Research On Important Feature Preserving Color Image Enhancement Method

With the rapid development of computer vision,the application field of image enhancement technology is more and more extensive.The essence of image enhancement is to maintain or enhance the important features of the image by adding specific information or performing specific transformations.Among them,color transfer and underwater image enhancement are research hotspots in recent years.At present,the traditional color transfer method has problems such as insufficient layering,loss of details,and inability to preserve the color information of the target image.In addition,the traditional underwater image enhancement method is prone to problems such as poor visibility,color distortion,and insufficient contrast.In view of the above problems,color transfer and underwater image enhancement are deeply studied in this paper with the clue of important feature preservation.The main research contents and innovations are as follows :1.Aiming at the problem that the traditional color transfer method has poor layering and the result image cannot preserve the color information of the target image,a new adaptive color transfer method(CAMR method)is proposed in this paper by combining context awareness and morphological reconstruction.Firstly,the saliency detection algorithm based on context-aware is used to obtain the salient regions and non-salient regions of the source image and the target image,and then the color transfer is performed in the corresponding regions to improve the level of the result image.Secondly,an adaptive segmentation algorithm based on morphological reconstruction is designed.The segmentation matrix representing the main color information of the source image and the target image is obtained for color mapping,so that the result image can better retain the main color features of the target image.Finally,using the segmentation matrix of the source image and the target image,a new color transfer mapping strategy is proposed based on the optimal transmission theory.A large number of experimental results show that the CAMR method proposed in this paper can better preserve the color information of the target image while effectively maintaining the hierarchical sense of the resulting image.The resulting image has better performance in terms of overall visual effect and objective evaluation indicators such as CSP and PSNR.2.Aiming at the problems of poor visibility,color distortion and insufficient contrast in traditional underwater image enhancement methods,a new underwater image enhancement method(FVCA method)is proposed by combining adaptive bilateral filtering algorithm and variational contrast enhancement algorithm.Firstly,the color correction of the underwater image is performed by the red-blue channel prior algorithm to improve the color characteristics of the resulting image.Secondly,a variational contrast enhancement algorithm is designed to enhance the contrast of the color-corrected image to effectively highlight its structural features.Then,the Laplacian pyramid algorithm is used to fuse the saliency weight,brightness weight and saturation weight of the image to improve the visibility of the result image.Finally,the image is processed by the improved adaptive bilateral filtering algorithm,so that the resulting image is clearer and more harmonious.A large number of experimental results show that the proposed FVCA method can effectively solve the problem of color distortion while improving the contrast and clarity of the image.The resulting image has achieved better results in both subjective and objective evaluation.