Anna - Finalizing My Detection System


I am in the last few stages of my project. To verify all of my detection systems, I have been using a novel method proposed in a Nature article, which has close to 100% accuracy in identifying plastics. This method uses Nile Red dye to color the plastics and proceeds to look at it with multiple fluorescent lights. It stated that the light that worked the best was blue-green light.


I have already tested both the navigation system and the detection system underwater, and their waterproofing is perfect. Once I have made sure my detection is as accurate as possible, with the use of the method I mentioned above, I will put it into the ocean again to see how well it is working with detecting plastics underwater.


The best part of this process was whenever my programs didn't work, I would always find a way to solve that problem. Once I do that, it gives a great feeling of success and is very motivating. For example, when I was writing my code for the part of my detection system that looks at unnatural colors, I found a chart for the version of HSV that is usually found in photoshop. I didn't realize that OpenCV has it's own separate HSV colors, so what happened, was the program didn't detect exactly the colors that I wanted. When I finally managed to make my program work, the next day, I managed to finish the next step in one sitting.


At the end of the day, it was always great to go over everything with my mentor. She is so supportive of everything I've been doing and helped me a lot with making my ideas plausible. Without her, my project would not be what it is right now.