Currently working on a camera for my dorm room that can detect if a person is at the door and alert me via an iOS app. The doorbell will run on an Raspberry Pi and the image detection will be done on OpenCV and the app will be built in Python. The goal of this project is to create something comparable to Amazon's Ring Device
Installed and implemented OpenCV, achieved facial recognition via CascadeClassifier within OpenCV, achieved facial recognition with bounding boxes.
I was completely unaware that OpenCV comes out of the box with image classification tools, I was always under the assumption that additional libraries had to be installed. Hoping to get the recognition to be less touchy, perhaps slowing the frame rate of the video capture down so that events such as taking a picture or sending an alert via webhooks does not happen multiple times per instance that someone is at your door.
Get OpenCV to take an image and save it locally when it detects a person in the frame longer that a couple seconds. As well as have the computer chime when a picture is taken or something to alert someone at the door their picture has been taken.