Inspiration
The idea for this project came to life from our experiences using the MTA subway system. There are many times where trains are delayed or even stuck in a place due to fire caused by garbage on the train tracks. As a result, we started to think of ways we can incorporate modern technology to solve this problem.
What it does
The Trash Detect application asks the user to choose an image or directory of images that will then be scanned by using the IBM Watson and Clarifai computer vision APIs. Computer vision helps to extract details from an image such as garbage and train tracks. If the computer vision analysis detects garbage near the train tracks, it alerts the user with a message prompt that trash is present.
How I built it
The application was built with python and uses the IBM Watson and Clarifai computer vision APIs to analyze images. The computer vision APIs extract information on the image which are then used as conditions to trigger message alerts. When the APIs detect trash or litter, the user is alerted with a message saying that trash is present.
Challenges I ran into
We did not have much experience with using python and the computer vision APIs so we had to learn a lot of concepts throughout the project.
Accomplishments that I'm proud of
Were proud of the fact that the program works almost flawlessly and includes a native graphical user experience.
What I learned
We learned how to use python and computer vision APIs.
What's next for Trash-Detect
The next step for Trash-Detect is to incorporate machine learning APIs to make the detection more accurate and learn over time.
Built With
- clarifai
- ibm-watson
- pillow
- python
- tkinter
Log in or sign up for Devpost to join the conversation.