Our inspiration came from the never ending quest for a safer internet experience. We have all experience the excessive use of vulgarity or obscene imaging in today's world wide web. Censible is our method of cutting down on these forms of media and providing a more enjoyable experience.
What it does
Censible is a web browser extension meant to provide a better search experience for the user. Censible will monitor webpages for the user and remove user selected key words and blur images relating to those keywords. Censible can be used by parents to protect their children's devices from inappropriate content by setting their own custom boundaries. Censible has workplace applications as well, users can create a list that can block distracting data from their view.
How we built it
We used an open source git repo called "Spoiled" which could parse html for a list of banned words and black out all the text around the words and any images near them. We expanded upon this by making images blur with a gaussian effect and incorporated google cloud's Vision api to look for labels in the images. These labels were also compared with the banned word list, making the censoring more specific. We also added an embedded options menu where the user could choose how much to blur images.
Challenges we ran into
This project encountered many roadblocks. The first of many was implementing the Google API. Our team lead looked into the problem and came up with a solution after many hours of research. Another challenge was allowing the user to create multiple lists for different work settings. Unfortunately our group ran out of time before we were able to complete this task. In the future, we look to completing this task along with other unfinished tasks.
Accomplishments that we're proud of
We are very of the Google Cloud Vision API that we implemented into the system. This API looks at every image on a webpage and then generates a list of words relating to the image. Those keywords are then checked against the users blocked list and will blur out the image if any words in the list match the block list.
What we learned
What's next for Censible
In the future, we would like to see Censible expand to the point where users can add multiple lists for different settings. This way Censible could be used in both an educational and workplace setting. The user would just need to switch to their separate list for that task.
Log in or sign up for Devpost to join the conversation.