An approach of fog detecting magnitude using referenceless perceptual image defogging

The increasing economy is resulting in heavy pollutions and foggy environments in my regions, which raises up a problem of detecting the magnitude of the fog or smog. Creating an alert system to classify the magnitude of fog is for numerous fields, such as the transportation security, manufacturing management, and disease prevention. However, current fog detection solutions require a long training process from feature extractions to classifications. This paper focuses on the issue of the fog detection and proposes a new approach for classifying the magnitude of fog using the technique of image defogging. The proposed method, called Image Defogging-based Detection for Magnitude of Fog (IDDMF), emphasizes the perceptual fog density considering image corpus from both foggy and fog-free images. The experiment evaluations have provided an assessment for demonstrating the performance of the proposed approach.