| With the development of science and technology and the expansion of demand,the ocean and underwater world is more and more valued by people.In order to fully understand and explore the underwater world and carry out various underwater operations and research,we must rely on comprehensive underwater scene information.Image and video,as one of the important media for human to obtain the effective underwater information,plays an important role in our underwater activities in various fields.However,compared with the air medium,the attenuation effect of water on the light is more serious,which is mainly manifested in the selective absorption and strong scattering of the light.And then,this effect results in poor image quality of underwater image,color distortion and poor visibility and other image degradation phenomena.In view of the problems in the current underwater vision system,this paper designs a method of underwater image video clarity,which uses the relevant methods in computer vision to process the underwater image and video.This paper includes two parts: the algorithm of single underwater image and the algorithm of underwater video.The specific research contents are as follows:1.Aiming at the problem that underwater images are similar to atmospheric fog maps,which is caused by the absorption and scattering of the medium,the image quality degradation is analyzed.On this basis,an underwater clarity processing algorithm based on the dark channel prior atmospheric dehaze model is proposed.The algorithm improves the dark channel prior dehaze model from the characteristics of underwater image processing Finally,through the experiment comparison,the good performance of the method in the single underwater image processing is verified;2.In order to solve the problems of video frame sequence jam,tone inconsistency and long processing time in underwater video processing,this paper proposes an improved underwater video clarity processing method to solve the above problems based on the single underwater image processing and integrating the slow change correlation between video frames.This method analyzes that the main cause of the problem is the estimation of the global background light intensity,and improves the process.Based on the estimation area of the previous frame of the video sequence,the domain search quickly locates to the estimation area of the current frame.This method effectively reduces the estimation difference,shortens the time used,and solves the above problems.At the same time,through visual studio 2010 development platform,relying on computer open-source visual library OpenCV,and through MFC programming,this paper designs and realizes user-friendly software system,which makes underwater image and video processing algorithm combined with software programming through the form of interface interaction,provides users with a variety of functional modules,reflecting its practical application value. |
