Search NASA⌕ Search

SEARCH · Search NASA

Results for “handwritten documents”

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.

Procedure Parsing: A Method for Parsing Handwritten Documents into Computer-Based Procedures

The nuclear industry is heavily procedure driven, where almost everything has a step-by-step instruction that is expected to be followed in detail. Historically, these procedures were printed on paper copies. Recently, the industry transitioned towards electronic copies (i.e., PDFs on tablets). One major drive for this transition is the introduction of human error and loss of situation awareness when using paper copies. However, electronic copies of documents inherently have the same error traps as their paper cousins. Therefore, there is an increased interest in a way to utilize the information in the step-by-step guidance, but to present it in a dynamic manner that guides the user and adapts to any encountered conditions. Researchers at Idaho National Laboratory propose a flexible, automated method based on document parsing and augmented by natural language processing (NLP) techniques, to address these shortcomings and capitalize on these recent advancements in machine learning. The proposed method provides a cost-effective solution for computer-assisted procedure parsing of hand-written control room procedures, originally authored in Word or PDF formats, into instructions that can be displayed as computer-based procedures (CBP) in a modern graphical user interface. The researchers devised, implemented and demonstrated the Operating Procedure Extender for Novel Systems (OPENS) method in 2020. The key to OPENS is to map the original procedure text into a context-free grammar, tying content to equipment, locations, and other steps, actions, etc. This formal grammar is then used to isolate and define keywords and actions verbs, such as “measure” or “evaluate” and tie them to specific equipment referenced within that step or located in other steps, substeps, actions, subactions and tables throughout the procedure. OPENS generates an abstract syntax tree from the document which it uses to store a copy of this information in the open-standard, machine-readable and human-readable file formats XML and JSON. The XML is useful to preserve the relational aspects of the procedure for referencing tables and branching information so the user can be directed to the next appropriate active step based on the values entered for that step and previous steps. The JSON is useful for storing and exchanging data objects used to track responses to previous steps and state changes in simulated environments. In future iterations, these formats can also be used for storing more detailed information about input during plant operation or simulation. The techniques the researcher developed could further be improved by integration of recent advancements in machine learning. NLP methods could standardize documents, correct for grammatical error, and provide automated semantic validation. The researcher expects that self-supervised techniques applied to collections of natural language instructions could strengthen the model with broader context. All these methods together give us a practical way to automatically extract protocols from documents and user interactions, empowering researchers, procedure writers and nuclear operators while moving the industry forward.

99 GENERAL AND MISCELLANEOUS↗

SharedCanvas: A Collaborative Model for Medieval Manuscript Layout Dissemination

In this paper we present a model based on the principles of Linked Data that can be used to describe the interrelationships of images, texts and other resources to facilitate the interoperability of repositories of medieval manuscripts or other culturally important handwritten documents. The model is designed from a set of requirements derived from the real world use cases of some of the largest digitized medieval content holders, and instantiations of the model are intended as the input to collection-independent page turning and scholarly presentation interfaces. A canvas painting paradigm, such as in PDF and SVG, was selected based on the lack of a one to one correlation between image and page, and to fulfill complex requirements such as when the full text of a page is known, but only fragments of the physical object remain. The model is implemented using technologies such as OAI-ORE Aggregations and OAC Annotations, as the fundamental building blocks of emerging Linked Digital Libraries. The model and implementation are evaluated through prototypes of both content providing and consuming applications. Although the system was designed from requirements drawn from the medieval manuscript domain, it is applicable to any layout-oriented presentation of images of text.

Computer Science↗

What Hansel and Gretel’s Trail Teach Us about Knowledge Management

Background At Idaho National Laboratory (INL), we are on the cusp of a significant era of change. INL is the lead Department of Energy Nuclear Research and Development Laboratory, focused on finding innovative solutions to the nation’s energy challenges. Not only has the Laboratory grown at an unprecedented rate over the last five years, but also has a significant segment of its workforce that is ready for retirement. Over the next 10 years, it is anticipated that upwards of 60% of the current workforce at INL will be eligible for retirement. Since the Laboratory is highly dependent on the intellectual capabilities of its scientists and engineers and their efforts to ensure the future of the nation’s energy portfolio, this attrition of resources has the potential of seriously impacting the ability of the Laboratory to sustain itself and the growth that it has achieved in the past years. Similar to Germany in the early nineteenth century, we face the challenge of our self-identity and must find a way to solidify our legacy to propel us into the future. Approach As the Brothers Grimm set out to collect their fairy tales, they focused on gathering information from the people that were most knowledgeable in the subject. For them, it was the peasants, with their rich knowledge of the region’s sub-culture of folk lore that was passed down from generation to generation around the evening fire. As we look to capture this tacit knowledge, it is requisite that we also seek this information from those individuals that are most versed in it. In our case, it is the scientists and researchers who have dedicated their lives to providing the nation with nuclear energy. This information comes in many forms, both digital and non-digital. Some of this information still resides in the minds of these scientists and researchers who are close to retirement, or who have already retired. Once the information has been collected, it has to be sorted through to identify where the “shining stones” can be found. The quantity of this information makes it improbable for an individual or set of individuals to sort through it and pick out those ideas which are most important. To accomplish both the step of information capture and classification, modern advancements in technology give us the tools that we need to successfully capture this tacit knowledge. To assist in this process, we have evaluated multiple tools and methods that will help us to unlock the power of tacit knowledge. Tools The first challenge that stands in the way of success is the capture of information. More than 50 years of nuclear research is captured in log books, microfiche, and other non-digital formats. To transform this information from its current form into a format that can “shine,” requires a number of different tools. These tools fall into three major categories: Information Capture, Content Retrieval, and Information Classification. Information Capture The first step is to capture the information from a myriad of sources. With knowledge existing in multiple formats, this step requires multiple approaches to be successful. Some of the sources that require consideration include handwritten documents, typed documents, microfiche, images, audio and video feeds, and electronic images. To make this step feasible for a large body of knowledge requires automation.

99 GENERAL AND MISCELLANEOUS↗

Automatic recognition system for document digitization in nuclear power plants

With the increasing number of data-driven models in nuclear applications, large volumes of numerical data are required to accurately model and predict the health status of a plant component. However, many historical operation logs that contain useful information are not fully utilized due to the lack of a systematic approach of digitization. To overcome this issue, this study proposes an automatic pipeline for extracting information from handwritten tabular documents collected from nuclear power plants. In our pipeline, we first denoise scanned documents with morphological operations, and then extract relevant parts from individual pages using both traditional computer vision and neural network methods. Handwriting recognition is applied to obtain text and numbers. As the most challenging step is how to crop only relevant information, the main focus of our paper is to detect tables and cells from scanned handwritten documents. Here we evaluate the efficiency and accuracy of our proposed method on handwritten operational reports obtained from a real-world case study. The results demonstrate the high accuracy and practicality of our proposed method.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Documenting the NASA Armstrong Flight Research Center Oblate Earth Simulation Equations of Motion and Integration Algorithm

A desire for more complete documentation of the National Aeronautics and Space Administration (NASA) Armstrong Flight Research Center (AFRC), Edwards, California legacy code used in the core simulation has led to this e ort to fully document the oblate Earth six-degree-of-freedom equations of motion and integration algorithm. The authors of this report have taken much of the earlier work of the simulation engineering group and used it as a jumping-o point for this report. The largest addition this report makes is that each element of the equations of motion is traced back to first principles and at no point is the reader forced to take an equation on faith alone. There are no discoveries of previously unknown principles contained in this report; this report is a collection and presentation of textbook principles. The value of this report is that those textbook principles are herein documented in standard nomenclature that matches the form of the computer code DERIVC. Previous handwritten notes are much of the backbone of this work, however, in almost every area, derivations are explicitly shown to assure the reader that the equations which make up the oblate Earth version of the computer routine, DERIVC, are correct.

integrators↗

Standard classification of software documentation

General conceptual requirements for standard levels of documentation and for application of these requirements to intended usages. These standards encourage the policy to produce only those forms of documentation that are needed and adequate for the purpose. Documentation standards are defined with respect to detail and format quality. Classes A through D range, in order, from the most definitive down to the least definitive, and categories 1 through 4 range, in order, from high-quality typeset down to handwritten material. Criteria for each of the classes and categories, as well as suggested selection guidelines for each are given.

Tausworthe, R. C.↗

Water Quality Sampling Locations Along the Shoreline of the Columbia River, Hanford Site, Washington

As environmental monitoring evolved on the Hanford Site, several different conventions were used to name or describe location information for various sampling sites along the Hanford Reach of the Columbia River. These methods range from handwritten descriptions in field notebooks to the use of modern electronic surveying equipment, such as Global Positioning System receivers. These diverse methods resulted in inconsistent archiving of analytical results in various electronic databases and published reports because of multiple names being used for the same site and inaccurate position data. This document provides listings of sampling sites that are associated with groundwater and river water sampling. The report identifies names and locations for sites associated with sampling: (a) near-river groundwater using aquifer sampling tubes; (b) riverbank springs and springs areas; (c) pore water collected from riverbed sediment; and (d) Columbia River water. Included in the listings are historical names used for a particular site and the best available geographic coordinates for the site, as of 2009. In an effort to create more consistency in the descriptive names used for water quality sampling sites, a naming convention is proposed in this document. The convention assumes that a unique identifier is assigned to each site that is monitored and that this identifier serves electronic database management requirements. The descriptive name is assigned for the convenience of the subsequent data user. As the historical database is used more intensively, this document may be revised as a consequence of discovering potential errors and also because of a need to gain consensus on the proposed naming convention for some water quality monitoring sites.

54 ENVIRONMENTAL SCIENCES↗

Generative large language models for predictive maintenance planning

Maintenance planning and the generation of necessary components for tasks can prove time-consuming and complex. Automating the creation of recurring or similar tasks by leveraging previous planning packages and data, while uncovering insights to automate planning package generation, presents an opportunity to conserve valuable time and resources. This work aims to harness the textual and probabilistic capabilities of large language models (LLMs) to automate the generation of planning packages. Utilizing diverse data sources ranging from raw data to handwritten text, both singular and collaborative LLMs are trained and tested. Results demonstrate their capability to generate essential planning package components, effectively replicating the statistical patterns in the data. This demonstrates the use of these tools inside a digital asset for automated planning. This work outlines a methodology for constructing datasets, a training suite, and evaluation methods for LLM-based textual and conversational planning tools utilized in an asset digital twin. Results indicate that the fine-tuned models generate estimated planning information within the statistical ranges observed in real maintenance data. The models achieve high accuracy (>90%) in document question-answering and instruction generation tasks. Furthermore, the conversational retrieval-augmented generation (RAG) assistant system achieves 100% document retrieval accuracy, while conversational information capture exceeds 98% across the majority of work-package assistant modules.

97 MATHEMATICS AND COMPUTING↗

The use of ion mobility spectrometry and gas chromatography/mass spectrometry for the detection of illicit drugs on clandestine records

Illicit drug distribution has over the past decade grown tremendously from simple 'drug pushing' where drugs were distributed from poorly organized individuals to today's well organized and well financed drug cartels. This change to a more 'corporate-like' atmosphere has resulted in a greater use of record keeping to monitor the profits generated. The use of record keeping by drug distributors is not restricted to high level drug smugglers but is used at all levels within the distribution network. Dealers at all levels including street dealers are generally 'fronted', given on consignment quantities of drugs that they in turn sell to customers, thereby requiring the need for records to keep track of drug sales versus liabilities. These records because of their illicit nature are often encrypted to hide the fact that they are indeed records of drug transactions. The creation of a handwritten notation concerning a drug transaction is normally brought on because of a purchase or sale. In a sale, this is commonly accomplished through a consignment, or the designation of a quantity to a customer to whom that amount has been 'fronted'. Because this activity generates a debt, it follows that an accounting for payments made, as well as new transactions completed, is only logical. One of the most common means of representing these is through an 'accounting flow', in which payments are subtracted from a running balance while new sales are added to it. The examination of illicit drug records has been the key to the prosecution of numerous federal, state, and local drug cases for a number of years. The Document Section of the FBI Laboratory, through its Racketeering Records Analysis Unit (RRAU), has been involved in such analytical efforts since 1983. Detailed analytical research brought about an evolution in the systematic approach utilized in the RRAU since that time. The close proximity of the drugs to the records often results in trace drug evidence being transferred to the records. The detection of trace drug residue on surfaces by ion mobility spectrometry (IMS) is well documented in literature. The following procedure will deal primarily with the newer techniques of trace drug analysis and drug record analysis developed by the Chemistry/Toxicology Unit of the FBI Laboratory since the more traditional techniques of latent finger print analysis and document analysis are well known.

Donnelly, Brian↗

Automated metadata, provenance cataloging and navigable interfaces: ensuring the usefulness of extreme-scale data

The MPO (Metadata, Provenance, Ontology) Project successfully addressed the goal of improving the usefulness and traceability of scientific data by building a system that could capture and display all steps in the process of creating, analyzing and disseminating that data. Throughout history, scientists have generated handwritten logbooks to keep track of data, their hypotheses, assumptions, experimental setup, and computational processes as well as reflections on observations and issues encountered. Over the last several decades, with the growth of personal computers, handheld devices, and the World Wide Web, the handwritten logbook has begun to be replaced by electronic logbooks. This transition has brought increased capability such as supporting multi-media, hypertext, and fast searching. However, content creation and metadata (a set of data that describes and gives information about other data) capturing has for the most part remained a manual activity just as it was with handwritten logbooks. This has led to a fragmentation of data, processing, and annotation that has only accelerated as scientific workflows continue to increase in complexity. From a scientific perspective, it is very important to be able to understand the lineage of any piece of data: who, what, when, how, and why. This is typically referred to as data provenance. The fragmentation discussed previously often means that data provenance is lost. As scientific workflows move to powerful computers and become more complex, the ability to track all of the steps involved in creating a piece of data become even more difficult. It was the goal of the MPO (Metadata, Provenance, Ontology) Project to create a system (the MPO System) that allows for automatic provenance and metadata capturing in such a way to allow easy searching and browsing. This goal needed to be accomplished in a general way so that it may be used across a broad range of scientific domains, yet allow the addition of vocabulary (Ontology) that is domain specific as is required for intelligent searching and browsing in the scientific context. Through the creation and deployment of the MPO system, the goals of the project were achieved. An enhanced metadata, provenance, and ontology storage system was created. This was combined with innovative methodologies for navigating and exploring these data using a web browser for both experimental and simulation-based scientific research. In addition, a system to allow scientists to instrument their existing workflows for automatic metadata and provenance is part of the MPO system. In that way, a scientist can continue to use their existing methodology yet easily document their work. Workflows and data provenance can be displayed either graphically or in an electronic notebook format and support advanced search features including via ontology. The MPO system was successfully used in both Climate and Magnetic Fusion Energy Research. The software for the MPO system is located at https://github.com/MPO-Group/MPO and is open source distributed under the Revised BSD License. A demonstration site of the MPO system is open to the public and is available at https://mpo.psfc.mit.edu/. A Docker container release of the command line client is available for public download using the command docker pull jcwright/mpo-cli at https://hub.docker.com/r/jcwright/mpo-cli.

97 MATHEMATICS AND COMPUTING↗