An Imaging Prognosis Model for Particle Pollution

Although the ongoing evolution of evaluation methodologies within the realm of Artificial Intelligence has urged the growth of numerous solution models for environment, a definite solution to accurately predict and forecast the environmental conditions still persists. Currently, evaluating the amount of particle matter(PM) is done using control stations that are set up in specific locations and can only measure air quality to a certain degree. In this paper, IoT and ML techniques are employed to predict PM emissions and concentration levels. The proposed prognosis approach leverages ML algorithms and AI techniques to construct an evolving technique for predicting the concentration of PM2.5 from the image data, utilizing data from the regions with the highest observed PM2.5 levels. The results of the ML analysis indicate that image-based estimation methods offer accurate and precise measurements for PM concentration.