Underwater thermal disturbance image restoration system and method based on deep learning

Abstract

The invention discloses an underwater thermal disturbance image restoration system and method based on deep learning. The system comprises a barrel body, a target plate is arranged at the center of the inner wall of the bottom of the barrel body, and heating rods are arranged at the two sides of the target plate, and are used as heat sources for forming thermal disturbance. An underwater camera is arranged at the position, located over the target plate, of the inner cavity of the barrel body, and an electric telescopic rod is arranged at the top end of the underwater camera and used for obtaining target plate images at different distances. Compared with an undisturbed image, the underwater thermal disturbance optical imaging method selects a proper image quality evaluation standard based on an image processing technology, analyzes the law of underwater thermal disturbance optical imaging, designs a deep learning model based on a deep learning method according to the obtained law, and restores the underwater thermal disturbance image so as to obtain clear underwater scene information.

Description

technical field [0001] The invention relates to the technical field of underwater thermal disturbance image restoration, in particular to a system and method for underwater thermal disturbance image restoration based on deep learning. Background technique [0002] The ocean occupies 71% of the earth's surface. In the future, the exploration of the unknown ocean will undoubtedly be one of the major topics of human research. However, when image acquisition is carried out underwater, the image quality will be affected by many factors, such as water turbidity, water absorption, and uneven temperature distribution of the water environment. Underwater thermal disturbance is one of the important factors. It is caused by the temperature field formed by the underwater heat source, accompanied by the dynamic field caused by the uneven density change of the underwater environment. According to Snell's law, when the water body continues to flow, the refractive index of light in the wa...

AI Technical Summary

Problems solved by technology

However, most of its research objects are in an ideal environment, and there are very few image restorations in complex water environments.

However, in the case of turbulent water flow, when there is a heat source in the imaging optical path, the heat flow superimposed on the water flow will produce a non-uniform dynamic field, which will directly lead to distortion and blurring of the collected image. not uncommon in

Images