Publication

Abstract : In the context of wildlife monitoring, the challenge of real-time animal detection in surveillance systems has been addressed using FPGA-based hardware acceleration. To this end, this paper proposes a novel system leveraging a modular architecture to enhance detection efficiency while maintaining low latency and power consumption. The design integrates key modules, including edge detection employing Sobel filtering for object boundary enhancement, motion analysis based on frame differencing to identify significant movements, and object classification using predefined size and motion thresholds. A parallel processing strategy is utilized to optimize computational performance. Simulation and validation conducted in ModelSim demonstrate the system’s capability to detect and track animal motion efficiently in various scenarios and dynamic conditions. Metrics such as detection responsiveness and computational efficiency evaluate the system’s performance. Comparative experiments underscore the scalability and practicality of the proposed FPGA-based approach in addressing human-wildlife conflicts, particularly in conservation areas and public spaces. The results indicate that the system provides a cost-effective and energyefficient alternative to traditional CPU and GPU solutions for edge computing applications in real-time wildlife monitoring.