Integrasi ESP32-CAM WROVER dan Radar LD2410C untuk Klasifikasi Manusia dan Hewan Berbasis Edge Impulse

Shalma Zopi Habibah (Politeknik Negeri Padang)
Yulindon Yulindon (Politeknik Negeri Padang)

JURNAL ILMIAH PENELITIAN MAHASISWA
Vol 4 No 3 (2026): JUNI

This study aims to design and implement a human and animal classification system based on ESP32-CAM WROVER and LD2410C radar sensor, supported by Edge Impulse and integrated with the Telegram Bot API. The system operates in real-time, utilizing image data from the camera and presence data from the radar to improve classification accuracy. Data collection was conducted directly at the testing site to ensure the model adapts to real environmental conditions. The implementation results show that the integration of camera and radar successfully overcomes the limitations of camera-only systems, particularly under low-light conditions. Furthermore, the use of Telegram as a communication medium provides practical remote monitoring without additional applications. This system demonstrates strong potential for applications in smart homes, security, and wildlife monitoring with low cost and high flexibility.