Slope hidden danger identification method and device based on visible light image and infrared image

Abstract

The application relates to the technical field of water conservancy engineering safety monitoring, and discloses a slope hidden danger identification method and device based on visible light images and infrared images, which comprises the following steps: using an image feature extraction network to respectively perform feature extraction on a visible light image of a current slope and an infrared image of the current slope, respectively obtaining a feature vector of the visible light image of the current slope and a feature vector of the infrared image of the current slope, splicing the feature vector of the visible light image of the current slope and the feature vector of the infrared image of the current slope to obtain a current fusion feature vector of the current slope, inputting the current fusion feature vector of the current slope into an artificial intelligence model after training, and generating a hidden danger identification result of the current slope through the artificial intelligence model after training. The application reduces the acquisition time of the hidden danger identification result of the current slope, and is beneficial to improving the acquisition efficiency of the hidden danger identification result of the current slope.