Cost-Efficient Waste Management through AI: An Environmental Accounting Framework Integrating Mask R-CNN and Graph Neural Networks for Industrial Solid Waste Analysis

This study proposes an innovative environmental accounting framework that combines computer vision with cost-benefit analysis to optimize industrial solid waste management. By integrating Mask R-CNN and Graph Convolutional Networks (GCN), we develop a multidimensional feature extraction system that achieves 96.33% classification accuracy while reducing manual auditing costs by 82% compared to traditional methods.