Color Editing For Complex Natural Images With Repeated Scene Elements

In recent years,with the rapid development of the information technology,the internet technology and digital media technology,the images and video have widely applied in people's lives.The popularity of cameras,camcorders,cell phones,computers,etc.also make people have more and more video image resources.And the demand for processing of visual media has become increasingly strong.In order to meet the needs of thousands of users,we further enhance the full use of visual media and the sharing of communication.And digital media processing technology has also attracted more and more researchers.As a field of computer vision related to color processing Technology has become a hot research,more and more applications involve in film and television,interactive entertainment,animation creation,remote sensing communications,medical imaging and other fields.Although more and more research scholars have embarked on the research of color processing technology,people often inevitably contains a large number of repeated scenes elements when they deal with the target image in the process of color processing technology.These numbers of repeated scenes elements and complex characteristics,the existing technology is still unable to completely solve the key problems of uneven color,border color mixing,coloring inefficient and color overflow.Therefore,this paper focuses on the study of color editing for complex natural images with repeated scene elements and aims to use the similarity between duplicate scene elements in the image and reduce the user's interaction.In this paper,we propose the approach of color editing for complex natural images with repeated scene elements,which is divided into three steps:image sub-block region generation,image sub-block region clustering and image color transmission.And the main contents are as follows:1,In this paper,we first use the linear iterative clustering method to divide the complex natural image with repeated scene elements into a certain number of sub-blocks.In the color editing process,we can use the image sub-block region to take the place of the image pixel,which greatly reduces the complexity of color editing.2,In the process of sub-block clustering of complex natural images,a simple fuzzy C-means clustering method is used,and the color of the target image and texture features of that are analyzed.Even if the natural image has complex repeated scene elements,we also accurately cluster the image sub-block area in a small amount of user interaction.3,In the process of image color transmission,the color assigning mechanism of local feature similarity and global color optimization mechanism are proposed,which can effectively maintain the color continuity of local area and the consistency of global color in color editing results.At the same time,in the local feature similarity distribution mechanism,we filter dissimilarity pixel points through using the setting of the confidence,which can effectively eliminate the impact of pixel points of different classification in the image sub-block area.The color editing method proposed in this paper is mainly applied to the color editing of complex natural images with repeated scene elements.We analyze the quality,the efficiency and the feasibility of the image color editing by collecting a large number of different kinds of complex natural images.The experimental results show that this method can not only reduce the computational complexity effectively,but also improve the accuracy of color editing in a small amount of user interaction.It can ensure the consistency and continuity of color transmission in color editing results and get high quality color editing results.