MSI Image Processing, From Low-Contrast Signal to Readable Writing: Extracting and Visualizing Text and Features from MSI Data

With Ivan Schevchuk and Keith Knox

In this workshop we will learn how to manipulate multispectral image data to recover textual information and compare this process to manipulating RGB images. Participants will learn how to interpret multispectral data and image processing results that visualize material properties of inks, writing supports and more. We will get hands on experience in recovering a few different types of text and other written information, using image processing tools such as HOKU, ENVI and GIMP that are available for accomplishing this task.


Ivan Shevchuk

Since 2016, Shevchuk has worked at the Centre for the Study of Manuscript Cultures at the University of Hamburg in Germany as an imaging specialist (technician) responsible for multispectral imaging and image processing for the recovery of lost, damaged and erased writing. He previously worked with synchrotron and free electron laser light source designing, building and maintaining experimental set-ups. His background is in unfinished mechanical engineering. 

Keith Knox

Keith T. Knox is the Chief Science Advisor for EMEL, the Early Manuscripts Electronic Library. He is a graduate of the University of Rochester, with a B.S. in Electrical Engineering and a Ph.D. in Optics. Since the 1990s, he has recovered lost text from several manuscripts, including the Archimedes Palimpsest and the Diary of David Livingstone. Keith is the author of Hoku, a general-use, Java-based software application for processing multi-spectral image data of damaged or erased manuscripts.