Research Of Sunflicker Removal Method Of Underwater Image

Both in neritic regions and abyssal regions, the underwater imaging system iseasily disturbed by some interferences, such as low visibility, low contrast, complexnoises, geometric distortion and the non-uniform illumination etc. However, theunderwater imaging system in neritic regions is easily affected by dynamic fluctuantsunflicker which enormously degrades the quality of the image. This thesis focuseson the study of this issue, and ultimately a sunflicker removal algorithm based onwavelet transform is proposed.First, the mathematical model of the underwater image is studied in this thesisbased on the classical image model. And then an improved image model isestablished to describe the image with sunflicker. The sunflicker removal algorithmbased on wavelet transform is proposed. The analysis of the sunflicker image modelis given in this thesis. Several images are collected in actual underwaterenvironments. The image acquisition system platform is built and the distortion ofthe camera len is tested. The actual images are preprocessed through image featurematching, wrong matching removal and perspective transform, to acquire theregistration image sequences on the VS2010software platform. On the Matlabsoftware platform, the registration image sequences are then used to test the validityof the sunflicker removal algorithm proposed in this thesis. The experiments showthat the algorithm in this thesis can remove the influence of the sunflickereffectively while keeping the main image character details.