Friday, July 29, 2016

Day 18 (July 29th - We're Pretty Much 1/2 Way There)

Morning Meetings:
  • Eastman House Summary
    • Good: Very knowledgeable guides, good lunch, great bus (there was way more room than we needed)
    • What should be done for next time: More time to explore other exhibits in the house, have the guide from the archive show us more of his work and not just explain it.
    • I was really happy to go to the Eastman House for a second time. I went there once before with my Photography 2 class and had the chance to look at the exhibits showing at that time as well as the rest of the museum areas. We did not have a chance to go see the archives however.
    • The archive area that we interns were given a tour of was pretty interesting. There is a lot of tech and chemistry involved to repair damaged artworks. I kept thinking of Oxford's really expensive camera that they had bought to image their library's ancient documents and how it could possibly help with the Eastman House's picture and artwork restoration. Oxford's camera is extremely expensive though, so I don't think that buying one and using it would be very practical.
  • slow and steady. I have been running mostly Classifications, PCA and SAM on my image chips for the past few days. I will eventually re-run everything of mine through ACE, and MF to pull more stuff out. I really should also get some practice in with using Di's mask code and Anna's Blur and Divide spatial method. I will get to those eventually.
  • Madi left early for a family trip (like, right after lunch). I think I can manage without her, maybe... (Come back soon!!!!)
  •  None of us interns can believe that we are half-way through our internship; our time together to laugh, play and work. I started to feel that sensation of "The End" today. Everyone comes up to work in the Reading Room by now for at least a few hours a day.

*************Emily: Rio Olympics this year: The Russians, Zika and Katy Perry. Good Lord! (This is just a summary of what she said and how she feels. Feel free to tell me to take this bullet out or edit it, Emily).************************

  • Cookout day. I met Aaron (RIT honors incoming freshman) and James (Byron Bergen Junior in the fall). They both work with coding. Aaron has been working with Biotechnology on DNA strands. He has already completed around twenty. James is really interested in Computer game management and design, although Byron Bergen does not have a strong technology program. I wish them both all of the best.
  • We interns played some volleyball and Frisbee. I ended up joining some of the grad students in their Frisbee game/ toss. The intern Frisbee was way too light for me to throw and I can't throw Frisbees well under trees.
*********** I think almost everyone has stopped working by 3:00 and is now taking a break by playing different, old platform games. YES!!!! In their defense, everyone has been staring at some shape or form of code for the past week. We needed a break from work. *******************8

Thursday, July 28, 2016

Day 17 (July 28 The Eastman House)

Everyone came in around 9:15 this morning instead of the usual 8:45. All of us high school interns, one RU student, Mr. Pow and Mr. Callens went to the Eastman house for a tour of one of their exhibits and a tour of their picture restoration area (in the guarded archives). Memories of my previous visit there with my Photography 2 class continued to flood back during the tour. Once we got to the archives though, I was out of familiar waters. Our tour guides were absolutely marvelous.

Afternoon work:
  • Madi fixed my bad Kmeans habit
  • I finished running image 7 through all of the programs
  • I started running image 8
  • Most of the sea looks the same by now. There were few interesting blotches in the last few image chips.

Wednesday, July 27, 2016

Day 16 (July 27 Don't forget the Outlines!)

Morning Intern Meeting:
  • We don't know who will be presenting today
  • Field trip tomorrow
  • Outline and Cookout on Friday
    • For Outline: Keep it Simple
  • RIT +1 (Tuesday) 
  • Ithaca Teganic Park water level at all time low
  • Code Academy
  • Movie Night
Morning Meeting with Mr. Messenger:
  • See "Outline for July 29 (web readable)"
Lunch Talk:
  • Perform Lab: Why Eyes Do Eyes look where they do? (I wrote 3 pages of notes.)  
  • What in the image drew a person's attnention?
    • Giving tasks matter! (different question phrasing leads to different results/focus areas)
    • Humans are subject to motor delays
    • Ball catching simulation game with Motion Capture (virtual reality headset)
    • Testing eye Fixation, Saccade and Pursuit
  • Typed up outline
  • continued working on image 6 (not much was found)

  • worked through image 7

  • Anna kindly reviewed the Blur and Divide process with Madi and me before lunch. Madi helped me again with it after lunch and now I know that I should not try processing more than one band image at a time.
  • Brain officially overloaded for the 5th workday in a row.

Outline for July 29 (web readable)

Enhancement of Hyperspectral Imagery of Historical Documents
Presentation Outline

Abstract/ Project Introduction and Description (1)
  • Historical Documents
  • Spectral Imaging
  • Spatial/ Spectral Image Processing
    • Spectral (color
    • Spatial (Brightness and location)

Hyperspectral Imaging (1)
  • (define, what does it mean-Collecting hundreds of colors)
  • Pictures of Oxford Set-up
  • Example of spectra of images

Processing (2)
  • Overview
  • ENVI
  • Classification
  • PCA Signature Matching (SAM, ACE, MF)
  • Spatial Processing

Classification (1)
  • (K-means, Mahalanobis Distance, etc.)
  • Example (picture of each)

PCA (1)
  • (exploration and visualization)
  • Example
Di’s masks in PCA (picture and explanation) (1)

Signature Matching (1)
  • SAM [ACE, MF]
  • Reference spectrum to compare
  • Examples (pictures)

Spatial (1)
  • Blur and Divide
  • Examples (pictures)

Gough Map (1)
  • History
  • What’s in it
  • What we’re looking for
  • Data

Pictures (3)
  • Big picture (high resolution)
  • Big picture (HSI [“data” picture file]))
  • Zoom ins

Results (Pictures, Spectral Graphs, etc.) and summary (7-10)

Tuesday, July 26, 2016

Day 15 (July 26 Let's Try Blur and Divide)

Morning Intern Meeting:
  • Same events as yesterday
  • Updates on everyone's progress:
    • Mr. Pow: getting ready for incoming freshman with Bob and preparing the CIS annual report
    • Nathan, Alice and Maria: finished going over interns' recorded data from eye tracker experiment
    • Emily: something about RZ Pisces and BC Pisces, calibration changes and R,I and G Bands, for now whe's back to reading about Jet Stream identification
    • Nick: turning black and white images to color, working on the Model format of the game (game functions not controlled by player)
    • Madi and I: Learned more Spectral programs, just learned some Blur and Divide for spatial programming, still working on the Gough Map, maybe will get working on the Codex
    • Zihao: asked to borrow material samples from a library, took around 200 samples through low spectrum
  • Outlines of our final slide show due this Friday on blog (it will be subject to change later.)
       Debate over Pokemon movies and who ate the Boston Cream doughnut. (Thanks again Bob for the donuts!)

        Continue to process chip files from the Gough Map while having occasional conversations. I could not get the Blur and Divide programs to work so I'll have to ask Ana for help later. I ran SAM and PCA on image 4 and Classifications, SAM and PCA on image 5. Image 5 did not have any interesting hidden features. It did give me a headache though.

Monday, July 25, 2016

Day 14 (July 25 The Sky Finally Opens)

Almost everyone got caught in the 8:10 - 9:00 AM morning downpour.

Morning intern meeting announcements:
  • This Week's events
    • Wednesday Lunch Talk
    • Thursday Field trip to Eastman house (leave by 9:30)
    • Friday Cook Out (please no rain or blazing heat)
  • Next Week:
    • Aug. 2 RIT +1
    • Aug. 3 Library Talk
    • Aug. 5 REU Presentations (open to the public)
Morning adviser meeting:
  •  Spectral Matched Filter (MF) - finds center of data by applying a perpendicular vector to the vector connecting a pixel's coordinates to it origin. (higher number = pixel spectrums are more similar to target pixel)
  • Adaptive Coherence/Cosine Estimator (ACE) - combo of MF and a cone (attached to the perpendicular vector)
  • SAM - (lower number = pixel spectrums are more similar)

Madi, Di, Mr. Messenger and I visited Anna's lab in the lower level of the CIS building. She uses Spatial ENVI programs to enhance and better contrast old (Blur and Define), faded texts unlike Di, who uses Spectral ENVI programs and a MATLab computer code to get a similar result. Spatial processing is different from Spectral processing in that one involves creating contrast based on a pixel's color spectrum while the other involves creating contrast based on a pixel's brightness. (Greater contrast sharpens the image details to be less blurry).

Madi and I have been receiving a lot of crash courses in learning the ENVI programs and what those programs do. Most of the mathematical details were not explained to us or have flown over Madi and my heads (due to their high level of difficulty) along with some detailed concepts so I'm not confident in giving a good description of how everything works.

Friday, July 22, 2016

Day 13 (July 22, Lights, Music, Chicken)

Our morning intern meeting topics:
  • Weather for the cookout (hot, maybe rain?)
  • Library Lecture/presentation about RIT online databases on August 3rd
  • End Presentation tips and works:
    • peer review of run through
    • think about the title now
    • be sure to acknowledge the sponsors and corporations who are funding your research groups
    • invites won't be sent until next week (invite guests, family, friends and counselors?)
  • Finding Undergrads for Imaging Science
 Mr. Messenger reviewed how to make spectral profiles of pixel colors and explained the Radiometry process and equation used to Madi and I. My Y-value numbers ended up being not quite right because I made my chips from file copy named "raw" of the Gough Map section while I should have been using the file called "data". I had no idea how the two files were different until he explained it to us. The the "raw" file had the overall observable light intensity while the "data" file is a documented "part" of the overall observable light intensity. I had been using the "raw" file for my chips which turns out to be alright for the ENVI programs, but not for spectral profiling.

These two graphs contain different spectrum data samples, but the y-axes are different.

"raw" pixel spectrum (above)

"data" pixel spectrum (above)

Mr. Messenger also helped Emily figure out a problem with the images she is processing of the night sky. What I heard of that conversation was that there was a piece of thread on the the telescope lens when the pictures were taken, but it only appears in some of the images. He introduced Emily to a pixel-sample process to try (I don't remember any specifics). I like the idea that the same process ideas can be used on both the macro and micro images.

Cecilia (aka CiCi), Emily, Madi and I headed over to the Interfaith Center with REU student  John to help him collect data for his music research algorithm. We H.S. interns took turns playing the piano (in the Interfaith Center) and the violin (CC and I brought ours) while John collected sound recordings and data on his computer to process later. This included playing the violin and piano at the same tempo both together and individually, and recording a duet between Madi (piano) and CC (violin) playing the Swallowtail Jig. (I happened to find it among the music papers stored in my violin case. It was fun to listen to!) It turned out that I was the only person who knew how to tune the violin strings by ear (without the help of fine tuners). Emily had a way better sense of when the individual strings were in tune though. Each of us H.S. interns were given a chicken drum-stick during the lunch cook-out in return. They took a while to cook, but they were delicious! (Thank you again John!!!)

Mr. Messinger had time today to read over the Gough Map article that Di gave Madi and I last week and revised his thinking of how to go at cleaning up the Gough Map. By the rate of our work, I think all of us can get through most of the Gough Map while Madi and I are around.

Thursday, July 21, 2016

Day 12 (July 21 We All Laugh)

Our day started off with the question "What happened at your school this year?" The most common topics that were brought up were: pulled fire alarms, senior pranks, school lunches and complaints about high school teachers. Madi and my morning meeting with Mr. Messinger involved growing interest in the seemingly unchangeable ocean area in the Gough Map section that I had been working on the last two days. I had found small bits of text that had faded into what looked like brown splotches. Both yesterday and today I can not recall all of my findings, but the images I saved have helped me recall some of them.I tried showing Di later this morning and he was interested as well.

Most of the work day for Madi and I was reworking and posting the chips we had already created. I have been submitting my images into the shared google power point and Madi has been working on running other programs on the chips she processed the last two days. She had already submitted the chips she had worked on thus far. I tagged Di into Madi and my joint presentation so if he sees anything or wants to look at our work, it should be there.

Mr. Messenger had to leave a couple of hours early today, but Madi and I are hard workers while he is gone. All of the interns had lunch at the Student Union. (We all experienced waves of contagious, uncontrollable laughter for both hilarious and stupid reasons.) then Madi and I worked our way across campus to find the place to register another parking permit for my dad's car. (The one that I should be driving by now. It's one thing to drive remote control vehicles. It's another to actually be in the vehicle while driving.)

My Created Chips for This Week

Day 11 (July 20 MRI Interns Talk Banned Color Pigment Materials)

This morning started out with the intern morning meeting introducing the lunch speaker Mr. Kanan's, our field trip to the Eastman House next Thursday (the original plan of going to an observatory was canceled), and we discussed who would bring what Lunch Friday. The morning meeting with Mr. Messinger will stop in periodically to see what we have found. He's not as busy today.

Morning Work: Listening to CC and Zihao talk of how and where to get color pigment samples (Paris Green, etc.) I finished running SAM on MapEast image 2. Worked and finished all programs on MapEast image 3 (sea) (Located below images 1 and 2)

Lunch and Lecture: Mr. Christopher Kanan's labwork (Machine Learning and Visual Recognition)

Afternoon work: walk-around, running RGB on different program runs of image 3 to make out scribbles/writing. Created chips 4-8. Presented work to Mr. Messenger.

Wednesday, July 20, 2016

Day 10 (July 19th, The Drone and the Tree/ Work continued )

Our morning began as usual: Morning meeting with all fellow interns (Aug. 2nd is the Bring a Friend day), and morning meeting with Mr. Messinger (keep working with the new section of the map).

Madi had already started working on creating chips of the land area of the new section of the Gough Map, so I started running programs on chips of the sea. I think it's the English Channel. But I'm not the greatest with European Geography so it's the portion of the sea that is southeast of England. I think I cleaned up a depth line (a boundary line that may have served to show how deep the sea would be). But nothing else showed up otherwise. I only managed to get to two 800x800 pixel chips through all of the programs that I know thus far.

I took a break to be a test subject for the Visual Perception Lab just before lunch. I almost spilled the beans afterwards. The interns had a little bit of trouble with hooking me up to the Eye Tracker. I can't talk about what I saw so that the experiment materials can be reused (Phewy).

I ended the day with an Entrepreneurship talk that turned into a "Presenting to Non-Science Audiences" presentation. I had to leave in the middle of it though. The points in this presentation this should help with all of our presentations at the end of this internship.


PAC in RGB with Enhancement

Monday, July 18, 2016

Day 9 (July 18th, Abstract peer review day)

I started the day by getting to the internship three minutes late and forgetting that all of the interns and the internship advisors would be in Bob's workroom to review our abstracts thus far. (Thanks Emily for texting me your location. I know you read everyone's blog at least twice a day so you'll get this message, eventually.) Everyone picked apart everyone's abstracts for a total of about 35 minutes. The one outstanding abstract was written by Nieles (note to self: check spelling later) if I can remember correctly. His area of the astronomical imaging internship group involves creating a fun astronomical education game that is played somewhat like or except with cooler features (images and graphics of the cosmos are added to a collection/ gallery and are earned as rewards, educational information integration in the gaming features and gameplay, etc). I hope I understood his explanation alright. (Note to self 2: check Nieles's abstract again sometime.) I sooooooo want to try out this game once he gets the basic mechanics finished. I definitely would not be able to do what Nieles or Emily are doing if I got in with the Astronomical Imaging advisors. I like working with data and images as well as learning about the stars and their properties. But, I definitely need visuals to work with and pure representative data and coding are not my areas of expertise. I actually have nearly no coding experience at all.

Back to the topic at hand:
 Other than a couple of minor grammar problems, CC and Zihao's abstract was flawless. Madi and mine was alright ... it definitely needs to get cut down and re-worded. I thought we signed up for an Imaging Science Internship, not an English Writing Course focused on the field of Science. (I'm joking of course. It's too serious and quiet in the workrooms with everyone using earphones to listen to music individually. LIGHTEN UP PEOPLE!!!!!!!! Geez!!!)

Madi made Oreo balls and they were delicious! I only had cake balls and weird cheesecake balls before. Madi's recipe trumps them all!

Ok, Ok. I'll try to stay on topic for the remainder of this post. Sorry, there's something about the concept of Monday that I don't like.

                  I had to leave early today for a couple of medical appointments around town, so I was only able to "work" for a couple of hours today. Mr. Messinger introduced another ENVI program to Madi and I today, Anomaly Detection. It functions by finding a mean and standard deviation of the pixel properties (colors and brightness) in an image and then comparing each individual pixel to the determined mean. So, the program pretty much works Z-scores for the entire image. (I remember Z-scores very faintly from the end of the Algebra 2/ Common Core class this past year. I don't really want to relive my days in that class. Moving on -->)  It comes up as RX Anomaly Detection in the ENVI --> Spectral -->Tools. RX stands for Reed and Xioli. Mr. Messinger also showed Madi and I how to create our own chips and how to Profile the data pixel colors (figure out their reactivity to different wavelengths of photons). I think I have my facts straight at this point.
                 Di uploaded another segment of the Gough Map onto my temporary user computer account so I'll "play around" with it tomorrow. Madi definitely has a head-start on the work. And Mr. Messinger confirmed my fear: I should not have tried to reboot the computer in the reading room on Friday. I had no idea that those things were THAT sensitive. Someone else from the internship tried to reboot the Shut-Down reading room computer at the end of the table last week. (It's a different one from the one I glitched, and you know who you are.) and it was also showing raw code. So .... I'll go have a talk with the tech-support crew tomorrow if the computers were not fixed and give them an apology for my ignorance of computer technicalities.

Friday, July 15, 2016

Day 8 (Abstract for 7/18)

Madeline Loui and Allyse Toporek

Enhancement of Hyperspectral Imagery of Historical Documents

       Several Weeks ago, Dr. Messinger attended a conference at Oxford Univeristy to present new computer processing methods to enhance the faded and illegible writings on images collected of historical documents. The work he presented was of images the university sent to him of the Gough Map of England. This document is around 600 years old and has undergone much work (updates) since its first creation. However, some of the ink has faded with time or other environmental factors and what remains of the ancient writing and pictures are grey, organic, unrecognizable structures or nothing at all.

      We have worked to learn those computer processes and programs in the ENVI software and custom processing code created by Di Bai, a PhD student in CIS. We are using them on smaller sections of the overall Gough Map in order to identify the original, hidden texts and figures. These include, but are not limited to, Classification processes, Principal Components Analysis (PCA), and Spectral Angle Mapper (SAM). Our finished works will be sent along with those of Dr. Messinger and those of Di Bai to Oxford for analysis and, hopefully, description. A broader goal is to assess the different methods and their capabilities.

Day 8 (power point vs. libre office)

Our morning began with talk of the "Bring a Friend Day" for later on in the internship. This would give us a chance to show our research and results to them and stimulate their interest in science.

The majority of Madi and my work time was taken up putting the remainder of my images onto our power point presentation for Mr. Messinger to look through and incorporate into the power point compilation to be sent to David and Nick at Oxford. (My images from days 5-7, PCA w/masks and SAM results) We all have a little idea of what they want to analyze, so fingers crossed. Madi and my PCA and SAM results have given the better results. We also figured out how to put a picture of the overall Gough Map onto the slide show to indicate where each of our chips were. (Whoever invented screenshot is a lifesaver, at times.) We compiled a total of 100 slides (way more than for my APUSH Debate this year ; P)

(Madi and I are still enemies of the copycat LibreOffice on the computers we use. We still prefer good old Microsoft Powerpoint.)

The Neapolitan ice cream was a hit during lunch. It had kind of melted while in the cooler outside though. There's a little left over that will be left in 2nd floor break room for anyone who wants it. Someone else should be in charge of bringing the ice cream next week. The lunch today was fun again, although a bit windy with some rain and thunder for the first half-hour.

Chip 10 (Vanilla colored)

Chip 10 (Cookies 'n Cream color )

Thursday, July 14, 2016

Day 7 (Rerunning SAM)

We all started our morning meeting with donuts (Thanks Bob!) and going over what to put in our first abstracts for our final presentations (due tomorrow). By what everyone has been saying, a lot of people in the same research groups will have the same abstracts since their work are pretty much the same, except for Emily and Niels in Astrological Imaging. Niels has been working on coding and program creation for the group. Emily, meanwhile, will probably be doing data reduction and photometry. This would be after she has finished reading all of the manuals and textbooks necessary to fully understand her work (which is still ongoing since we all first started our work).

The entire morning was made up of re-running all of our images from yesterday (Madi chips 1-5 and me chips 6-10) through SAM (see Day 6) and coming out with full processed images. It took me the first ten minutes or so to figure out how I was still getting the band of black on the top of the image (which I still don't understand) but gave up and began running a more manual-like version of SAM on ENVI called SAM Target Finder with BandMax. That worked but I still felt like I am falling behind Madi in processing speed. I then realized that she was running less pixel sample trials than I was.

Original (infamous) Chip 10
(The red wall is Hadrian's Wall. To the left is Scotland and to the right is England)

SAM Chip 10 red pixel sample (with enhancement) from yesterday

SAM Chip 10 red pixel (with enhancement) from today

(These images look better on the ENVI viewing pages)

We all gathered around 11:45 to head to Global Village for lunch for a nice break. (We missed the lunch-rush by about 15 minutes.) It was nice to feel a breeze and the sun since all of us are somewhat cooped-up in our work areas (but by no means bored).

The afternoon was mostly finishing up the SAM reruns and typing up a draft of our abstract. All of us interns have been putting in a lot of time with our research groups and we have been discussing a little of what each of us have been putting in our blogs (since we can all see each others from the website , now an operational shortcut). Since Madi finished her portion of the power point early and her work, she went looking through previous interns' pages. She noticed that during previous years' internships, there was a "Bring Your Friend Day" and a "Biking Day". I'm really wondering if we will be able to do these this year since I like this idea.

Day 6 (SAM and Madi Describes "The Secret History")

We interns started our day planning who would bring what for our weekly Friday picnic. I volunteered to bring the ice cream this time. I couldn't remember if the ice cream name was pronounced as tripolitan or neapolitan. I figured it out later.

Mr. Messinger taught Madi and I the process and ideas behind SAM the program. It stands for Spectral Angle Mapper and it functions to search for pixel similarity to a reference point. All of Madi and my work for today was running the SAM program on the data chips given us on day one. For some reason though, the top 1/4 of all of our processed chips were cut off from the SAM processing and appeared as a black strip at the top of the viewing page. Mr. Messinger could not figure out exactly what Madi or I did to have a part of the image cut out. For most of the afternoon, Madi and I put together a power point of our best saved image results.

(After SAM and correct orientation)

Madi attempted to give a detailed summary about the book "The Secret History" before lunch. This took about an hour with interruptions and questions which included confusion about who died in the book and which characters played what parts in the plot. After this we had lunch and attended a talk about Eye Movement and Illusion of Continuity by Jeff B. Pelz. He talked about "optical illusion" (which is realy neural illusion), The Foveal Compromise, coping systems (Dual Retina and Eye Movement System), and the uses of an eye tracker camera in his area of work. This gave us interns an inside look of what the Visual Perception lab work was. The interns working in the lab could not disclose information to us about our work because the other H.S. interns will be test subjects for their experiment.

Tuesday, July 12, 2016

Day 5 MatLab and Coding Implementation

Di introduced his code on MatLab to Madi and me.

His code is designed to block pixels on an image. In the context of our work, the prominent red and green inks on the Gough Map would be blocked out of the image clips that Madi and I work on. Blocking the red and green inks would allow the PCA program to run a better data analysis on the more subtle and less prominent areas of ink.

Di's code then converts the final image into a file that the ENVI program can read and the PCA program and enhancement tools can be run on the processed image.

The orientation of the image was originally given to Madi and me rotated. After running PCA and Di's code, the images are flipped and rotated again. If you track the areas that look shadowed or smudged on the original image onto the enhanced images, some cursive-like writing can be made out. The bottom two images are my best enhanced images today.

(According Mr. Messinger, images of our results are alright to be added to the blogs. The raw codes are not.)

Original Image

After Di's Code/ After running the PCA program

After PCA Enhancement (black and white) 

After PCA in RGB
(blue ink writing was originally barely visible)

In other news:

  • The fire alarm went off unexpectedly around 11:00 and resulted in a visit from the Henrietta Police Dept. and some of the security guards on campus. (Thanks for coming!) This brought up my question of: Where should the interns gather outside in an emergency like this?
  • A presentation was given between 4:00 (16:00)and 5:00 (17:00). Madi and I could not attend sadley.

Saturday, July 9, 2016

Day 4 (Yay, first Monday)

Mr. Messinger introduced the Principle Componants Analysis (PCA) program to Madi and I during our morning meeting. The PCA programs are designed to identify a cooridnate system to best fit given data. In the case of using ENVI for color identification, each color in a document are a coordinate and PCA changes the coordinate system (the coordinates of the graph) to best fit the color data distribution.

So the majority of our work today was running Di initially gave us through PCA. We were able to clear up a lot more of the images and found more hidden text with PCA and image enhancement tools. We also checked some of each other's work and took a long walk at the end of the day to clear the headaches we got while staring at the computer screens for hours on end.

Day 3 (July 8th)

Madi and I are quickly assimilating to a kind of routine by now. The group has the 15 minute opening meeting, then Madi and I have a morning meeting with Di and Mr. Messinger about our progress with the Gouge Map, other interesting parts of the map they have found thus far through their research with Oxford, and what ENVI programs and tools Madi and I would be working/practicing and all of the details behind the programs. The rest of the day would be used running the ENVI programs we know on different image files that contained sections (chips) of the Gouge Map. Di would check up on us about every hour and a half to check our progress (or whenever Madi or I called him because I thought I froze the computer : | ) and we end the day with a 4:00 meeting to go over the day's work.

There was another Ph.D. defense at 2:00, however neither Madi nor I felt up to going to it after being fascinated and somewhat lost by Lady's presentation on her work (see Day 2). Even though it would have probably been interesting to check this one out, we couldn't find the room until Mr. Messinger told us we were on the wrong floor and we just felt the need to get back to the chips we were working on. I reasoned that we had an hour and a half lunch period for the Friday cookout and volleyball in the muggy weather and a stroll around the building was the only afternoon break that we should afford. It was a pretty good day for the planned event and it was interesting to chat with some of the attending college students.
In my opinion, my work on the chips was not as fruitful as Madi's thus far. But this is just my perspective and it has only been the third day of work for both of us. (Love the steep learning curve [double thumbs up].) She did clean up an image to find hidden writing at the end of the day which was extremely exciting for the both of us. It looked like original town names that were smudge out by an excessive amount of map-use and then someone rewrote the names with the red ink used to label all of the other towns and cities on the map. Another "stick man"-like blob also appeared. It kind of resembled the one that Di had discovered on another portion of the map before Mr. Messinger's meeting in Oxford.

To Madi and I, the resulting image with the discovery looked like a mistake. Madi had the program identify six different colors but an overall image with the six "false" colors, used to mark the areas with similar color shades, wouldn't process through ENVI. Only black and white images of each color detection would show individually. Her results also did not in any way resemble the work results I had from the same chip. I should not have been so quick to regard the result as a mistake. Mr. Messinger's explanation of the significant finds really surprised Madi and I. He continued to talk with Di about further work on the chip and area of the discovery which both excited and overwhelmed us interns. Every time we have a meeting, Mr. Messinger readjusts what Madi and I should look for on the Gouge Map during program processing and how it can be done. There is a "program processing" phase and an "analysis of the results" phase. Madi and I have been working on the "program processing" learning. For next week, I am eager to continue refining the ability to effectively analyze the results of the program processing.

Thursday, July 7, 2016

Day 2

Today started out with planning for the cookout lunch that the supervisors scheduled for this Friday (tomorrow). After a quick, 15-minute morning group meeting, Madi and I split up so that one of us could grab our locker keys and the other could notify Mr. Messinger that the other was on the way. Madi forgot where Bob's office was so I ended up grabbing the keys for us and was a couple of minutes late.

During our morning meeting with Mr. Messinger, he told us of ENVI's origins and gave a quick overview of what Madi and I learned yesterday using ENVI's classification programs. Interestingly, ENVI was first designed by a duo in Colorado for geological use. ENVI was later implemented in the field of astrology in the 90's. It was around then that, Mr. Messinger came across the program (if I remember the story correctly). The program was eventually sold and the creators now live comfortably.

ENVI was originally designed to be used by Ph.D. knowledge-level professors, but later versions were written with ease of access in mind for a wider variety of users. Di had Madi and I try the new version available on the computers we were using (I think he said the version was something like 8.5) and they worked alright. The screen appears way more like Photoshop and, in my opinion, displays many functions more openly than the "classic" version. But Madi and I switched back to the ENVI -classic (I think it was a version 5) for we were originally given our first instructions based on this older version and none of us could figure out where some of the basic functions were located on the newer version. (For example, we could not locate "True Color" to change the original images from black and white to color.)

At 10 A.M., Madi and I went to see a Ph.D. defense by a woman named Lady. (My apologies for misspelling your name Ma'am and I am so sorry for walking into your presentation late.) Her presentation was about an application that would improve Remote Sensors. We could not understand the vast majority of the mathematics she was presenting due to our inexperience with college-level mathematics, so I can not justly summarize what Ms. Lady presented. (I am also not certain if any of the specific information she presented is considered to be ready for publication.) However, we both listened attentively and I tried to take notes on a few parts that I had somewhat understood. I still do not completely comprehend about half of what I wrote (5 single sided composition notebook pages written total) however thanks to Mr. Messinger's explanations over the past two days, I did understand Ms. Lady's idea of targeting pixels with like-color on maps and trying to eliminate a great deal of background pixels, false positives, and no matches. Mr. Messinger later explained that our research will eventually be moving towards that stage.

For the remainder of the day, Madi and I practiced using different Classification programs on the same image chip (picture-section) of the Gouge Map. I adopted Madi's idea of taking screenshots of our trials and documenting them as we worked on them to help us later when we will have to give final presentations. We both struggled with the labeling and organizing of our files and I know that some of my trials are lost as undecipherable files. (They were mostly trial command inputs so it was alright, I think.) We pushed through and did our best.
Mr. Messinger took a look at my files at the end of the day and pointed out what Madi and I should be looking for when comparing the different color-gradient image copies that we produced. A lot of the images I had were the same as Madi's. Whenever I look at the processed images, I see them resembling really flashy-colored wallpaper. Mr. Messinger's points just made me more interested in jumping back on ENVI and trying again. (I am still nervous when I am under observation though : I)

(OOH last little note: Picnic/Cook-out tomorrow! 11:45-1:00 last time I heard. All are welcome!!!!!! I bought what I volunteered to bring and a little more since I heard some of my other interns say they were not sure if they would remember what they would bring.)

Wednesday, July 6, 2016

Day 1

Joe Pow and Bob Cullen started this first day with a tour of the Chester A. Carlson Imaging Center. All of the interns and I were then led to a red barn on campus to do some fun teamwork-building activities for the rest of the morning. All of the activities were very thought provoking and fun to figure out. The temperature quickly rose from comfortable to just-bearable faster than I thought, but it did not take away from the team-building, in my opinion.

After lunch (provided by Mr. Pow and Mr. Cullen, thank you for this!) Madi Loui and I were introduced to our advisor for Document Restoration, David Messinger. He and his grad student Di re-introduced Madi and I to their current project works: the Gough Map (of England) and the Archimedes Palimpsest. Mr. Messinger told us of his trip to Oxford and the overwhelming positive response he received to the results of his research on the Gouge Map. (What he was saying brought back many pleasant memories of sophomore year world history classes; in which we analyzed the significance of different historical documents that survived through the unforgiving dimensions of time and space.) Di then introduced Madi and I to the general works of the wonder-program that all of us would be using: ENVI. After getting over the fear my possible failure to use its programs, Madi and I quickly absorbed Di's instructions on how to use ENVI. The computer program appears to me like a cousin of Photoshop but requires more manual control of its tools, and more patience and practice to efficiently using it.

I love the idea that Madi and I will be able to help Mr. Messinger and Di with their research in this area of imaging science. I can't wait to practice more tomorrow!