An artificially intelligent deeplearning app for securing your home and surrounding. The deeplearning based tensorflow object detection app identifies humans/car/bikes through your mobile camera lenses. The tensorflow object detection app limits itself by suppressing a few functionalities such as intimating the users, raising the alarms whenever the object of interest is detected. Being a demo its accuracy of prediction has been limited. Hence, prediction below 0.9 should be deemed as incorrect. A clean, perfect solution can be obtained from the authors upon request.
Consider a scenario wherein someone enters your home while you are away. The bad intent of the the intruder can be ruined by installing this security system which identifies the person and sends you an alert.
Few questionnaire:
a). Deep Learning - Deep learning is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms. (learn more on Wiki)
b). Object Detection - Making a machine learn to classify objects just like human do right from childhood. Involves a lot many things. (More link here)
c). Tensorflow - TensorFlow is an open-source software library from google which provides a platform to develop and deploy your own deep learning models on various devices.
Tensorflow Github Link - https://github.com/tensorflow/tensorflow/
This app github link - https://github.com/agabhishek/ArtificialIntelligenceHomeSecurity (for linux system only)