Air Visibility Detection Based on Convolutional Neural Networks

Abstract

The reduced visibility caused by smog often causes serious traffic accidents, and even brings a large number of casualties and incalculable economic losses. With the development of science and technology, haze visibility detection methods have become a research hotspot in the field of image processing and computer vision. Due to the low accuracy, poor generalization ability and long time consumption of traditional image recognition methods, this paper uses deep learning theoretical knowledge to establish a convolutional neural network model, process and identify smog images, and select test set data to verify the model and detect the level of air visibility.