Google Glass and paleontology collections: what we learned, Part I

PFL16A_Glass

Author: Kelsey

Part I in the Google Glass series. Other posts: Part II.

The Lowdown: Google Glass has remarkable potential as a curation and documentation tool, but what it gains in efficiency it loses in picture resolution and lack of updates. Before we acquire our next one, we will wait for a newer version with a better camera, but we are stoked by this new piece of technology.

Background: This summer the Nonvertebrate Paleontology Laboratory (NPL) acquired Google Glass, version 2 of the explorer edition. NPL is part of the University of Texas Jackson School of Geosciences, where I go to grad school and study Australian agamid lizards in all their cranial kinetic glory. Full disclosure: I was working at NPL and had suggested to Ann (Curator and Collections Manager) earlier in the year that we try out this new technology. The idea of augmented reality or forehead cameras is not a new one—sci-fi writers have been heralding their coming for over half-a-century—but here was a chance to test a tangible piece of the future.

We had one “simple” goal at the beginning of our adventure: test the camera and video for curation potential. Ten to twenty thousand fossils are added to NPL every year. Only 1/8th of these are digitally recorded in our database. Inventory is a careful balance of speed and detail. Whole drawer contents and individual specimens are often recorded. Our fleet of staff and volunteers have begun using cameras, ipads, smart phones, and now Google Glass. In science (and, I suspect, academic institutions in general) simple goals often turn into reticulating fractals of fascinating sub-tests, sub-questions, and side studies. Fortunately, that’s why I got into this business.

Requirements: We are interested in devices that reliably and repeatedly capture images with a high enough resolution that all text in the field of view is readable and the fossils are individually recognizable. These photos would then be saved in our database for future research and inventory reviews

Stats: The Google Glass Explorer Edition comes with a 5MP fixed-focus CMOS camera capable of taking 2560 x 1888 resolution images. The fixed focus means the glass is set to capture as great a depth of field as possible and will not adjust, automatically or otherwise. Glass will tune the ISO (shutter and aperture controls) from as low as 60 to a high of at least 960. Videos are shot in 720p only. The aperture size of about f/2.5 with a focal length of 2.7 mm.

Start: I found taking pictures with glass is like switching from a go-cart to a normal car, you have to get used no longer aiming for the middle of the road. When you are wearing glass, the screen is above your right eye (NOT in front of it), and the camera lens is to the right of that, so you will have to aim your head at left side of the drawer while taking the picture. The camera app for our edition does not have an aiming feature, so getting the correct angle and resolution takes practice, patience, and intuition.

Observations: Hands-free is great! When I used the iphone to take pictures (more about that in a part II), I had to constantly put down the phone to move drawers or reposition specimen labels. It was incredibly handy to have both my hands free. Additionally, glass really is comfortable to wear.

OkGlassOk Glass, point the laser at…

However, I found I was never truly hands free. For every single picture you have to backtrack (the “swipe down” action) to the glass home screen (above) and ask it to take a picture. Glass saves all pictures, but immediately prompts you to share an image immediately after you capture it. This could be solved with a simple picture app that bypasses the social media features. Once I get my mad programming skills up to snuff, this is one of the first projects I’d like to tackle.

Pictures: I found the large depth of field meant I often underestimated how much of the drawer was in view, or I’d overcompensate and get WAY too close and personal with the fossils, which just resulted in close up shots of fuzzy fossils. Most of these can be solved with practice and the addition of aiming software.

AimingTroubles

Aiming Troubles

The camera is very sensitive to light levels and has no internal regulatory mechanism. Even a slight adjustment in head angle can make the difference between a dim or overexposed picture. Wearing a baseball hat or wide-brimmed hat does not help, only squishes the glass down to uncomfortable angles.

 TooBrightTooDim

This analysis? Just right.

I also had a problem tilting my head to the right. I suspect most people to not hold their heads perfectly upright, which results in a tilt to the picture. More on how I dealt with this problem in an upcoming blog post, “Keeping Level-Headed.”

Looking down to photographs low drawers caused the glass to slip, so I added a lanyard in the back. This was easy for one side, but the battery on the right necessitated a duct tape solution. The addition does not compromise comfort too much and helps really secure the glass to my head. Fashion may also be compromised to a certain degree.

GlassLanyard2

Google Glass: Nerd Edition

Overheating was also an issue. Our collections spaces (“the cages”) are, for the most part, not climate controlled and summertime Texas heat and humidity are high even early in the morning. However, even in the climate controlled areas, continuous use causes glass to flash a warning message after about 15 to 20 minutes. This is an oft-cited problem in the glass community, and comes as a consequence of clashing optimal operating temperatures.

Is the text readable? Sometimes. The smaller the text and the worse the lighting conditions, the more likely it is to lose data. Also, it’s impossible to check the pictures until they are loaded onto a larger screen. On the other hand, when I had to leave a project half-way through and wanted to pick up where I left off, I could simply review the most recent pictures or videos and quickly start where I left off.

 DocumentComparison

 If I never see this label again…

So, is NPL a victim of the Gartner Cycle  or are we pushing the boundaries of museum science? Once I had a handle on the initial pros and cons (hands free, trouble aiming, trouble with light balance, overheating, and resolution) we decided to conduct a more formal study comparing the iPhone 5 to Google Glass. See “What we learned: Part II” for results and my preliminary conclusions!

Questions? Comments? Leave your reflections below.

Advertisements

5 thoughts on “Google Glass and paleontology collections: what we learned, Part I

  1. Kelsey, I love the way you’re thinking here. I’d love a future where the glass was connected to the RFID tags I want to place on each specimen, so that the specimen record would automatically pop up in your HUD when you hovered over it. The RFID tags would make checking specimens in/out and doing inventory so much easier, too. Plus: harder to steal specimens!

    On the topic of head-tilt. Having had the strabismus surgery twice, I now know a lot about our eye muscles. One of the symptoms of my strabismus was a permanent head tilt. I had a problem with the inferior oblique, which works in part by twisting the eye when the head tilts (did you know our eyes have limited auto-levelling for head tilt!?). Because my inferior oblique was spasming, I automatically controlled it by tilting my head. My head tilt was so ingrained that it took me several months after the surgery to correct it.

    I’m not saying your head tilt is a problem, but it would be worth having it looked at by an ophthalmologist.

    1. Hmmmm… interesting. Would the drawers have RFID tags too? I am picturing an automated system that could go get the fossil you request (like some libraries are doing with their stacks now). NPL is working on an arcGIS map of the collections as well. I could show you sometime.

      Also, I am training another person on Glass. I want to see if they start tilting their head as well. Eh… maybe I’ll just have the whole lab try it out. I’m always looking for a bigger sample size!

      -K

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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