Deep Reinforcement Learning Enabled Smart City Recycling Waste Object Classification

The Smart City concept revolves around gathering real time data from citizen, personal vehicle, public transports, building, and other urban infrastructures like power grid waste disposal system. understandings obtained the can assist municipal authorities handle assets services effectually. At same time, massive increase in environmental pollution degradation leads to ecological imbalance is a hot research topic. Besides, progressive development of smart cities over globe requires design intelligent management systems properly categorize depending upon nature biodegradability. Few commonly available wastes are paper, paper boxes, food, glass, etc. In order classify objects, computer vision based solutions cost effective separate out huge dump garbage trash. Due recent developments deep learning (DL) reinforcement (DRL), object classification becomes possible by identification detection wastes. this aspect, designs an intelligence DRL recycling (IDRL-RWODC) model for cities. goal IDRL-RWODC technique detect objects using DL techniques. encompasses two-stage process namely Mask Regional Convolutional Neural Network (Mask RCNN) classification. addition, DenseNet applied as baseline RCNN model, Q-learning network (DQLN) employed classifier.Moreover, dragonfly algorithm (DFA) hyperparameter optimizer derived improving efficiency model. ensure enhanced performance technique, series simulations take place on benchmark dataset experimental results pointed better techniques with maximal accuracy 0.993.