Abstract:
With increasing global concerns over waste management, the issue of efficient and effective garbage disposal has gained significant attention in recent decades. This study aims to address the challenges of manual labor-intensive garbage classification, which can be time-consuming and inefficient, by proposing an intelligent garbage classification system based on machine learning (ML) technology. The proposed system employs a STM32F103ZET6 microcontroller as the main control module and OpenMV4 H7 camera module to capture image data of garbage samples. BRISK feature extraction algorithm is employed for object detection and recognition, while the JQ8400 voice module is utilized for speech broadcast functions. Through testing, it was found that the system's accuracy in accurately categorizing garbage into four classes - kitchen waste, harmful waste, other waste, and recyclable waste - reached up to 98%. This demonstrates that the proposed system is simple to operate, reliable in its performance, and cost-effective compared to traditional manual methods.
