Detection and monitoring of products inside chillers are important for efficient storage management, especially in the retail, food, and pharmaceutical sectors. Manual checking methods, which are still widely used, have several limitations, such as requiring additional labor and being prone to human error. To address these issues, this study develops an automatic, microcontroller-based system capable of detecting products without direct human intervention. The system employs an ESP32-CAM as the main processing unit for image capture, combined with a VL53L0X Time-of-Flight LiDAR sensor for distance measurement. Image data are processed using binary thresholding and simple blob detection to identify object regions, while distance data from the LiDAR sensor are used to validate object presence within a defined detection range. The object status is determined from the detected image region and the measured distance, and the result is displayed locally through a web server. The two data streams are processed in combination so that product presence can be identified while maintaining a compact and low-power implementation inside the chiller. Implementation results show that the system can detect objects within an effective range of approximately 10 cm to 1 meter, achieving a detection success rate of around 90 percent with a false-detection rate of less than 5 percent, and a system response time of 1-2 seconds from data acquisition to display. Therefore, the combined ESP32-CAM and LiDAR-based system can improve the efficiency of product-presence monitoring in a chiller, while still providing opportunities for further development through advanced AI-based visual analysis and IoT integration.
Copyrights © 2026