3D-CNN Architecture to Improve the Classification Accuracy of the Real-Time Images from IOT Devices

Abstract

The classification of real time images from the fast data capturing devices in Internet of Things (IoT) environment is a critical task. It requires suitable processing and development of a model for increased accuracy in classifying the objects in real-time. Therefore, the necessity in improving the accuracy of classifying the instances is needed post performing the modelling, building and development of a model. In this paper, a three-dimensional (3D) Convolutional Neural Network (CNN) is developed to increase the process of classification for the objects in the real-time environment. The objects needed to train the classifier is supplied and the model is built in python environment. The results show an increased classification accuracy in detecting multi-objectives than the state-of-art models.