Skip to main content

Day 15

Today at our morning meeting we saw a presentation given by one of Joe’s undergraduate seniors for his senior project. His presentation involved analysis of how phones take pictures in various environmental conditions and are able to analyze text. It was interesting as I had never learned about his before. Also I attended another seminar given by a professor involved in image analysis of documents with cultural significance. Using various imaging and lighting techniques he was able to decipher text that may not seem visible at first on valuable documents. A document that I thought was cool was Archimedes Palimpsest. Later I continued to work on the pigment mixture data sets.

Comments

Post a Comment

Popular posts from this blog

Last Day

Today, was the final day of my internship. This was a great opportunity and I had my final presentation today.The presentations started once family members, faculty and staff, and other professors arrived and I was fourth to present. I gave my presentation and afterwards was relived that it was over. Also, congressman Joe Morelle made an appearance, which was exciting. We all got to meet with him and talk to him about the internship. Afterwards he gave a small speech and we took some pictures. Later we were all presented with our certificates of achievement. The internship was a unique and rewarding experience that will help me in the future. 
Post a Comment
Read more

Day 12

Today one of my fellow interns described her lab group in the morning meeting. She discussed some of the code she was working on and explained different aspects of the code she was writing. Later, I started developing a method to test the effect “noise” was having on the MOUSE. The “noise” is what occurs as part of the spectra but could affect the pairwise algorithm, so we wanted to see how large an impact it was having on the algorithm. After going through some of the spectra of one of the mixtures and creating varying amounts of noise, we realized that the affect was not as large as we thought. Then I took spectra of some of the mixtures again to run in the algorithm as some of the mixture had anomalies. One of the mixtures so far of the reruns was successful as it only returned percentages of the pigments in that specific mixture, which indicates that the algorithm is working effectively.
Post a Comment
Read more

Day 17

Today at our morning meeting we went over the last outline of a fellow intern. We discussed a little about the presentations and some suggestions. I continued working with the data and forming representations and compiled the data together. Also, I went to a masters thesis defense in the college of engineering, which was cool. The defense was about shadow detection and different algorithms of improving shadow detection in various systems. The defense also mentioned a lot about convolutional neural networks and other coding methods. Later, I worked on gathering spectra for some samples and tried to improve the MOUSE functions, so the spectra would generate more accurately.
Post a Comment
Read more