Research On Cross-medium Imaging Recovery Technology Based On Structured Light

The random fluctuation of the water-air interface in cross-medium imaging will change the relative position of the light entering the imaging surface of the camera,which is manifested in the image that the image is deformed and distorted,which will cause the imaging target to be unrecognizable and hinder the further imaging of the imaging target.Research and exploration.Since there is no law to follow the fluctuations of the water surface,the image deformities are also random and irregular,and cross-medium imaging recovery is very challenging.The research of cross-media imaging recovery technology not only has a key impact on the detection effect of underwater environment,and the development of this technology can also enhance the country's ability to control the territorial waters.Therefore,it is of great significance to carry out research on cross-media imaging recovery technology.Structured light is the current mainstream 3D measurement solution.In this paper,structured light refers to light patterns that carry special information.The characteristic of structured light energy to carry specified light information has important enlightening significance for the research of cross-medium distortion image restoration algorithm in this paper.Aiming at the problems existing in the current cross-medium imaging image restoration algorithm,this paper proposes an image registration algorithm based on structured light and improves it.The algorithm proposed in this paper can greatly shorten the time required for image processing while ensuring the effect of image restoration.The main tasks are as follows:(1)A brief classification and generalization of the current research status of crossmedium imaging image restoration is carried out,the more classic algorithms are listed,and then the advantages and disadvantages are analyzed.(2)The imaging model of the camera and the light propagation characteristics are studied,and the random fluctuation of the water-air interface will cause serious geometric deformation of the cross-medium imaging target.(3)Aiming at the problem of iteratively obtaining reference images based on the image registration restoration algorithm,an image registration algorithm based on structured light is proposed.This algorithm eliminates the repeated iteration steps of the image registration restoration algorithm.It is necessary to input a distorted image to restore the distorted imaging object,which greatly improves the efficiency of image processing.(4)The algorithm proposed above is verified by simulation and compared with the restoration algorithm based on image registration in terms of restoration image quality and time processing efficiency,which proves that the algorithm can guarantee the speed of restoration image quality.Has an absolute advantage.(5)In order to further shorten the image processing time,a local weighted average polynomial fitting algorithm based on structured light is proposed.It is an improvement to the image registration algorithm based on structured light.The difference is that the local weighted average polynomial fitting is used.To replace the registration part of the algorithm.(6)After the registration algorithm based on structured light and the local weighted average polynomial fitting algorithm based on structured light are implemented in C++,experiments are carried out.Experimental results prove that both algorithms have good recovery effects,and the improved algorithm can indeed further shorten the processing time.