Stories in Stone - Mysterious Petroglyphs and Inscriptions in the NortheastApril 1 2018
Over the 25 years of exploring the northeast, we’ve encountered petroglyphs and post-European settlement Inscriptions. Though some are still part of the popular culture, most have been lost to the deep forest of New England. There are many interesting interpretation for some of the known inscriptions in the northeast. While others are still somewhat of a mystery.
The mission of this expedition will be to locate and document all known and newly discover carvings across the northeast. We will also be researching historical records for any that have been lost and long forgotten. Once we have cataloged a petroglyph or inscription, then the hard work begins. Based on local records, archeological data, legends and lore, we will try to unravel their story. We’ll try to put them into historical context as best as possible with the available information. In the end, the story behind some will remain a mystery. I’ll leave the speculation of their origin to others.
Accept contributions for your expediton by providing us a few details. We will create an account on your behalf at WePay. If you haven't already registered with WePay, they will send you an email to complete your registration.
Making the Invisible Visible Pt III
Pictographs in the northeast are rare. The few I am aware of are from the colonial period of New England. Unlike petroglyph, pictographs are just a surface coating and tend to be far less durable. Pictographs that have survived in America are either in protected areas such as caves and shelters or exist in dryer climates.
Though New England’s harsh weather appears to have erased the pictographs, we are hoping that some of the surface coating might still faintly remain on the rocks surface. If so, we might be able to reveal the remaining coating with an image enhancement program called DStretch. This program has been very successful in drawing out details in an image that are invisible to the human eye. This tool applies a decorrelation stretch algorithm that was originally developed by NASA JPL. Simply, an algorithm that will find and extract a weak signal from the noise is applied to the color data. In this application the weak signal is the difference in hue that is too subtle for the human eye to notice. Once this weak signal is detected, the contrast for each color is stretched to equalize the color variances. This tool has buttons for a variety of colorspaces that the decorrelation can be performed in and give different results. The results I’ve seen with pictographs in the western United States is no less than amazing.
Though I have not been able to use DStretch on any of the known colonial pictographs, out of curiosity, I applied the tool to photos from known Indian shelters I’ve visited over the years. To my surprised I saw results with one of them. On what appeared to be bare rock, a have hidden text was revealed. At Abrams Bedroom in Swansea MA, the boulder at the entrance to the shelter was shown to previously have the word CAVE painted on it. Since it appears that it had been spray painted onto the rock, it could be from no earlier than the late 40s. Though the find has no historical value, it showed the easy of use of this tool, and its value in the field & to review older images for unknown pictographs.
DStretch is also available for free.
Below is a video demonstration of DStretch being applied to the rock outside of Abrams Bedroom in Massachusetts.
Making the Invisible Visible Pt II
Another way to accomplish virtually the same results you get with MatCap Shading & Specular Enhancement of a 3D model, is through Polynomial Texture Mapping (PTM). This is a technique currently used by archeologists that work with difficult to see inscriptions. With this technique you do not create a 3D model, but instead take a series of photos of your subject from a stationary camera. While photographing your subject, you illuminate it with a flash following a planned pattern. With this tool you do not move the object when examining it, you instead move the position of the light source. Just like with the 3D model, you can also apply a Specular Enhancement. The results I’ve seen so far are very impressive. One advantage this technique has over using the 3D model is that the object is always stationary. As you manipulate the light to reveal features on the rocks surface, the fixed position of the rock makes it easier to see & interpret them. The disadvantage with this technique is it is much more difficult and time consuming. Free software available for PTM and the improved version know as Reflectance Transformation Imaging (RTI).
We decided to give RTI a try this weekend. To do this we needed a camera, tripod, an independent flash/light source, and two black cue balls. Positioning the cue balls near our subject, we proceeded to do a series of photos moving the light sources position for each one. The light source follows a dome pattern, best to be described as a series of positions along each spine of an umbrella.
Our first attempt was unsuccessful. When processing our images with the RTI software, we encountered an error. The issue was most likely due our rush to get a first try under our belts, poor planning, and not following the recommended workflow. We will continue attempting to master this technique, so we can evaluate if it will be a useful tool for our expeditions.
Below is a Video Demo of Reflectance Transformation Imaging (RTI) provided by Cultural Heritage Imaging (CHI) a non-profit organization.
Making the Invisible Visible
The goal of this project is to not only rediscover and document the inscribed rocks, but to also view them with fresh eyes, through new technology and photographic techniques. MatCap Shading and Specular enhancement of 3D models and Polynomial Texture Mapping (PTM) are two techniques we are investigating, to leverage details from the weathered and difficult to see inscriptions. Since there is a handful of recorded pictographs in the northeast, that were last known to be barely visible, we will also be experimenting with an image enhancement technique known as Decorrelation Stretch (DStretch). Over the next few posts I will give a brief overview of each of these techniques. During the debrief of this expedition we will provide a short report on them and what we believe are the best practice when applying them.
Though we’ve already gotten excellent results when 3D modeling sites through photogrammetry, capturing an inscription can be much more difficult. The rough texture, and highly irregular coloring of the stone can make it difficult to see an inscription. Also, a greatly weathered inscription can often be impossible to notice. To eliminate these issues, we need to strip away the texture & coloring, and better highlight even the shallowest inscription. Using features in 3D modeling called MatCap Shading and Specular enhancement, we should be able to do this. Matcap Shading would replace the photo-texture on the model with a smoother silver/gray shade, and apply a fixed light source. This should eliminate a lot of the visual clutter in the model. Next, we can apply a specular enhancement, which will make the surface of the model more reflective, highlighting even a shallow marking on the rock with a shine or shadow. This technique can be applied by anyone using any of a variety of free 3D modeling software, and viewers.
We’ve begun to dig thorough our own records and organize the inscriptions we’ve already visited or are aware of. Currently we have added 76 petroglyphs and inscriptions to the list, with many more to go. We’ve put them in a database, recording information including our data sources and the currently concluded context. We’ll be using this as a road map to plan our expeditions to locate the inscription starting this spring. Once an inscriptions is found, we'll use photography techniques to best capture even the most weathered rocks. As we go along we update the database, and provide a field report of our findings. When possible we will construct 3D model of the inscriptions.
Here is a link to an example of a 3D model we’ve already done for the recently found In Hoc Signo Vinces Inscription. https://sketchfab.com/models/7e3172d765f845aebea9bcc34f50feb9/embed