To predict the deposition of air pollutants on trees Leaf in high traffic area using Machine Learning Algorithm

Globally, as contemporary society continues to grow and urbanization picks up speed, the issue of air pollution is becoming more and more pressing. The two most pressing issues that need to be resolved by humans right now are techniques for forecasting the level of air quality and figuring out the appropriate governance. An increasing number of machine-learning techniques have been applied in recent years to address the air quality prediction issue. Prediction results are, however, significantly impacted by the unpredictability of environmental changes and the challenge of accurately anticipating quantitative values. This research proposes to use a machine learning method to predict air pollution on tree leaves in heavy traffic. The pollutant deposition level is determined using high-resolution camera photos of the roadside tree's leaf. In this paper, we used Linear Regression algorithms with optimization methodologies PSO, DPSO, and FODPSO to predict pollutant levels from leaf pictures, and the AQI value from (HSPCB) values. As a result, the Fractional-Order Darwinian PSO outperforms all other PSO algorithms in terms of accuracy, stability, and speed, making it a preferred choice when accuracy and execution time are of concern.