Overview

 

The Beyond the Box Digitization Competition will award up to $1 million to the person or team who creates an automated technology that increases the speed and accuracy of digitization of a drawer of insect specimens and their associated data.

The Competition is a joint effort of the U.S. National Science Foundation (NSF) and the American Institute of Biological Sciences (AIBS), with NSF serving as the Sponsor and AIBS serving as the Organizer.

Entrants will use a standardized drawer of insect surrogates (referred to as "specimens" within this document) and related surrogate objects, pinned in multiple unit trays with labels stacked and arranged in several "typical" and challenging configurations. The drawer's specimens must be imaged with no human intervention, and label data must be automatically captured and translated into digital format using automated methods such as optical character recognition (OCR) and natural language processing (NLP). The Competition encompasses visualization of obscured objects, digitization of labels, digitization of specimens, and OCR. Data dictionaries may be utilized to assist the OCR translation. The solution may tap into existing natural history collections data dictionaries, create comprehensive new data dictionaries, or leverage a combination of the two resources.

Prize Parameters

  • The positions of specimens, unit trays, labels, and other elements in the Entrant's drawer must be those outlined in the specifications for the surrogate drawer. The only allowed human intervention is the placement of the entire specimen drawer in the imaging station developed by the Entrant. Human manipulation of the positions of specimens, labels, unit trays, or any other elements within the drawer is not permitted once the specimen drawer has been placed in the imaging station.
  • Once the specimen drawer has been placed in the imaging station, direct contact with the specimens is not permitted, either via human intervention or the technology. The position of the specimen on the pin may not be manipulated in any way.
  • Specimens must not be damaged in any way.
  • Labels must not be damaged in any way.
  • If a pin is removed from its original location by mechanical means, it must be returned to the precise original location and position.

Submission Materials

All Entrants must submit the following materials via the submission Web site:

  1. A video demonstrating the process of scanning the entire surrogate drawer and all of its specimens and related labels and capsules. The video must be split screen, with the left-hand side of the video showing the technology/device in operation, while the right-hand side of the video must show the digitization of each specimen and associated labels and capsules in real-time. Videos should be a minimum resolution of 720p. The video will be uploaded as a private video to YouTube (via the submission Web site), and once submitted, the Entrant will not be able to view the video on YouTube.

    Uploaded videos must be in a format indicated as acceptable by YouTube: https://support.google.com/youtube/troubleshooter/2888402?hl=en/

  2. Three images each of one specimen from each of the trays, as described in the Level 1 Achievement. The specimen selected from each tray is up to the Entrant, except for the Tray 4 specimen, which must be either specimen 3 or 5.

    1. One image must be the single dorsal image of the specimen.
    2. One image must be the single lateral image of the specimen.
    3. One image must be the single lateral view of the specimen plus any associated labels and capsules.

    This equates to a total of 24 images (three specimen images X eight trays). The images submitted must be generated with lossless compression. The image file format submitted must be non-proprietary. The images, as indicated in the Levels of Achievement, must have a minimum resolution of 5760 x 3840 pixels. The actual captured images from the device do not have to meet the above specifications of lossless compression. This specification is only provided for purposes of the submission aspect of the Competition. Captured images are required to be lossless, and should not be compressed.

    Captured images: Those images captured directly by the digitization technology. These images must be captured by the technology in an uncompressed, lossless format with a minimum resolution of 5760 x 3840 pixels.

    Submitted images: Those images submitted with an entry solely for the purpose of aiding in judging. These images must use lossless compression with a minimum resolution of 5760 x 3840 pixels and be submitted using a non-proprietary format.

  3. Three images of the technology, with an element in each image that provides an indication of scale. Image files must not exceed 10 MB each. The image file format must be non-proprietary.

    1. One image must be a side view of the device.
    2. One image must be a top view of the device.
    3. One image must be a front view of the device.

  4. One image of the surrogate box, clearly showing each specimen. The image file must not exceed 10 MB. The image file format must be non-proprietary.

  5. Log-in and access information for:

    1. The database
    2. The Web interface, assuming the Entrant believes their technology fulfills the 6th Level of Achievement

    Use Appendix K, Form 1 to submit this information. This document (which will also contain the marketing plan referenced below) must be submitted as a PDF.

  6. A 4-page (maximum length) marketing plan discussing and demonstrating how the technology would be made available (could be at cost) to collections to assist them in solving this national problem. Be sure to include associated costs with marketing the technology, as indicated below. The marketing plan will be evaluated based on the following:

    • The plan appears well thought out and appropriate/reasonable
    • The plan proposes potential methods for deploying the technology
    • The costs and estimates seem reasonable
    • The plan is comprehensive

    The marketing plan must be submitted in a PDF format, using 12-point Arial font, with one-inch margins. Use Appendix K, Form 2 to submit this information. The marketing plan may contain images, tables, etc., as necessary but these must be included in the 4-page limit.

  7. An export of the associated metadata information in an xls format. The file must include the following:

    1. A drawer identifier
    2. A timestamp showing when the data capture process began
    3. A timestamp showing when the data capture process was completed, when the drawer can be safely removed
    4. A specimen counter indicating how many records are in the database for this drawer
    5. An operator identifier indicating who was running the machine

Judging

All submissions will be evaluated by a panel of judges based on whether they have or have not met the specified Levels of Achievement and their associated parameters, and will be scored based on the quality of meeting these parameters. Entrants must meet all Level 1 through Level 4 parameters to qualify as a finalist.

The number of judges evaluating each submission will be based on the number of submissions received. Finalists will be invited to an on-site demonstration and evaluation of their technology for 1) their original surrogate drawer and 2) an actual insect drawer with real specimens (provided to the finalists on-site).

If the selected winner fulfills all of the Level 4 Achievements (but not all of Levels 5 or 6), they will be awarded $500,000 USD. If the selected winner fulfills all of the Level 5 Achievements (but not all of Level 6), they will be awarded $750,000 USD. If the selected winner fulfills all of the Levels of Achievements set forth for the Competition, including all of the Level 6 parameters, they will be awarded $1,000,000 USD. Only one winner will be selected.


For full contest rules, visit the Complete Contest Rules and Materials section.

 

 

GET NOTIFIED  Sign up to get notified about the competition.

National Science Foundation Logo    American Institute of Biological Scienes Logo