A Waste Detection and Separation System Based on Image Recognition and Embedded Artificial Intelligence

Abstract

Currently, waste management is one of the most critical issues in Thailand and globally due to the continuous increase in waste volume and improper waste management practices. According to statistics from the Department of Pollution Control, Thailand's waste volume has been increasing annually. As of the latest data in the year 2022, Thailand's total waste volume reached 25.70 million tons per year. This substantial amount of waste can have various adverse effects on the environment, leading to issues such as disease transmission and pollution. To address this problem, there has been ongoing development of technology for waste management and sorting. This presents an intriguing alternative for tackling waste management challenges. If waste can be managed properly and efficiently, it can have positive economic impacts and contribute to environmental development. Recognizing the significance and benefits of waste segregation and proper management, a system for waste detection and sorting based on image recognition and embedded artificial intelligence has been proposed. The system processes images to identify and classify waste, providing notifications regarding the type of waste detected. The model has been trained using deep learning technology, convolutional neural networks, and computer vision-based artificial intelligence. It is operationalized with a prototype robotic platform utilizing highperformance microcontroller boards for processing. The model has been trained with datasets of various types of waste, enabling it to accurately detect and recognize hazardous waste through sensors and robotic platforms. This system has demonstrated superior performance in waste detection and recognition and can operate effectively in diverse real-world environments. Therefore, the waste detection and sorting system based on image recognition and embedded artificial intelligence can help mitigate the risks associated with waste inspection and promote proper and efficient waste management.