My research project focused on solving a common but serious problem: helping Alzheimer’s patients take their medicine correctly. People with Alzheimer’s often forget when or how much medication to take, which can lead to health problems, hospital visits, or even dangerous situations. To help fix this, we developed a smart pill dispenser that uses a special computer vision technique called the Discrete Cosine Transform (DCT) to recognize pills based on their shape and size.
I chose this research topic because I’ve always wanted to build something that makes a real difference in people’s lives. I’ve seen how hard it is for elderly patients to manage their medications, particularly when memory loss is involved. The idea of using technology, especially computer vision, to solve this problem felt meaningful and exciting.
At first, I thought the capstone would mostly be about building hardware, but it turned out to be more about analyzing data and writing code to process images. That shift surprised me, but it helped me grow a lot as a student. I had to teach myself new concepts in signal processing and image analysis, which was challenging but also rewarding.
The hardest part was working with image data. It wasn’t always easy to get clear images of pills or extract the right features using DCT. Small things like lighting or camera angle could affect the results. On the other hand, once we figured out the right steps, the actual coding and testing went more smoothly than I expected. I also found that writing about the results and explaining them in simple terms helped me understand the work better.
In the future, I’d love to expand this project by connecting the dispenser to a mobile app for caregivers. The app could give updates, track missed doses, or even show live camera feedback. I’d also like to train the system to recognize more types of pills and work under different lighting conditions.
What I learned from this project is the importance of persistence and creativity. Research doesn’t always go the way you expect, and sometimes what seems like a small discovery—like a data pattern—can lead to a big improvement. I also learned how powerful it is when engineering and healthcare come together. This capstone taught me not just how to build a smart device, but how to think deeply about the people who will use it.
No comments:
Post a Comment