August 11th, 2017: Day 27

Today was the fastest morning meeting ever. Matt handed out papers and then we all got to work. I worked on Roger's data for the morning. Yesterday Paige and I put together another folder of results for him so today was just finishing the final three files. I finished the files during lunch so Paige and I got more things from Roger. We did another practice presentation with Matt and Joe. We got some good notes and critiques. The presentation is just going to keep being updated, likely up until Wednesday night. Today's run through was even shorter than last time so we added a few more slides, including an acknowledgement slide. I included Roger, Dave, Robert/Bob, and Todd Hanneken. We're probably going to add more people to it but that list is a good start. I helped David Lewis with some stuff, mainly finding things in a lab. It was better than nothing. Monday I'm going to help him set up his experiment, I'm pretty thrilled.

Comments

July 20th, 2017: Day 11

The morning meeting wasn't too bad today, we finished up the abstracts and had a good chat. Paige and I had the most eventful morning we've ever had, we got to go to all four floors and visit all of the labs trying to find people. We also talked to Ryan to let him know how disappointed we were with him but he's fine. We've been putting the final touches on the presentation and it looks pretty good. Peter, Ashley, and Tristan came up for lunch and we analyzed life's deepest mysteries like if Peter's shorts were shorter than mine, which they definitely were, and each others social media pages. It was a pretty good lunch. I've been doing a lot of reading today in the absence of work. Going back and forth between a dry book and a semi-dry presentation is absolutely exhilarating. Paige has been making a lot of small changes to the slides and I'm good with that, they aren't major and she's right to do most of them so I'm glad she sees some of these

Abstract Rough Draft

Document Restoration's main focus is to study old documents and analyze them using several techniques. We are currently working on finding corrections using pigment analysis and imaging enhancements on the Selden Map of the Ming Dynasty dating back to the 1600's. We will also be working with faded and illegible texts. These documents have either been erased, covered up, or have faded. Spectral imaging creates more precise and useful information than regular red, green, blue (RGB) imagery. For example, hy per-spectral and multi-spectral imaging are used to make items more legible and more visible to the human eye. Multi-spectral imaging uses less than 50 color bands during processing, while hyper-spectral imaging uses several hundred bands during processing. Other analysis techniques used to get more information about a part of a document include Principal Components Analysis (PCA), Spectral Angle Mapper (SAM), and Mahalanobis Distance. These techniques can also be applied t

July 19th, 2017: Day 10

During the morning meeting today Joe attempted to look up my house of Google Maps but unfortunately, the car did not drive down my street. We did three more abstracts today and they were all pretty good. Paige and I will be the second one done tomorrow so we'll give it a final review later today. Dave isn't here today so we're on our own and doing extra things to add to the presentation. I just finished a slide that has 12 chips of the map all compared to the red pigment of the chrysanthemums in Japan. I'll add that in at the bottom after I finish this. Paige is working on comparing chips of the map to a blue pigment and using the island off Southwest Sumatra as reference. Yesterday she did the same thing using green pigments and put together a map and I'm attempting to do the same but there are far less red pigment on the map so the SAM image is usually white with a small amount of black or grey. I'll add that image in later too. I guess all we have left to do

RIT Summer Internship Blog

Search This Blog