Smart Wildlife Surveillance Leveraging Deep Learning and Sensor Data Analysis
Silpa Chaitanya P, Kurapati Pujitha, T. Revathi, Durga Bhavani Darla, Anusha Siddham · 2024
Wild animal detection is critical for wildlife conservation and human safety in locations where humans and wildlife interact.The purpose of this paper is to create a system for detecting animals that combines PIR,NOIR camera and ultrasonic sensors with a Raspberry Pi controller. These devices collect and transmit photos and environmental data to a central server. YOLO is a deep learning model that is used on the server to recognise and classify wild animals according to predetermined categories by analysing incoming photos in realtime.Automating the detection of wild animals is intended to support conservation efforts and improve public safety in areas where people and wildlife frequently clash. The device can identify the presence of animals in real time by merging data from PIR and ultrasonic sensors. The sensor data is analysed using advanced algorithmic techniques and machine learning to separate out natural environmental occurrences from encounters with wildlife. The system activates actions like warnings or deterrent devices in reaction to the detection of a wild animal, assisting wildlife monitoring and conservation measures. It reduces the requirement for constant human presence, efficiently monitors sizable areas, and compiles useful data for ongoing research. This ground-breaking method aids in the conservation of wildlife.