The inspiration was the edge detection that we were interested in and we coded it using Java because since we were in first year, this was the only language that we knew. We used OpenCV and Tess4J to scan the digits that were showing up. We noticed that the digits were way too quick so we had to slow it down using a timer. And after that, we had to use different Java imports to scan the numbers and put them into a GUI in real-time. The trouble that we had was we did not know how to import the Tess4J libraries and we ran into some issues such as the video not displaying on the GUI and it was detecting different numbers and symbols so we had to limit the digits that it scans from 0-6 so that it accurately scans the video. For denoising, what we did is we used OpenCV and its denoiser.
We learned how to import different libraries in IntelliJ and we also learned how to use the different libraries. We deepened our understanding on how Java works as a whole.
The challenges that we faced were importing the libraries and choosing the right detector to detect the digits. Before, we tried to use Humble which greatly failed and we resorted to using Tess4J. We were also having difficulties with FFmpeg. But overall, diamonds are made out of pressure.
-
-
Note: To run the repo, you must have OpenCV and Tess4J in your computer with the path files directed to them respectively. I did the demonstration so that you can just look at the code because installing them is a hassle.
Log in or sign up for Devpost to join the conversation.