Gesture Controlled Home Automation using CNN

As technology is rising, more advancements are made in making the life of people easier, by providing methods for easy monitoring and managing. In this paper, a Home Automation model is designed to provide ease of control of home appliances, using an android application. The elderly and physically challenged people can perform their day-to-day activities efficiently. In previous methods, accelerometers are used to monitor the activity, which are accurate but are not flexible and portable. The proposed system detects the gestures given as input by the user and controls the home appliance. The client interface is responsible for capturing the input gesture from the user, using an android application and uploading it on the raspberry pi server. Raspberry pi acts as an important preprocessor. Backend Processing involves image preprocessing, training the CNN model, and prediction of image class category of input gesture image. Based on this predicted class of image, the respectively assigned action takes place at the home interface.