Beatrice Casey, Weeks 7+8: The Grand Finale

If any of you had met my mother, then you know that her biggest fear is someone (more often than not, me) being hungry. Going home on the weekends is really nice because I get to spend time with my family and do a lot of fun things, and not to mention we have so many fruits and vegetables that we could probably open our own farmer’s market. Monday mornings, however, usually consist of my mother stuffing my bag to the brim with fruits and snacks and water, until the point where I have to tell her to stop because it won’t all fit. I am not complaining one bit, I actually find it quite funny, but I worry that she worries too much about me being hungry. I don’t know how to convince her that I am okay and that she doesn’t need to give me so much food, but if it makes her feel better to give me all this food, then I’ll let her.

Original Histogram of the distribution of pixel intensities.
My final weeks in EXP were probably the most exciting and stressful of them all. I had to finish my project and show it to my professor, but seeing the results was amazing. Knowing that the graphs I saw were ones that the code that I wrote produced made me very proud of myself. This was by far the most difficult thing I have ever done and I am more than happy that I was able to pull through and figure it out.

I moved out of my sister’s house on the 28th of July and packed the night before. It took a while, mostly because I kept procrastinating by watching Netflix or YouTube and eating snacks. But I fortunately managed to finish up at a reasonable time and I went to sleep. My mom and brother picked me up the next day and helped me move out. My mom was more than happy to be taking me back home, but my sister was sad to see me go. Now that she is in her junior year of college, she is getting more and more busy, which means I get to see her less and less. I am really glad that I was able to spend this summer with her and have her by my side while I worked at UPenn.

Histogram After the Census Transform-
showing the distribution of the transform across the image.
Although I moved out on that Friday, I still went back on Monday to finish up some work. That would have been my last day, except for the fact that one part of my code needed to be finalized and I didn’t get the chance to finish it before I had to catch my train. I had to return on Thursday one last time (because on Tuesday and Wednesday I was in Atlanta visiting colleges) to finally submit my project and say my goodbyes. I arrived in the lab and told Dr. Taylor that I was ready to show him my project and so he sat down with me and had me explain what my code did, line-by-line, so he could make sure that I actually knew what I was doing. Fortunately, I passed the test and he was pleased with the results. After I explained my code, Dr. Taylor explained to me once again what my project would be helping with so I fully understood what I had done. So what did I end up doing, you ask? Well, I wrote a program that would perform a census transform on an image and return a graph showing the distribution of the census across the image. The census transform takes a 3x3 pixel patch and compares the surrounding pixels to the center pixel (changing the number to 1 if it is greater than the center pixel, and 0 if it is not) and returns an 8 bit binary that represents the pixel patch. Then, it is converted into a decimal number and stored in an empty matrix that is the same size of the original image in the spot where the comparison took place. It then produces a graph (above) showing how many times a specific number from the census transform occurs. This is helpful in Stereo Vision because when calculating the Hamming distance, a value that occurs less often is more valuable and useful than a number that occurs more often, since if the value occurs less often and there is a match across two images, it is more likely that the two spots are the same.


Overall, my experience has been amazing. I am so thankful that I was able to be part of such an amazing program and experience and I cannot believe that I managed to succeed in this project. This experience made me realize that computer science is indeed the field I want to pursue and I am beyond excited to learn more in this field.

Comments

Post a Comment

Popular posts from this blog

Deja Cunningham, Week 1, Training, Training, Training!

Image
Edwin L. Jones Research Building (My lab) My first week at Duke has been both exciting and tedious. I first entered my lab last Monday at 9 am and began my experience by going to a lab meeting. My PI introduced me to the rest of the lab members who I would possibly be working with. My lab is pretty small.  There are a total of 7 people including me, the PI, a lab technician, two PhD students, a grad student, and a post doc. They were all very welcoming and were surprised by how many techniques I had already known how to do (thanks to biotech!). The lab meeting was about two hours long and consisted of the lab members presenting on the progress/new discoveries of their project. In the beginning, I understood a small part of each of their projects, but after a while it all sounded pretty foreign to me. I was at least a little proud that I could understand pieces of it. After the lab meeting, my PI Dr. Shinohara showed me around the lab and told me all the things I needed to do...
Read more

Emily Guo, Week 1&2

Image
I've just finished my second week at the lab. The transition has been a bit rough, mostly because the project my graduate student is working on is quite different than what I had prepared for. It also took some time to get situated in New York and to adjust to living on my own. I'm living in an apartment with three med students in Tower 3, on Haven Avenue. It is about half a mile's walk from my lab and has a nice view, although it is quite noisy at night.  As it turns out, my grad student, Alyssa, actually lives in Tower 2, right next door. There are grocery stores and restaurants all along Broadway, two streets over. The Hudson and the George Washington Bridge from my window Tower 3 (in the back) Dr. Tycko's lab is located on the sixth floor of the Irving Cancer Research Center on St. Nicholas Avenue: Everyone usually gets to the lab at 10:00 in the morning, and I usually stay until about 6:00 p.m. The first day, however, I came at 8:00, and had to w...
Read more

Keeley Garvey, Week 1, von Frey X 320

Image
View from my desk My first week at Duke has been exciting and informative. On the first day, I entered the Snyderman Genome Sciences Research Building 1 with chocolate and nerves. My grandmother is Italian and insisted that I come with gifts on the first day. The grad student I am working with, Mr. Yul Huh, told me not to be nervous and gave me a tour of the lab. It was all very overwhelming at first, but he made me feel comfortable. He told me that a student from Dartmouth started a week earlier than me and would be working with me in the lab. After lunch that day, Shawn, the student from Dartmouth, sent me a few primary articles that would be necessary to read because they are the foundation of my project. For the most part, Shawn and Yul are the ones I go to if I need help or don't understand something because there is somewhat of a language barrier in my lab. However, there are a few other labs next to us that are doing similar pain research, so I could go to them as well....
Read more