Border-Square net: a robust multi-grade fruit classification in IoT smart agriculture using feature extraction based Deep Maxout network

Internet of Things (IoT) is a distributed system of interconnected tools, such as people, animals, wireless devices, and agents, called nodes. In IoT, clustering is a data collection process that reduces energy consumption by forming IoT nodes into clusters. In the clustering, all nodes are arranged into virtual clusters, while one node acts as the Cluster Head (CH). The correct selection of the cluster head reduces the energy consumption. Now a day, IoT is being distributed in environments, such as the smart agriculture sector or forests. Fruit quality classification is a significant task in the supermarket, factories, as well as other industrial applications. Accordingly, fruit classification mechanism helps cashier of supermarket to find the species and prices of fruits. Various fruit quality classification approaches are developed to find quality of fruit. Accordingly, an efficient fruit quality classification method is modeled by Border Square Optimization-based Deep Maxout network (BSO-based Deep Maxout network) classifier. The proposed Border Square Optimization (BSO) approach is designed by the incorporation of Border Collie Optimization (BCO) with Least Mean Square (LMS) algorithm. It is necessary to select the energy-efficient node as CH, as the process of routing the fruit image to the sink node is done through CH. With the features acquired from fruit image, the multi grade classification of fruit quality is done by the Deep Maxout network model in such a way that training practice of deep learning classifier is accomplished by BSO model. The proposed approach achieved superior performance in terms of throughput, energy, delay, and accuracy with the values of 0.6759, 0.6753 J, 0.3659 s, and 0.9467.