Design and Implementation of an Automated Control System for Anomaly Detection Using an Enhanced Intrusion Detection System

Abstract

In forest zone and the field of agriculture, the conflict between human beings and wild animals is a vital problem to be addressed since it results in the extinction of various resources and the lives of human being is also in danger. The farmers are the major victims in such issues where they lose their crops to be cultivated, livestock, other related properties and resources. Henceforth, such regions and zones should be monitored round the clock to avoid the intrusion of any wild animal that could probably create a negative impact in the field and human lives. The Smart intrusion detection system can be implemented in the field with sensors and controllers that can detect the presence and movements of anonymous objects. This smart system will be implemented with a mechanism to detect and classify the objects with deep learning algorithm on further running of the system, using Content-Based Image Retrieval (CBIR) the captured image is investigated and classified based on the extracted features and alarm only if the intruder is recognized to be an unauthenticated object. Instant notifications will be triggered to the farmers through their mobile applications and the intruder will be reinforced to move out of the field with the alarming sound, which could annoy the object based on its type.