Inspiration:
The eye is the most important sense of human being because it allows us to observe the beautiful outside world and process information. What if computers also have this ability? It opens so much potential for what humanity can achieve automatically. That's why we are eager to learn and apply a technology called Computer Vision, especially the fast-rising potential of automotive car and AI. Computer vision is an idea of giving computers "eyes" (Yes! Imagine how much data we can get from a daily basis). When we're at VandyHack, Hillman group challenged us to use a camera that put on top of a square to read, measure any hand, and return the gloves size for them. It is indeed a perfect opportunity for us to practice our Computer Vision knowledge. If the camera acted like our eyes, it would have been an easy task. That’s exactly what we tried. We help computers achieve vision by making it run through many breakdown steps.
What it does:
When users put their hand (palm down) on the flat surface of the box, the camera will read and measure their hand size. Then, the software will return the recommended gloves size for them.
How we built it:
We use mainly OpenCV library and Python language. In a simple term, we initially turn all frames of the video into black and white, so the computer can process more efficient. Second, we draw an imaginary static triangle according to the box’s white surface (assume the camera and the box always remain the same position). Third, we draw an imaginary horizontal line that preferably cut the hand into half. Using that line, we can calculate the size of hand. Finally, we double that size, compare with the gloves database, and return suggestions for users.
Challenges we ran into:
The calibrating camera results in even sides of the square on the pictures, so we had to draw a static square. Pictures captured by the camera appeared to have many noise can affect our measurement. Therefore, we let users pick where they want to draw the line to get the best result. And of course, we only have a limited amount of time to complete a computer vision project.
Accomplishments that we're proud of:
During the coding project, our group were lacking python knowledge. However, we were able to learn what we need in a short period of time, help each other to identify errors, and put everything together. And overall, we have a working project.
What we learned:
We learned to utilize rich sources from the Internet, teamwork, and encourage each other
What's next for the impeccable hand:
There are still many factors that can affect the result we get. Therefore, if we have more time, we will be able to come up with more cases and testing to perfect our product
Log in or sign up for Devpost to join the conversation.