Automated Identification and Classification of Polymeric Waste Wrappers Using Convolutional Neural Networks and IoT-Based GSM Communication

Polymeric waste, especially plastics, is a pressing global environmental challenge. This pioneering research amalgamates computer vision, ResNet50 architecture, ReLU activation functions, and IoT communication to transform polymeric waste detection and management. Leveraging Convolutional Neural Networks (CNNs) like ResNet50, the model achieves outstanding accuracy, surpassing 98.7% in identifying diverse polymeric waste items. The integration of ReLU activation functions enhances the network's ability to discern intricate waste patterns, resulting in an impressive F1 score exceeding 98%. Training with an extensive and meticulously annotated dataset ensures adaptability in recognizing varied polymeric waste forms, guaranteeing high accuracy in real-world applications. The IoT integration through a GSM module has played a pivotal role in enabling real-time communication upon waste detection. Recent successful tests showcased the system's seamless transmission of messages from the detection setup to a control room upon polymeric waste identification. Specifically, real-time wrapper detection illustrated the system's efficacy by promptly notifying the designated control room, showcasing its practical applicability and swift responsiveness. This fusion of computer vision and IoT technologies establishes a versatile framework applicable across diverse environments, reshaping waste management in urban areas, public spaces, and recycling facilities. Its scalability positions it as a promising solution to curb polymeric waste proliferation, proactively reducing the ecological footprint of plastics and fostering environmental conservation. This research introduces a transformative paradigm in polymeric waste management. The integration of advanced technologies not only accurately identifies and categorises waste but also triggers immediate responses, marking a significant stride towards efficient and sustainable waste management practices globally. The successful incorporation of IoT via GSM communication, enabling real-time alerts to a control room, highlights the system's practicality and readiness for immediate integration into waste management infrastructures.