Inspiration
In today's busy world we are always troubled with handling keys, or handing it to other house members. As a remedy to this problem companies introduced the notion of "smartlocks": which work basically on patterns and biometrics instead of traditional keys. But we feel that these "smartlocks" aren't that smart anymore. In today's era where automation is at its zenith, we feel that people should have more information and control over their houses even when they are not physically present there. As a remedy to this problem we put forth "Digital-Attendant", which is a security system that sits on top of the smart locks and gives the user his desired freedom and control.
What it does
It is an intelligent system that coordinates with the smart lock and notifies the user about person's visiting his house in his absence. Whenever a person comes closer to the door it detects this event and sends out a notification to the user's smartphone, the user then has the option to converse with this person over a video call or opens the door and let the person get in, or ignore. So even if someone who's not registered with the biometric system can get in without the user personally coming down.
How we built it
We used openCV for image recognition, android apps for user's smartphone, and we integrated everything from sensors to mobile phones using raspberryPI. The sensor at the user's door identifies the event and triggers a signal to raspberryPI. The PI inturn decides on what to do based on the captured image and the output from image recognition algorithms. The user is notified and his decisions are sent back to raspberryPI which inturns actuate the sensors to do the neccesary action.
What we learned
We learned a lot about integration, since our product has multiple devices, api, and applications, integration was a challenging task.
Challenges:
Start from the ideation, brainstorming for a good idea which will be helpful and useful to the masses. Coming to implementation challenges, it has many moving pieces v.i.z. Pi, sensors, camera, automated locks, infra red sensors, two mobile app, Computer Vision, Cloud server, dashboard. Having all these pieces working in harmony proved to be a challenge.
Accomplishments:
We were able to get an accuracy of about 90% and successfully completed other major parts and integration of the whole idea.
What's next for CodeMusic
We plan to build a good and rich home automation system. Complete with many more features. This is the first step we have taken toward home automation and I look forward to automate lots of other components and eventually add in Cognitive and AI to the system and also enhance it with an Intelligent Personal Assistant.
Log in or sign up for Devpost to join the conversation.