Digital Image Analysis with ImageJ: Applications in Conservation

Online, March 7, 9, and 11, 2022 with live sessions at 12:00 - 4:00 p.m. EST
Led by Lai Ding

This workshop introduces FIJI ImageJ, an NIH developed open-source digital image analysis software, to the art conservation community. ImageJ is an open-source program for digital image analysis with a robust user-base in the medical and scientific imaging communities. Some of the tool’s functions, such as comparing multiple images and measuring shapes within an image are familiar concepts in conservation. While these assessments are usually carried out by observation or approximation, ImageJ introduces consistency and speed to the processes. Applications in conservation and preservation are growing and this course will offer an opportunity for beginner and intermediate users to learn from an expert in the field. After the workshop, the participants will be able to perform basic functions independently and they will have enough proficiency to ask questions in the user community or to resolve issues by consulting available documentation.

The workshop’s three sessions will focus on:

• March 7: Introduction to basic digital image analysis techniques and ImageJ. Topics will cover measurement, filter based de-noise, background subtraction, and automatic object identification. These concepts/techniques are not limited to ImageJ but can also be applied to other image analysis software as well.
• March 9: Advanced topics and case studies. Topics will include morphology (shape) filters, segmentation/threshold, image alignment/registration, and designing image analysis protocols. We will also touch base on the ethics of image processing.
• March 11: The final day of the workshop consists of 20-minute one-on-one ‘office hours’ with the instructor to discuss participants' specific concerns regarding their own project and images. To best utilize this time, participants are asked to send the instructor sample images with a project description by February 28 so Dr. Ding can review them prior to the workshop. Dr. Ding may also use the sample images as case studies in the March 9th session.

All participants will receive workshop material including practice image sets before the workshop. Though this workshop is aimed as an introduction workshop on ImageJ to beginners, the instructor highly recommends participants to practice on ImageJ before the workshop. The workshop will use FIJI ImageJ, all participants need to install the FIJI ImageJ and make sure it works on their own computer. FIJI ImageJ works both on PC and Mac, though PC is preferred for this workshop.

The live sessions for the workshop will take place in Zoom and automated captions will be available. The first two live sessions (March 7 and 9) will be recorded and accessible to participants shortly after the session is complete.

This program is supported by a collaborative agreement between the National Center for Preservation Technology and Training (NCPTT) and the Foundation for Advancement in Conservation (FAIC). FAIC was created by a grant from The Andrew W. Mellon Foundation and is supported by donations from members of the American Institute for Conservation and its friends. Workshops are made possible with the assistance of many AIC members, but no AIC membership dues were used to create or present this course. FAIC relies on your contributions to support these and its many other programs. Learn more about donating to the foundation here.

Thank you to the dedicated members of the FAIC Collaborative Workshops in Photograph Conservation Advisory Committee for their work in organizing this workshop.

Working with Data (“1s and 0s”): September 16

• Data types (strings, booleans, floating point, integers, etc.)
• Spreadsheet terminology (cells, rows, tables, headers, etc.)
• Character encoding - what is it and why does it matter?
• Data formats/serialization - why it’s good to use .csv, etc.
• Nulls vs blank/empty and why does it matter?
• Regular expressions (regex) - what they do, where do they appear?
• Activity: Identifying data types with provided worksheet

Exploring and Standardizing Data with Open Refine: September 23

• Live demonstration, creating a project in Open Refine
• Faceting - built in functions and some basic GRELs (Refine regex)
• Editing and clustering - built in functions and some basic GRELs
• Activity: Following short list of recipes for Open Refine with participants' own dataset or a provided dataset

Vocabularies and Reconciliation: September 30

• What are vocabularies and why you should be aware of them - benefits of aligning your data with others
• Standard vocabularies in the field, how to access and use them
• Live demonstration, Open Refine reconciliation service
• Activity: Explore pros/cons of reconciliation options (VIAF, Getty Vocabs, Wikidata, etc.)

 Mini Project (no live session)

• Participants work with their own dataset or one provided, draft project document that describes dataset with cleaning and reconciliation goals. Create Refine project and complete as many cleaning steps as possible with focus on columns for reconciliation (could be artist names, materials, or something else). Define standards to reconcile to and why they chose. Opportunity to schedule a meeting with instructor/assistant to discuss progress and challenges

Case Studies and Discussion: October 14

• How have these ideas been applied in practice?
• What were the experiences from the mini-project?
• What implications does this content have on your ways of working? 
• Optional: Submit results from the mini project to get feedback from instructor/assistant