Using spectral imaging in DH

Despite the vague definition of Digital Humanities (DH), we can all agree that it involves using electronic tools of some sort to facilitate different stages of research in the field of Humanities.

As such, I’d like to introduce the “Hyperspectral Imaging”: a combination of a special imaging technique and image processing software which could significantly expedite data collection process while providing information that cannot be observed by naked eyes. Hyperspectral Imaging currently has case applications in agriculture, mineralogy, chemical imaging, astronomy, etc.

What is Hyperspectral Imaging?

Similar to a regular imaging system, think you cellphone’s camera, Hyperspectral Imaging cameras take a snapshot what they are pointed at. However, in contrast to the typical camera sensors or human eyes which see the incoming light at only three visible wavelengths (Red Green Blue), a hyperspectral imaging camera records the intensity of the electromagnetic waves coming in across the whole spectrum.

The goal is to capture the spectrum of all the points in the image. Processing the spectrum of the entire image can provide the necessary information for detecting objects in the image, chemical and physical  composition of the objects in the image, age estimation for the objects, etc.


Image reprinted from Adimec Blog [Link]

Use case in Astrology

As a part of a project that I worked on a few years ago, I developed a computer program that processed high spectral images taken from slices of meteoroids to determine the minerals in them and segment the image into pieces signifying mineral chunks in the rock. This was achieved by first segmenting the image into similarly looking spectra groups, and then comparing each group with a database of the spectra of known minerals.

Despite my short involvement in the project, I managed to develop the software which could group the minerals in the rock with high accuracy and almost fully developed the software to compare the spectrum for each group to determine its composition.

This slideshow requires JavaScript.

This turned out to be a fairly straight-forward to develop in Python and the final product could provide invaluable insight into the composition of the imaged rocks in a matter of less than a second.

Proposed use case

There have been recent advancements in incorporating new technologies to help with documenting architectural establishments. Such technologies help with cases where collecting large amounts of data is necessary. Additionally, such tools, for instance, digital tablets, can facilitate tasks such as marking original, previously reconstructed, and recently reconstructed sections of walls and buildings without harming the building.

Despite the incredible productivity increase that existing tools offer task of documenting a large establishment could take weeks or even months. This is where I see Hyperspectral Imaging extremely helpful as it could reduce the data collection time even further. Once properly developed, a computer software can process all the images from the building in seconds and automatically mark the areas of interest on the image in addition to determining the chemical composition and estimated age of each segment.

This could simplify documenting an archaeological site to simply taking images of the site with a hyperspectral camera and then uploading the images to a computer. Within seconds the proper software can produce insights that would be equivalent to months of manual work and lab experiments. I believe this could be an extremely helpful tool for the people in the field by providing more through data and reducing the labor work, allowing the scholars to focus on understanding the data and intellectually engage with the data.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s