Real-time long-wave infrared fog penetration performance enhancement algorithm

Abstract

The invention discloses a real-time long-wave infrared fog penetration performance enhancement algorithm. The algorithm comprises the following steps: 1) reading a long-wave infrared foggy image and along-wave infrared fog-free image of the same scene point and initializing the long-wave infrared fogless image; 2) obtaining a dark channel graph by using a minimum filtering optimization method; 3)obtaining an atmospheric light value by using a brightness maximum mean value solving method; 4) obtaining transmissivity by using a transmissivity optimization solving method; 5) solving a fog-freeimage by using a formula; the invention has the advantages that the defogging effect is good, the contrast ratio and the dark tone of the image are improved, the gray scale loss and the color spot generation are prevented, the relationship with the expected image is established, and the algorithm time is shortened as much as possible.

Description

technical field [0001] The invention belongs to infrared image processing technology, in particular to a real-time long-wave infrared fog penetration performance enhancement algorithm. Background technique [0002] Image defogging technology is to remove the interference of fog in the image by certain means, so as to obtain high-quality images, in order to obtain satisfactory visual effects and obtain more effective image information. Research on infrared image defogging technology can be widely used in video surveillance, satellite remote sensing, vehicle safety assistance, military investigation, urban traffic, etc. Certain reference value. Foggy image processing can be divided into two categories: foggy image enhancement and foggy image restoration. Among them, image processing-based enhancement methods include global enhancement and localized enhancement, while physical model-based restoration methods include partial differential equation-based , based on deep relation...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a real-time long-wave infrared fog penetration performance enhancement algorithm to solve the problem of image degradation and blurring in different degrees of long-wave infrared foggy images collected in various concentrations of foggy weather, and the image appears darker in tone , the details are not obvious, the contrast is low, it is easy to produce halo and color spots, the dynamic range of the gray value is too small, and the interaction with the desired image cannot be achieved.

Images