skip to main content


Title: Computational metadata generation methods for biological specimen image collections
Metadata is a key data source for researchers seeking to apply machine learning (ML) to the vast collections of digitized biological specimens that can be found online. Unfortunately, the associated metadata is often sparse and, at times, erroneous. This paper extends previous research conducted with the Illinois Natural History Survey (INHS) collection (7244 specimen images) that uses computational approaches to analyze image quality, and then automatically generates 22 metadata properties representing the image quality and morphological features of the specimens. In the research reported here, we demonstrate the extension of our initial work to University of the Wisconsin Zoological Museum (UWZM) collection (4155 specimen images). Further, we enhance our computational methods in four ways: (1) augmenting the training set, (2) applying contrast enhancement, (3) upscaling small objects, and (4) refining our processing logic. Together these new methods improved our overall error rates from 4.6 to 1.1%. These enhancements also allowed us to compute an additional set of 17 image-based metadata properties. The new metadata properties provide supplemental features and information that may also be used to analyze and classify the fish specimens. Examples of these new features include convex area, eccentricity, perimeter, skew, etc. The newly refined process further outperforms humans in terms of time and labor cost, as well as accuracy, providing a novel solution for leveraging digitized specimens with ML. This research demonstrates the ability of computational methods to enhance the digital library services associated with the tens of thousands of digitized specimens stored in open-access repositories world-wide by generating accurate and valuable metadata for those repositories.  more » « less
Award ID(s):
1940233
NSF-PAR ID:
10383304
Author(s) / Creator(s):
; ; ; ; ; ;
Editor(s):
Adam, N.; Neuhold, E.; Furuta, R.
Date Published:
Journal Name:
International Journal on Digital Libraries
Volume:
23
Issue:
4
ISSN:
1432-5012
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Conference Title: 2021 ACM/IEEE Joint Conference on Digital Libraries (JCDL) Conference Start Date: 2021, Sept. 27 Conference End Date: 2021, Sept. 30 Conference Location: Champaign, IL, USAMetadata are key descriptors of research data, particularly for researchers seeking to apply machine learning (ML) to the vast collections of digitized specimens. Unfortunately, the available metadata is often sparse and, at times, erroneous. Additionally, it is prohibitively expensive to address these limitations through traditional, manual means. This paper reports on research that applies machine-driven approaches to analyzing digitized fish images and extracting various important features from them. The digitized fish specimens are being analyzed as part of the Biology Guided Neural Networks (BGNN) initiative, which is developing a novel class of artificial neural networks using phylogenies and anatomy ontologies. Automatically generated metadata is crucial for identifying the high-quality images needed for the neural network's predictive analytics. Methods that combine ML and image informatics techniques allow us to rapidly enrich the existing metadata associated with the 7,244 images from the Illinois Natural History Survey (INHS) used in our study. Results show we can accurately generate many key metadata properties relevant to the BGNN project, as well as general image quality metrics (e.g. brightness and contrast). Results also show that we can accurately generate bounding boxes and segmentation masks for fish, which are needed for subsequent machine learning analyses. The automatic process outperforms humans in terms of time and accuracy, and provides a novel solution for leveraging digitized specimens in ML. This research demonstrates the ability of computational methods to enhance the digital library services associated with the tens of thousands of digitized specimens stored in open-access repositories worldwide. 
    more » « less
  2. null ; null (Ed.)
    Biodiversity image repositories are crucial sources of training data for machine learning approaches to biological research. Metadata, specifically metadata about object quality, is putatively an important prerequisite to selecting sample subsets for these experiments. This study demonstrates the importance of image quality metadata to a species classification experiment involving a corpus of 1935 fish specimen images which were annotated with 22 metadata quality properties. A small subset of high quality images produced an F1 accuracy of 0.41 compared to 0.35 for a taxonomically matched subset of low quality images when used by a convolutional neural network approach to species identification. Using the full corpus of images revealed that image quality differed between correctly classified and misclassified images. We found the visibility of all anatomical features was the most important quality feature for classification accuracy. We suggest biodiversity image repositories consider adopting a minimal set of image quality metadata to support future machine learning projects. 
    more » « less
  3. Garoufallou E., Ovalle-Perandones MA. (Ed.)
    Biodiversity image repositories are crucial sources for training machine learning approaches to support biological research. Metadata about object (e.g. image) quality is a putatively important prerequisite to selecting samples for these experiments. This paper reports on a study demonstrating the importance of image quality metadata for a species classification experiment involving a corpus of 1935 fish specimen images which were annotated with 22 metadata quality properties. A small subset of high quality images produced an F1 accuracy of 0.41 compared to 0.35 for a taxonomically matched subset low quality images when used by a convolutional neural network approach to species identification. Using the full corpus of images revealed that image quality differed between correctly classified and misclassified images. We found anatomical feature visibility was the most important quality feature for classification accuracy. We suggest biodiversity image repositories consider adopting a minimal set of image quality metadata to support machine learning. 
    more » « less
  4. Collections digitization relies increasingly upon computational and data management resources that occasionally exceed the capacity of natural history collections and their managers and curators. Digitization of many tens of thousands of micropaleontological specimen slides, as evidenced by the effort presented here by the Indiana University Paleontology Collection, has been a concerted effort in adherence to the recommended practices of multifaceted aspects of collections management for both physical and digital collections resources. This presentation highlights the contributions of distributed cyberinfrastructure from the National Science Foundation-supported Extreme Science and Engineering Discovery Environment (XSEDE) for web-hosting of collections management system resources and distributed processing of millions of digital images and metadata records of specimens from our collections. The Indiana University Center for Biological Research Collections is currently hosting its instance of the Specify collections management system (CMS) on a virtual server hosted on Jetstream, the cloud service for on-demand computational resources as provisioned by XSEDE. This web-service allows the CMS to be flexibly hosted on the cloud with additional services that can be provisioned on an as-needed basis for generating and integrating digitized collections objects in both web-friendly and digital preservation contexts. On-demand computing resources can be used for the manipulation of digital images for automated file I/O, scripted renaming of files for adherence to file naming conventions, derivative generation, and backup to our local tape archive for digital disaster preparedness and long-term storage. Here, we will present our strategies for facilitating reproducible workflows for general collections digitization of the IUPC nomenclatorial types and figured specimens in addition to the gigapixel resolution photographs of our large collection of microfossils using our GIGAmacro system (e.g., this slide of conodonts). We aim to demonstrate the flexibility and nimbleness of cloud computing resources for replicating this, and other, workflows to enhance the findability, accessibility, interoperability, and reproducibility of the data and metadata contained within our collections. 
    more » « less
  5. Garoufallou, E. (Ed.)
    Flexible metadata pipelines are crucial for supporting the FAIR data principles. Despite this need, researchers seldom report their approaches for identifying metadata standards and protocols that sup-port optimal flexibility. This paper reports on an initiative targeting the development of a flexible metadata pipeline for a collection contain-ing over 300,000 digital fish specimen images, harvested from multiple data repositories and fish collections. The images and their associated metadata are being used for AI-related scientific research involving au-tomated species identification, segmentation and trait extraction. The paper provides contextual background, followed by the presentation of a four-phased approach involving: 1. Assessment of the Problem, 2. Inves-tigation of Solutions, 3. Implementation, and 4. Refinement. The work is part of the NSF Harnessing the Data Revolution, Biology Guided Neural Networks (NSF/HDR-BGNN) project and the HDR Imageomics Institute. An RDF graph prototype pipeline is presented, followed by a discussion of research implications and conclusion summarizing the re-sults.ite this need, researchers seldom report their approaches for identi-fying metadata standards and protocols that support optimal flexibility. This paper reports on an initiative targeting the development of a flex-ible metadata pipeline for a collection containing over 300,000 digital fish specimen images, harvested from multiple data repositories and fish collections. The images and their associated metadata are being used for AI-related scientific research involving automated species identification, segmentation and trait extraction. The paper provides contextual back-ground, followed by the presentation of a four-phased approach involving: 1. Assessment of the Problem, 2. Investigation of Solutions, 3. Implemen-tation, and 4. Refinement. The work is part of the NSF Harnessing the Data Revolution, Biology Guided Neural Networks (NSF/HDR-BGNN) 
    more » « less