DT PixelFlow: Artificial Intelligence and Application

March 23, 2021 | by Hannah Storch

DT PixelFlow Artificial Intelligence

Artificial Intelligence (AI) has great implications for cultural heritage preservation. Gathering metadata for a collection can be time-consuming and labor-intensive, often involving the individual knowledge and experience of one person. Without this identifying descriptive metadata, valuable information can be lost and collections can remain incomplete. AI can be used as part of a metadata workflow to reduce the cost and tediousness of enriching a collection with enhanced metadata records.

Along with creating digital preservation-grade derivatives and deliverables, DT PixelFlow can use artificial intelligence to describe an item’s material, text, and image content. This type of descriptive data extraction allows not only for the leveraging of existing assets but also for the salvaging of descriptive metadata information before it, or the context required to create it is lost to time and memory.

Material

While the material type is a common metadata field, it is often automatically generated with more generic information, such as text or film. By implementing AI analysis, DT PixelFlow has the ability to automatically suggest the material type and item categories with greater depth of detail. This is a capability we are currently developing and exploring as part of the PA ArCHER Grant with Smithsonian Center for Folklife and our partners RIVERai.  The goal is to have DT PixelFlow automatically determine the types of documents, such as pieces of correspondence or legal briefs, and then further categorize them into groups such as memos, contracts, or letters. Learn more about the PA ArCHER Grant and its progress here.

We are very excited to explore this capability with other institutions and collections. If you have a collection you think would benefit from large-scale automatic material-type description using AI, with human QC, please contact us.

Text

Along with providing Optical Character Recognition (OCR), DT PixelFlow is able to provide a deeper analysis of the text and written content of an image to provide valuable contextual information. By using entity extraction, DT PixelFlow provides context to information that might otherwise seem like unconnected data, such as recognizing an address, formulaic greeting, or date from a string of numbers and text. Similarly, this type of entity analysis can find known entities such as proper names, which could enable an institution to successfully search for and gather together all of the images relating to a particular person or place.

DT PixelFlow’s AI analysis is not only able to recognize and transcribe the text within an image but also understand the conveyed sentiment and style of the text, interpreting the emotions, such as positive and negative or happy and sad, behind them.

These kinds of deeper analyses are set up on a project-by-project basis to ensure the analysis is relevant to the collection, the institution, and the stakeholders of the results. If you think your collection might benefit from AI analysis of the structure, content, or sentiment of the OCR’d text, please contact us for a consultation and we’ll help you understand what is possible and, just as importantly, what is practical.

Image

Artificial intelligence can identify objects and individuals inside of photographic or pictorial collections. With digital records, if this information is not extracted, cataloged, and linked to the image, this descriptive information can be lost to volume – obscured by the sheer scale of images one might have to look through manually to find given content. We can provide object detection and/or face detection in DT PixelFlow. This enables us to isolate and identify objects that are both in the foreground and less prominent in images.

Object Detection with Artificial Intelligence
Object Detection: 2 tin cans, 3 people, 1 horse, 1 truck
Image courtesy of the Andy Warhol Photography Archive, Cantor Arts Center, Stanford University
Object Detection with Artificial Intelligence
Object Detection: 10 plates
Image courtesy of the Andy Warhol Photography Archive, Cantor Arts Center, Stanford University

DT PixelFlow also has the ability to identify and categorize general objects, locations, and information using keywords. This information can be general or tied to a specific institution or collection. For example, keywords for a collection of slides belonging to a natural history museum could have more refined and accurate metadata with keywords pertaining to that particular type of collection, location, or scientific study. Within the image, DT PixelFlow is able to recognize and extract both big depictions and small details, from the recognition of landmark features to facial features and human emotion. If there are specific individuals of interest, we can even train DT PixelFlow to automatically identify the faces.

Packaging

Once the descriptive metadata has been derived through AI analysis, it can be packaged in many different formats to make it more accessible to the user and institution. After it has been interpreted, the information can be embedded into the final image file, ensuring that the data is always linked to the image and that they can be updated together in the future, or output in other formats. For usability, this descriptive data could also be generated in a txt file, document format, or included in a new or existing spreadsheet. To learn more about how to make the most of your metadata, check out our recent metadata article.

Impact

Traditionally the accumulation of descriptive metadata has been a specialized as well as a tedious and labor-intensive process. With AI analysis and application, Pixel Acuity is able to maintain a high level of accuracy while increasing efficiency and accessibility, and as needed we can leverage our highly skilled staff to provide human QC on top of the automatic detection provided by our AI Combining our in-house software, our experience in cultural heritage collections, and our talented team, we are now able to assist in the preservation of collections through descriptive categorization and contextualization derived from AI research and analysis as well as digital surrogacy.


Contact us to learn more about our digital imaging services and how we can bring artificial intelligence to your workflow.

Managing and Mastering Metadata with DT PixelFlow

January 21, 2021 | by Hannah Storch

For cultural heritage professionals, metadata provides invaluable descriptive information about an object or resource, but it is also time-consuming to accumulate. Creating and maintaining metadata for a collection is an integral part of taking cataloging one step further and creating a digital collection. Metadata provides context for an item within a collection and can either be embedded in the digital file at or after the time of creation or maintained in a centralized location such as a database, DAM, CMS, or spreadsheet. Using RAW rapid capture imaging with metadata embedding capabilities and DT PixelFlow, Pixel Acuity has been able to automate much of the metadata creation process and create workable metadata formats for institutions. This helps us both reduce the cost of, and increases our accuracy and flexibility in, providing metadata services.

Defining Metadata

For the purpose of institutions such as libraries and archives, metadata can be categorized into four basic types: administrative, descriptive, preservation, and technical. Administrative metadata provides the provenance context information necessary to understand the information resources, such as past ownership and from where the resource came. Descriptive metadata describes a resource, its context, and identifying characteristics so that people can locate and search for the asset using subjects and keywords. Preservation metadata is the conservation information that can be used to protect the original resource from deterioration or degradation.  Finally, technical metadata is the information about the digital file that can allow the resource to be identified. When collecting metadata, institutions are faced with limited resources for staff, time, and funding, often having to choose between feasible or “good enough” metadata and comprehensive metadata. We, at Pixel Acuity, are able to use our DT PixelFlow scripting to alleviate some of the burdens of that choice, offering options for embedding and generating metadata to create and enhance institutional records.

DT PixelFlow Metadata Capabilities

Since technical metadata is information about the digital asset that is created during the collection imaging process, we are able to capture that information and embed it in the file derivatives themselves. Embedding this information ensures that it will not be lost, automatically links the data and the metadata, and ensures that the image and the metadata will be updated together. Typically this information is linked with the TIFF derivative file after being captured in the RAW but with DT PixelFlow we are able to embed it in almost all derivative formats including TIFFs, JPEGs, PDFs, and PDF/As. Technical metadata information can also be extracted and put into a spreadsheet, which can simplify data management and facilitate search and retrieval. Using DT PixelFlow we are able to generate spreadsheets containing this information in multiple formats, including Dublin Core, based on client preference in order to facilitate data retrieval and storage in accordance with their own institution’s Digital Asset Management System (DAMS) or collection information storage/organization type.

Mapping information from the IPTC metadata fields to the file derivative
Excerpt from a Dublin Core Metadata spreadsheet

Along with creating spreadsheets and ingesting metadata, DT PixelFlow can be used to enrich existing records or create a more holistic record, combining information from original inventories and documents with new information obtained at the time of capture and digital content creation. This includes the generation of basic descriptive metadata such as an object type or category, transcription of annotations on an item or its container/folder/box, and non-subjective evaluations such as page count. More advanced descriptive metadata information can require both organization-specific and subject-specific expertise. While it is not possible for us to offer such subject-specific expertise for every collection that we digitize, we are able to utilize and enhance records created by specialists. If a client is able to provide us with an inventory with existing information, such as an inventory spreadsheet or an XML format of a finding aid, we can extract information from that format and create a new record, with that information as well as the technical information about the digital asset obtained during the imaging process. This allows us to combine records to give a more comprehensive understanding of a resource, its place within a collection, and how it relates to the digital asset.

An example of descriptive information extracted from an xml format finding aid provided by the client

Case Study Featuring Metadata Mastery

Using our DT PixelFlow scripting, we are able to automate the often arduous processes of metadata embedding and creation, minimizing cost and labor for the institution and allowing it to focus on other aspects of digital collection creation. In just one example, prior to digitization, one of our clients had a cataloging inventory with folder-level descriptive information that listed the location information for the assets, such as box and folder number, as well as descriptive information including title, dates, location, collection, and series. Not only were we able to use our tools to embed that information into the files, effectively linking the original object to the digital asset, but we were also able to generate information about the filename of the digital asset, the number of assets for each folder, and generate checksum lists and titles for each folder. This additional information was added to the original inventory, giving a more holistic view of the original item and the digital asset within the collection and linking the two within our client’s DAMS.

Information from the client provided for PA to embed in the files
Information from PA provided to the client in spreadsheet format

Artificial Intelligence & Metadata

Pixel Acuity is also at the forefront of leveraging artificial intelligence to assist metadata workflows. We are working with the Smithsonian Center for Folklife and Cultural Heritage (recipient of our DT ArCHER Grant) to evaluate the effectiveness and accuracy of these methods. This effort deserves its own article, which we will publish later this year. For now, we can say that we don’t expect AI to be a magic wand that replaces expertise, experience, and careful execution, but we do expect it will be an enormously useful tool.

Summary

While metadata generation can be costly in terms of time, labor, and resources, it is crucial to capturing the context of items within a digital collection. Metadata is what allows scholars and researchers to search a collection for specific information and allows registrars, cataloguers, and collections managers to organize their data and collection information. With DT PixelFlow automation, we are able to effectively and efficiently assist our clients to have integrated metadata records, so that they do not have to sacrifice quantity or quality.

Ready to Learn More About Mastering Metadata?

We can help your next project be a breeze! Learn more about DT PixelFlow, project planning, additional services, and pricing by contacting us here.

Shining Light On Film Scanning

December 10, 2020 | by Hannah Storch

Shining Light On Film Scanning Banner

Preservation grade film scanning is no simple task, and it becomes considerably more complex in mass digitization projects with large collections. Transmissive materials (a catch-all term used to collectively refer to film, glass plates, and any other media designed to be viewed in front of a light source) present many obstacles in handling and imaging not found with reflective media, and there are other considerations in terms of digitization method and final image rendering. Pixel Acuity has spent the better part of a decade perfecting film scanning workflows that optimize efficiency, fulfill each client’s unique goals, and conform to the highest image quality standards.

Film Scanning Hardware

Challenges

Film collections are often in a delicate physical state and are susceptible to many types of physical deterioration. Film can degrade in many ways: delaminating, becoming brittle, distorting, and fading to name just a few. All of these factors result in the need for conservation-grade handling and extra attentive care during imaging, especially during rapid capture in mass digitization efforts. 

Because film must be handled with the utmost care, digitization workflows frequently require additional staff beyond an imaging technician/photographer, and once the film has been imaged, there are still many decisions to make regarding the presentation of the film, all of which require in-depth knowledge of software settings, workflows, and processing steps. Clients may want film presented as it appears to the eye, or want negative items converted into positive images and positive images color corrected. 

At Pixel Acuity, our team of experts uses their extensive knowledge and experience to resolve these issues and create the highest-quality preservation-grade digital surrogates.


Pixel Acuity Film Scanning
Pixel Acuity staff digitizes a negative.

Solutions

In order to provide the best care possible for the film during the digitization process, Pixel Acuity follows the same conservation principles that are used for in-person viewing. All working surfaces are cleaned on a regular basis and the trained object handlers handle the material with care and wear conservator-approved gloves.

In order to minimize potential damage or scratching of the emulsion of the film, Pixel Acuity uses film carriers, such as the Digital Transitions (DT) magnetic or glass carriers, that make minimal or no contact with the emulsion (pictured above). These carriers also help deal with physically distorted material, hand-cut film, and materials of differing thickness, such as glass plates and lantern slides.

Over years of working in the cultural heritage imaging space, Pixel Acuity has perfected imaging workflows for film, moving quickly, efficiently, and safely through the digitization process. By implementing these workflows, we are able to digitize transmissive material at an unparalleled rate, imaging approximately 2,500 35mm slides or 3,200 strips of film a day.

Pixel Acuity Film Scanning Quality Control

Using our extensive knowledge base, Pixel Acuity’s skilled imaging technicians are able to render film according to the client’s specifications and needs: either object reproduction, content reproduction, or speculative artist’s rendering.

Object reproduction imaging is a faithful reproduction of the entire physical object, as it would appear to the eye on a light table. 

Content reproduction involves producing a human-readable version of the image contained within the object, for example, a negative converted to a positive image, or a contrast adjusted version of a faded positive image. Color negative conversion is a particularly challenging task, with no one-size-fits-all solution. However, Pixel Acuity has developed several proprietary conversion methods born from extensive research and experience in the darkroom that provide excellent positive “print” image files from color negatives of all types.

A speculative artist’s rendering involves more creative license and agency on the part of the imaging technician as they attempt to recreate the image as they imagined the artist would have wanted their final product to look. This rendering method can produce final images that counteract the effect of years of age on the film itself and produce an image that is reminiscent of how the original film was most likely intended to look. For this type of bespoke imaging work, Pixel Acuity works with clients to research how the artist might have wanted the image represented to ensure accuracy in the alterations.


Glass Plate Digitized by Pixel Acuity

We Can Help With Your Collection

Pixel Acuity’s extensive experience in digitizing transmissive materials, our knowledgeable object handlers and photographers on staff, and our use of the latest imaging equipment and technological tools in the industry makes us one of the leading authorities on film scanning. 

Working with collections around the world, for institutions such as the Smithsonian Institution, The Getty, and so many others, Pixel Acuity has created digitization workflows that combat the challenges of such a potentially tricky material while optimizing efficiency, quality, and preservation.

To learn more about how Pixel Acuity and Digital Transitions can help you with digitization services, software, and consultations, please contact us.

Looking for more film scanning resources? Check out this new Film Scanning Knowledge Center by DT Cultural Heritage here.

Glass plate negative (right) picturing Abraham Lincoln was taken by Mathew Brady and was digitized by Pixel Acuity for the National Portrait Gallery.

The Word On Optical Character Recognition With DT PixelFlow

At The Phillips Collection Archive

November 19, 2020 | by Hannah Storch

DT PixelFlow

Pixel Acuity has offered the cultural heritage community unparalleled imaging and digitization services for the better part of a decade. Recently, we have added new automations and related offerings to our repertoire. One of the most impactful innovations in Cultural Heritage imaging technology has been the ability to use the next-generation Optical Character Recognition (OCR) in our DT PixelFlow software to turn typed and handwritten documents into searchable text. Pixel Acuity is now not only able to generate the highest quality digital images for cultural heritage collections but also to create searchable texts for the researchers and scholars who access these collections, revolutionizing the way that they conduct research.

The Phillips Collection Archive

One of our ongoing projects that leverages DT PixelFlow’s OCR capabilities is our project with The Phillips Collection Archive in Washington, DC. The Phillips Collection houses modern and contemporary art, while The Phillips Collections Archive contains materials pertaining to the museum’s founding director, Duncan Phillips, and his wife Marjorie. The Archive holds materials documenting the purchase of important pieces of modern and contemporary art from the 1920s to present acquisitions. The current project consists of digitizing  approximately 100,000 personal photographs and correspondence, pamphlets, and documents relating to the family and their work with various directors, artists, and galleries. By using DT PixelFlow’s OCR capabilities, The Phillips Collection Archive is able to transform their collection of typed and hand-written material into fully-searchable documents.


Optical Character Recognition (OCR) Application and Process

For our project with The Phillips Collection Archive, we are able to implement our OCR technology to create two different types of readable and searchable text files from our digital images – PDF/As and .txt files. We start by capturing the highest-quality and most consistent images of the material – the better the input the better the output – so we surpass preservation-grade digitization standards such as Metamorfoze-strict, FADGI 4-star, and ISO 19264 using RAW rapid capture photography to capture digital images. This enables us to preserve all of the information recorded by the camera sensor at the time of capture without applying compression or losing any information.

Once all of the images have been captured in the RAW format, they are ready to be run through DT PixelFlow in order to create the OCR’d derivatives. Due to our modern machine-learning approach, we are able to generate highly-accurate OCR’d text in multiple languages and output formats.  We also have the flexibility to create a controlled, topic-specific vocabulary, depending on the needs of the collection, which can be used to further increase the specificity and accuracy of the resulting text.

The resulting data learned during the machine OCR process is then encoded into an hOCR file, which can then be converted into the deliverables requested by the client. Our unique approach enables us to offer a wide range of deliverables, including but limited to, PDF, PDF/A, a METS/ALTO sidecar xml, and txt files.


An example of searchable handwritten text within OCRed material.

Derivatives and Deliverables

Since The Phillips Collection Archive aims to make the documents and correspondence of Duncan and Marjorie Phillips more accessible to researchers and scholars, they have opted for both PDF/As and txt files. The PDF/A format layers the OCR’d text over the image of the object and produces a document that researchers can use to search on their own devices and see matches in their original visual context, in the document itself (examples of typed and handwritten applications are pictured above). The txt file (one example is pictured right) extracts the text from the image and creates a separate file format, which can be utilized by other institutional systems such as text-analysis tools or word-cloud generators. The choice of these OCR’d deliverables, along with highest-quality preservation-grade digital images, will allow researchers to delve deeper into The Phillips Collection Archive and learn more about the history of the Museum and the relationships that formed its foundations. While it may have taken hours of painstaking research to further explore the relationship between the Phillips Collection and The American Federation of the Arts, with a simple keyword search, a researcher can now find all of the documents, both typed and handwritten, pertaining to the Federation or The Phillips with a click of a button. 


Customized Innovation

It is opportunities and projects like these that allow Pixel Acuity, as a company, to innovate new workflows and adapt new technologies to give our clients the best possible digitization services and imaging experience. We continue to promote advancements, such as machine-learning-powered OCR, within the cultural heritage community because, the bottom line is that the best deserves the best.

To learn more about how Pixel Acuity and Digital Transitions can help you with digitization services, software, and consultations, please contact us.

How Can Pixel Acuity Help Your Digitization Program In the Age of Social Distancing?

There are many unknowns in the world today and Cultural Heritage institutions everywhere are facing the same challenge – how to best serve the public and their mission while keeping their employees safe and healthy. We heard this first hand from many institutions during and following our recent webinar titled “Digitization in the Age of Social Distancing.” When coronavirus (COVID-19) struck, these institutions that are responsible for promoting the arts, history, and culture had to close their doors to the very public they serve. In order to function in the “new normal,” people became increasingly reliant on technology, using it not only as their primary means of gathering information but also interacting with the world. With this growing dependence on online and remote access, collection digitization and digital preservation have proven even more vital than ever before. At this moment in time, we as a Cultural Heritage community can come together, reaching the world and the public in new ways through collection digitization and online publication.


With decades of combined hands-on experience imaging cultural heritage in all its forms, Pixel Acuity offers highly specialized knowledge of the inner workings of large-scale digitization efforts as well as customizable workflows and production solutions to fit your institution’s individual needs. By working with you in assessing your individual digital imaging needs, Pixel Acuity can provide a tailored digitization workflow plan for either on-site or off-site digitization that will maximize efficiency while implementing OSHA and CDC guidelines to ensure a safe working environment for everyone involved. We are committed to providing the same high standard of care and production quality whether on-site or off-site. In both scenarios, we are dedicated to applying archival-quality methodologies and Federal Agencies Digital Guidelines Initiative (FADGI) standards to our imaging of collections to create preservation-grade images of the highest possible resolution and quality control.


Although some parts of the world are beginning to reopen to varying degrees, Cultural Heritage institutions are experiencing a “new normal” with reduced on-site staff and hours spent in-person with physical collections. We at Pixel Acuity understand that when every in-person hour counts, priorities have to be re-evaluated in order to optimize time spent with the collection. Within the digitization workflow, physical preparation and digitization of materials requires hands-on work with the physical collection, while tasks such as metadata entry, post-processing, and quality control (QC) can be completed remotely. Pixel Acuity provides both on-site and off-site digital imaging services to the Cultural Heritage community. We are able to install state of the art photographic equipment – developed by Digital Transitions – on-site at Cultural Heritage institutions, as well as provide highly qualified imaging technicians to implement digitization workflows and create high-resolution digital images. We also are able to arrange for off-site digitization production at any of our production facilities in Chantilly, Virginia; New York City, New York; or Los Angeles, California. With this service, Cultural Heritage institutions can send their collections to one of our facilities, where our certified art/object handlers and imaging technicians can digitize the collections and then return the digital files and physical collections back to the original institution. This allows Cultural Heritage institution employees to minimize the amount of time they have to spend on-site with the collection during the digitization process, and enables us to provide them with digital files that  can be worked on remotely post-digitization. Along with providing imaging technicians and object/art handlers for digitization production, Pixel Acuity is also able to supply qualified staff to assist with collection processing duties such as rehousing collections, barcode application, and metadata creation.  


Like everyone in the Cultural Heritage community, we at Pixel Acuity understand how difficult it is to serve the community and our clients in these uncertain times, and we are working harder than ever to provide customizable imaging solutions to help institutions reach their digitization goals. To learn more about our services or to obtain more information, please contact us.