Harris Corner Detection
Canny Edge Detection
Inspired by our own personal stories, we wanted to create a comprehensive solution for pressing problems we faced in our lives. One issue that was of paramount importance to us was the safety of our loved ones. We drew inspiration from how many accidents can be prevented with the proper measures in place. One of our members had a grandmother who suffered a fall and was seriously injured as her Cataract resulted in her being unable to see obstacles properly. We wanted a solution for all members in a family, ranging from the grandmother who has failing eyesight and cannot see corners of furnitures, to the young child who is going to school for the first time. We are selling a solution that offers a peace of mind to all members of the family.
What it does
Our application which comprises an app and a device, takes over the caretaking role of vulnerable members of the family. The wearable device can be made in the form of a watch for children, or a pendant for grandparents. The device can track a child’s location so parents can observe their location with just a click of their app. The device also has fall detection capabilities, so emergency services can reach the elderly who has suffered a fall in the least amount of time. By using wearing tech, it removes the need for the elderly or child, who might not be tech savvy (or if it’s a young children, their parents might not want to allow their use of a smartphone), to use their phones in the event of an emergency. We also hope to eventually expand our solution to also have accident prevention tools; for example, using the Canny Edge detection and Harris Corner detection techniques to identify edges and corners in a room so elderly members know where they should avoid, and parents know which corners they should edge-proof for their toddlers.
How I built it
We used OpenCV library with Python for the sharp edge monitoring for safety + Android Studio to create the application.
Challenges I ran into
As most of us were using OpenCV for the first time, we faced many compatibility issues when installing the computer vision library. We were also newcomers to this, hence the learning curve was steep and we spent a lot of time debugging the entire night. We did not get much (if any) sleep while rushing this project out in the limited time span.
Accomplishments that we are proud of
We managed to learn useful computer vision techniques and applied them to practical use within 24 hours. We also managed to survive on minimal sleep.
What I learned
Android sensors, OpenCV, session management for android.
What's next for Family Care
We aim to use SmartNation APIs, such as OneMap API in our app. By doing so, a shared database of the family’s favourite places can be stored. Using geofencing technology, we can also detect if users veer off-course, as well as save their location data.