose: This study presents a novel idea of using automated planning technique for parallel-opposed prostate plans with pencil beam scanning proton therapy. Methods: A random 5 clinical prostate patients treated with parallel-opposed modulated-scanning proton beam in our proton center were selected in this study. All of the 5 cases were re-planned with in-house developed python scripts integrated into RayStation 4.0 clinical version. This automatic tool was designed to perform patient DICOM data import, external structure contours, CT density assignment, prescription to PTV assignment, beam placement, optimization, and dose calculation. Results: The mean time to generate a complete treatment plan was 1 minute per plan. For the automatically generated plans, 5 out of 5 plans (100%) were deemed clinically acceptable by one experienced radiation oncologist. In addition, the automated plans were overall dosimetrically equivalent to the clinical plans when scored for target coverage, rectum and bladder dose. Mann-Whitney U tests did not show significant dosimetric differences for normal tissue structures between the clinical and automated plans. Conclusion: We have developed a robust and automated method for fully inversed planned paralleled-opposed prostate proton treatment planning. This efficient method can be readily integrated into clinical practice. The tool produces clinically acceptable plans using only manually contoured anatomic structures and simulation CT data set. We anticipate that this tool will not only improve patient access to high-quality proton treatment by simplifying the planning process and will also reduce the effort and cost of incorporating more advanced planning into clinical practice in the near future.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2013

Expediting Groundwater Sampling at Hanford and Making It Safer - 13158

The CH2M HILL Plateau Remediation Company (CHPRC) manages the groundwater monitoring programs at the Department of Energy's 586-square-mile Hanford site in southeastern Washington state. These programs are regulated by the Resource Conservation and Recovery Act (RCRA), the Comprehensive Environmental Response Compensation and Liability Act (CERCLA), and the Atomic Energy Act (AEA). The purpose of monitoring is to track existing groundwater contamination from past practices, as well as other potential contamination that might originate from RCRA treatment, storage, and disposal (TSD) facilities. An integral part of the groundwater-monitoring program involves taking samples of the groundwater and measuring the water levels in wells scattered across the site. Each year, more than 1,500 wells are accessed for a variety of reasons. Historically, the monitoring activities have been very 'people intensive'. Field personnel or 'samplers' have been issued pre-printed forms showing information about the well(s) for a particular sampling evolution. This information is taken from two official electronic databases: the Hanford Well Information System (HWIS) and the Hanford Environmental Information System (HEIS). The samplers traditionally used these hardcopy forms to document the groundwater samples and well water-levels. After recording the entries in the field, the samplers turned the forms in at the end of the day and other personnel posted the collected information. In Automating Groundwater Sampling at Hanford (HNF-38542-FP Revision 0, Presented at Waste Management 2009 Conference, March 1 - March 5, 2009, Phoenix, AZ), we described the methods, tools, and techniques that would be used in automating the activities associated with measuring water levels. The Field Logging and Electronic Data Gathering (FLEDG) application/database that automates collecting the water-level measurement data has now been implemented at Hanford. In addition to eliminating the need to print out documents, the system saves three-to-four man days each month for the field personnel taking the measurements and the scientists and administrators managing the data and the documentation. After the information has received technical review, FLEDG automatically updates the database for water-level measurements and loads the document management system with the completed sampling report. Due to safety considerations, access to wells is conditional. A spreadsheet with appropriate data not only lists the wells that are cleared for work, but also the safety personnel who must be present before work can start. This spreadsheet is used in planning daily activities. Daily plans are structured to ensure that the wells to be sampled are cleared for work and the appropriate safety personnel have been assigned and are present before the work starts. Historically, the spreadsheets have been prepared manually, and as a result, are potentially subject to human error. However, a companion database application has been developed to work with FLEDG - making the entire sampling process more efficient and safer for personnel. The Well Access List - Electronic, WAL-E, is a database that contains much the same information that was previously manually loaded into the spread sheet. In addition, WAL-E contains a managed work-flow application that shows the access requirements and allows for appropriate reviews of the compiled well. Various CHPRC organizations, including Industrial Hygiene, RADCON, and Well Maintenance and Sample Administration are able to enter and review the wells added or deleted from the WAL-E database. The FLEDG system then accesses this database information to identify appropriate support personnel and provide safety requirements to field personnel. In addition, WAL-E offers the assurance that wells have appropriate locks and are correctly labeled and electrically grounded as required, before well activities begin. This feature is an extremely important aspect of the FLEDG/WAL-E system because it adds another safety check to the work evolution and reduces the potential for unsafe conditions to exist that would lead to a stop-work situation. This paper will further describe the technical aspects of both electronic programs/databases and their successful application at Hanford. (authors)

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2018

SCIP: a single-cell image processor toolbox

Abstract Summary Each cell is a phenotypically unique individual that is influenced by internal and external processes, operating in parallel. To characterize the dynamics of cellular processes one needs to observe many individual cells from multiple points of view and over time, so as to identify commonalities and variability. With this aim, we engineered a software, ‘SCIP’, to analyze multi-modal, multi-process, time-lapse microscopy morphological and functional images. SCIP is capable of automatic and/or manually corrected segmentation of cells and lineages, automatic alignment of different microscopy channels, as well as detect, count and characterize fluorescent spots (such as RNA tagged by MS2-GFP), nucleoids, Z rings, Min system, inclusion bodies, undefined structures, etc. The results can be exported into *mat files and all results can be jointly analyzed, to allow studying not only each feature and process individually, but also find potential relationships. While we exemplify its use on Escherichia coli, many of its functionalities are expected to be of use in analyzing other prokaryotes and eukaryotic cells as well. We expect SCIP to facilitate the finding of relationships between cellular processes, from small-scale (e.g. gene expression) to large-scale (e.g. cell division), in single cells and cell lineages. Availability and implementation http://www.ca3-uninova.org/project_scip Supplementary information Supplementary data are available at Bioinformatics online.

Martins, Leonardo (ORCID:0000000333512910)↗
DOE OSTI2017

Automatic Attribute Profiles

Morphological attribute profiles are multilevel decompositions of images obtained with a sequence of transformations performed by connected operators. They have been extensively employed in performing multi-scale and region-based analysis in a large number of applications. One main, still unresolved, issue is the selection of filter parameters able to provide representative and non-redundant threshold decomposition of the image. This paper presents a framework for the automatic selection of filter thresholds based on Granulometric Characteristic Functions (GCFs). GCFs describe the way that non-linear morphological filters simplify a scene according to a given measure. Since attribute filters rely on a hierarchical representation of an image (e.g., the Tree of Shapes) for their implementation, GCFs can be efficiently computed by taking advantage of the tree representation. Eventually, the study of the GCFs allows the identification of a meaningful set of thresholds. Therefore, a trial and error approach is not necessary for the threshold selection, automating the process and in turn decreasing the computational time. It is shown that the redundant information is reduced within the resulting profiles (a problem of high occurrence, as regards manual selection). The proposed approach is tested on two real remote sensing data sets, and the classification results are compared with strategies present in the literature.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Computer-aided pulmonary image analysis in small animal models

Purpose: To develop an automated pulmonary image analysis framework for infectious lung diseases in small animal models. Methods: The authors describe a novel pathological lung and airway segmentation method for small animals. The proposed framework includes identification of abnormal imaging patterns pertaining to infectious lung diseases. First, the authors’ system estimates an expected lung volume by utilizing a regression function between total lung capacity and approximated rib cage volume. A significant difference between the expected lung volume and the initial lung segmentation indicates the presence of severe pathology, and invokes a machine learning based abnormal imaging pattern detection system next. The final stage of the proposed framework is the automatic extraction of airway tree for which new affinity relationships within the fuzzy connectedness image segmentation framework are proposed by combining Hessian and gray-scale morphological reconstruction filters. Results: 133 CT scans were collected from four different studies encompassing a wide spectrum of pulmonary abnormalities pertaining to two commonly used small animal models (ferret and rabbit). Sensitivity and specificity were greater than 90% for pathological lung segmentation (average dice similarity coefficient > 0.9). While qualitative visual assessments of airway tree extraction were performed by the participating expert radiologists, for quantitative evaluation the authors validated the proposed airway extraction method by using publicly available EXACT’09 data set. Conclusions: The authors developed a comprehensive computer-aided pulmonary image analysis framework for preclinical research applications. The proposed framework consists of automatic pathological lung segmentation and accurate airway tree extraction. The framework has high sensitivity and specificity; therefore, it can contribute advances in preclinical research in pulmonary diseases.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

On the measurement of shape: With applications to lunar regolith

With the renewed commitment from NASA and other commercial entities for a presence on the Moon, the importance of understanding the characteristics of lunar regolith and how to utilize it have become the target of increasing scrutiny. Much of what is known about lunar regolith was collected during and immediately after the Apollo program, however, analytical techniques and instrumentation have advanced in leaps and bounds in the subsequent decades. Specifically, dynamic image analysis systems have advanced to the point that millions of particles can have morphological characteristics automatically determined in relatively short time frames. Particle morphological data was collected on several lunar samples and a pair of widely used regolith simulants to ascertain the accuracy of these simulants and to explore statistical analysis methods of these large datasets. It is found that these morphology datasets can vary widely depending on the particle size of the particles, and simple averaging of the data skews the results heavily towards the numerically abundant size ranges, the fines. Different reporting methods are suggested to ameliorate these problems. When applied to the lunar regolith, the particles are noted to be less morphologically complex than initially suspected. Compared to the lunar material, the simulants are found to contain some more morphological variability and angular grains. Such difference is likely due to the wildly different comminution processes that these different powder systems are subjected to.

2D shape↗
DOE OSTI2025

Hardware acceleration for HPS algorithms in two and three dimensions

We provide a flexible, open-source framework for hardware acceleration, namely massively-parallel execution on general-purpose graphics processing units (GPUs), applied to the hierarchical Poincaré–Steklov (HPS) family of algorithms for building fast direct solvers for linear elliptic partial differential equations. To take full advantage of the power of hardware acceleration, we propose two variants of HPS algorithms to improve performance on two- and three-dimensional problems. In the two-dimensional setting, we introduce a novel recomputation strategy that minimizes costly data transfers to and from the GPU; in three dimensions, we modify and extend the adaptive discretization technique of Geldermans and Gillman [1] to greatly reduce peak memory usage. We provide an open-source implementation of these methods written in JAX, a high-level accelerated linear algebra package, which allows for the first integration of a high-order fast direct solver with automatic differentiation tools. We conclude with extensive numerical examples showing our methods are fast and accurate on two- and three-dimensional problems.

Fast direct solvers↗
DOE OSTI2025

Mesoscale Cellular Convection Detection and Classification Using Convolutional Neural Networks: Insights From Long-Term Observations at ARM Eastern North Atlantic Site

Marine boundary layer clouds are crucial in Earth's climate system. They frequently manifest as closed or open cell mesoscale cellular convection (MCC). MCC clouds are challenging to represent accurately in current climate models, highlighting the need for detailed observational data sets and in-depth analyses. This study utilizes over 8 years of observations from the U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) User Facility Eastern North Atlantic (ENA) site at Graciosa Island, Azores, to investigate these clouds. We first apply a convolutional neural network with a U-Net architecture to classify open and closed cells, marking the first application of such an approach for automatically detecting MCC patterns from ground-based radar measurements. This method addresses some observational gaps in satellite data related to low temporal resolution, nighttime challenges, and limited vertical structure capture. The analysis of the MCC cases shows clear differences between closed and open MCCs: Closed MCC clouds are characterized by lower cloud tops and bases, shallower cloud geometrical depth, weaker horizontal wind speeds, stronger atmospheric stability, and a more homogeneous liquid water path than open MCCs. Finally, we demonstrate two potential applications of our radar-based MCC classifications: (a) facilitating the investigation of aerosol-cloud interactions and (b) exploring meteorological factors along with MCC's evolution by integrating satellite imagery and back-trajectory analysis. The identified MCC cases offer a valuable resource for the scientific community to study MCC processes further and improve climate model accuracy.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Automated nuclear cloud feature extraction from film

Chemical, biological, radiological, nuclear, and explosives incidents require rapid detection and characterization for appropriate response. For a nuclear detonation, visible-light cameras may be used to locate the cloud and characterize fallout deposition when coupled with numerical models. Films from the United States’ nuclear testing era compose the only sizeable collection of imagery depicting high-yield detonations. These films offer unique insights into characteristics of flows involving scales that are difficult to replicate experimentally, and they are a valuable source of data for the validation of models for nuclear fallout transport, either as part of emergency response or forensic activities. In this work, we implement modern computer vision and machine learning techniques to identify and track the cloud automatically and subsequently determine the time dependence of some of its features. We trained a ResNet-18 image classifier on hundreds of images to categorize nuclear cloud morphology. Each category or cloud regime is determined by early cloud evolution and is associated to constitutive properties of the flow, such as distribution of vorticity. Next, we identified keypoint features using the KAZE algorithm and tracked these keypoints in the images, allowing us to determine the dimensions and velocities of the cloud across film frames. These measurements converted to real-world units provide valuable experimental data that can be used in the development and validation of nuclear cloud models. We compared the results of this method against manual cloud rise measurements from two different films. In one, our automated method accelerated the feature extraction process without sacrificing measurement accuracy.

Khristy, Joel [ORNL] (ORCID:0000000209963060)↗
DOE OSTI2015

SU-E-I-58: Experiences in Setting Up An Online Fluoroscopy Tracking System in a Large Healthcare System

Purpose: Transitioning from a paper based to an online system for tracking fluoroscopic case information required by state regulation and to conform to NCRP patient dose tracking suggestions. Methods: State regulations require documentation of operator, equipment, and some metric of tube output for fluoroscopy exams. This information was previously collected in paper logs, which was cumbersome and inefficient for the large number of fluoroscopic units across multiple locations within the system. The “tech notes” feature within Siemens’ Syngo workflow RIS was utilized to create an entry form for technologists to input case information, which was sent to a third party vendor for archiving and display though an online web based portal. Results: Over 55k cases were logged in the first year of implementation, with approximately 6,500 cases per month once fully online. A system was built for area managers to oversee and correct data, which has increased the accuracy of inputted values. A high-dose report was built to automatically send notifications when patients exceed trigger levels. In addition to meeting regulatory requirements, the new system allows for larger scale QC in fluoroscopic cases by allowing comparison of data from specific procedures, locations, equipment, and operators so that instances that fall outside of reference levels can be identified for further evaluation. The system has also drastically improved identification of operators without documented equipment specific training. Conclusion: The transition to online fluoroscopy logs has improved efficiency in meeting state regulatory requirements as well as allowed for identification of particular procedures, equipment, and operators in need of additional attention in order to optimize patient and personnel doses, while high dose alerts improve patient care and follow up. Future efforts are focused on incorporating case information from outside of radiology, as well as on automating processes for increased efficiencies.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2022

Scalable algorithms for physics-informed neural and graph networks

Physics-informed machine learning (PIML) has emerged as a promising new approach for simulating complex physical and biological systems that are governed by complex multiscale processes for which some data are also available. In some instances, the objective is to discover part of the hidden physics from the available data, and PIML has been shown to be particularly effective for such problems for which conventional methods may fail. Unlike commercial machine learning where training of deep neural networks requires big data, in PIML big data are not available. Instead, we can train such networks from additional information obtained by employing the physical laws and evaluating them at random points in the space–time domain. Such PIML integrates multimodality and multifidelity data with mathematical models, and implements them using neural networks or graph networks. Here, we review some of the prevailing trends in embedding physics into machine learning, using physics-informed neural networks (PINNs) based primarily on feed-forward neural networks and automatic differentiation. For more complex systems or systems of systems and unstructured data, graph neural networks (GNNs) present some distinct advantages, and here we review how physics-informed learning can be accomplished with GNNs based on graph exterior calculus to construct differential operators; we refer to these architectures as physics-informed graph networks (PIGNs). We present representative examples for both forward and inverse problems and discuss what advances are needed to scale up PINNs, PIGNs and more broadly GNNs for large-scale engineering problems.

42 ENGINEERING↗
DOE OSTI2014

Modeling the Virtual Machine Launching Overhead under Fermicloud

FermiCloud is a private cloud developed by the Fermi National Accelerator Laboratory for scientific workflows. The Cloud Bursting module of the FermiCloud enables the FermiCloud, when more computational resources are needed, to automatically launch virtual machines to available resources such as public clouds. One of the main challenges in developing the cloud bursting module is to decide when and where to launch a VM so that all resources are most effectively and efficiently utilized and the system performance is optimized. However, based on FermiCloud’s system operational data, the VM launching overhead is not a constant. It varies with physical resource (CPU, memory, I/O device) utilization at the time when a VM is launched. Hence, to make judicious decisions as to when and where a VM should be launched, a VM launch overhead reference model is needed. The paper is to develop a VM launch overhead reference model based on operational data we have obtained on FermiCloud and uses the reference model to guide the cloud bursting process.

Garzoglio, Gabriele↗
DOE OSTI2017

An automatic input/output test experience of FPGA-based logic controller with I/O stimulator - 282

This paper describes an automatic testing approach instead of a traditional manual test for an FPGA-based Logic Controller (FLC). It mainly describes an automatic testing of FLC using an I/O stimulator with the initiated manual action. The FLC was developed by Doosan Heavy Industry and KAERI in S. Korea. We also developed a test-bed like I/O stimulator for a burning test based on the operational scenario. Signal sources from the I/O stimulator contain an analog input, an analog output, a digital input and a digital output, a programmable power supply, an RS232C serial interface, and a dual port Ethernet interface. LabVIEW Programming was used for the automatic testing. The FLC consists of a general processor module (FPM-01), a complex processor module (FPM-02), analog input modules (FAI-01), analog output modules (FAO-01), digital input modules (FDI-01), digital output modules (FDO-01), a data link module (FDL-01), and a network module (FNW-01). Among these, in this paper, automatic digital I/O testing and automatic analog I/O testing are described. In addition, for the accuracy of the automatic testing, two graphic recorders are used. One is the output results of general processor based testing, and the other is the output results of complex processor based testing. These are different company products. Redundancy recording was found in the test results. The test results were synchronized internally with the existing I/O stimulator and externally with two graphic recorders. Calibrations of all of the equipment were made before the system testing. Before the testing started, a test plan and the test procedure were set up in advance, and we performed test scenario-based automatic functional testing for the FLC as the qualification process. It can be several executed depending on the scenario and conditional parameters of the test. For validation of the accuracy and reliability in our automated test, manual testing was also performed. Manual tests were conducted in two ways. The first one was a manual test using Fluke 754 and Fluke 715. The other one was using embedded manual features of the I/O stimulator. By performing the automatic input and output testing using the I/O stimulator, the time and expense of FLC testing can be reduced significantly compared with a conventional manual test. We rechecked the automatic test results using Fluke754 manually. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

A Commercial Building Plug Load Management System that Uses Internet of Things Technology to Automatically Identify Plugged-In Devices and Their Locations

Plug and process loads (PPLs) account for a large portion of U.S. commercial building energy use. There is a huge potential to reduce whole building consumption by targeting PPLs for energy savings measures or implementing some form of plug load management (PLM). Despite this potential, there has yet to be a widely adopted commercial PLM technology. This paper describes the Automatic Type and Location Identification System (ATLIS), a PLM system framework with automatic and dynamic load detection (ADLD). ADLD gives PLM systems the ability to automatically identify devices as they are plugged into the outlets of a building. The ATLIS framework takes advantage of smart, connected devices to identify device locations in a building, meter and control their power, and communicate this information to a central database. ATLIS includes five primary capabilities: location identification, communication, control, energy metering, and data storage. A laboratory proof of concept (PoC) demonstrated all but the energy metering capability, and these capabilities were validated using a series of system tests. The PoC was able to identify when a device was plugged into an outlet and the location of the device in the building. When a device was moved, the PoC's dashboard and database were automatically updated with the new location. The PoC implemented controls to devices from the system dashboard so that devices maintained correct schedules regardless of where they were plugged in within the building. ATLIS's primary technology application is improved PLM, but other applications include asset management, energy audits, and interoperability for grid-interactive efficient buildings. An ATLIS-based system could also be used to direct power to critical devices, such as ventilators, during a brownout or blackout. Such a framework is an opportunity to make PLM more widespread and reduce the amount of energy consumed by PPLs in current and future commercial buildings.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Automated Waterbox Inspection for Nuclear Power Plants Using Computer Vision - Based Change Detection

Nuclear power plant waterboxes require regular inspection for leaks, missing components, and structural damage during maintenance outages. Traditional manual inspection is time-consuming and poses safety risks from confined space entry. We developed an automated computer vision system for drone-based waterbox inspection in partnership with Florida Light and Power. Our approach uses feature detection and matching to identify critical changes between baseline and current inspection images, automatically flagging additions (leaks/debris), removals (missing plugs), and translations (displaced components) while compensating for drone movement and environmental variations. We systematically evaluated six feature matching methods, from classical approaches (SIFT+BF) to state-of-the-art neural networks (SuperPoint+SuperGlue), using both standard benchmarks (HPatches) and waterbox-specific validation with real-world augmentations. SuperPoint+SuperGlue achieved superior performance with 7.82 pixels RMSE and 100% success rate—2.8x better accuracy than our baseline. While the pre-trained model has commercial licensing restrictions for nuclear deployment, our findings validate this architecture for custom training. We implemented a real-time GUI demonstrating the SIFT+BF approach for immediate deployment, processing drone feeds at 30 FPS with color-coded change visualization. Future work includes training a custom SuperPoint+SuperGlue model on waterbox data and integrating Vision-Language Models for automated reporting and maintenance guidance.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗
DOE OSTI2021

Alert Classification for the ALeRCE Broker System: The Light Curve Classifier

We present the first version of the Automatic Learning for the Rapid Classification of Events (ALeRCE) broker light curve classifier. ALeRCE is currently processing the Zwicky Transient Facility (ZTF) alert stream, in preparation for the Vera C. Rubin Observatory. The ALeRCE light curve classifier uses variability features computed from the ZTF alert stream and colors obtained from AllWISE and ZTF photometry. We apply a balanced random forest algorithm with a two-level scheme where the top level classifies each source as periodic, stochastic, or transient, and the bottom level further resolves each of these hierarchical classes among 15 total classes. This classifier corresponds to the first attempt to classify multiple classes of stochastic variables (including core- and host-dominated active galactic nuclei, blazars, young stellar objects, and cataclysmic variables) in addition to different classes of periodic and transient sources, using real data. We created a labeled set using various public catalogs (such as the Catalina Surveys and Gaia DR2 variable stars catalogs, and the Million Quasars catalog), and we classify all objects with ≥6 g-band or ≥6 r-band detections in ZTF (868,371 sources as of 2020 June 9), providing updated classifications for sources with new alerts every day. For the top level we obtain macro-averaged precision and recall scores of 0.96 and 0.99, respectively, and for the bottom level we obtain macro-averaged precision and recall scores of 0.57 and 0.76, respectively. Updated classifications from the light curve classifier can be found at the ALeRCE Explorer website (http://alerce.online).

47 OTHER INSTRUMENTATION↗