Fractional stock exchange trading optimization trained deep learning for wild animal detection with WMSN data communication in IoT environment

The Recent development of Wireless multimedia sensor networks (WMSN), and the Internet of Things (IoT) have been improved for resolving day-to-day concerns in the agricultural field. Furthermore, agriculture fields near the forest areas face a serious hazard from wild animals that attack farms regularly. Because of the wide range of wild animal movement and physical sizes, wild animal detection, and monitoring are more complex. This research developed an IoT-enabled WMSN for wild animal detection using the deep learning (DL) technique. Initially, the IoT-WMSN simulation is initialed. The nodes gather the input wild animal data and the routing process is carried out to predict the best route. Later, the collected data by a node is routed to the base station by the proposed Fractional Stock Exchange Trading Optimization Algorithm (FSETO). Then, the adaptive median filter (AMF) is employed to remove the noise from the input wild animal image. The saliency map extraction finds the noticeable regions of the image in the visual field, and the wild animal is detected by FSETO-enabled deep convolutional neural network (deep CNN). Moreover, the detection is evaluated by precision, recall, and f1 score of 0.900, 0.897, and 0.918 respectively.