Elevated CNN Based Secured Sensor Image Data Communication for HAR: IIOT

Abstract

The Industrial Internet of Things (IIOT) is constituted by the development of the Internet of Things into the fields of manufacturing, monitoring, and management. It enables industries to make use of data communication and control by leveraging the capabilities of smart machines and real-time analysis. Our proposed work concentrates on the application of IIOT along with two contributions: the first one is to secure sensor based image, video and audio communication and the other one is the detection of human activity. Initially, the human activities data are captured by the camera sensors which is attached in the employee ID card that sensor monitors the employee activities as a motion image along with vibrations and sound signals. All this information is securely communicated to the smart office dashboard with the help of an efficient encryption algorithm namely Software Defined Network along with Stochastic Honey Overlapping Based Dual Helix Scan (SDN-SHODHS), that secures the communication in the access network. In the second phase to detect employee activity, we propose an efficient Elevated Deep Convolutional Neural Network that gives the best efficient result compared to all other existing works. Our proposed framework achieves secure sensor data image, audio and video communication and accurate human activity recognition in terms of encryption and decryption time, precision, recall and F-measure.