Research On Adaptive Data Visualization Technology And Implementation

In data visualization,color is an attribute that designers generally pay attention to.Visualization uses color to convey specific information inherited from the source data to the target audience,and the quality of its color coding directly affects the effectiveness of information delivery.Especially for visual charts of categorical data,the choice of categorical colors is a challenging task,as it often requires consideration of factors such as category perception and aesthetics.The task of color matching for visual charts can generally be roughly divided into two steps:palette generation and color mapping.However,most current research tends to focus on only one of the two.Among them,color matching suggestions mainly focus on the distinguishability between categories,the harmony of overall colors,and the semantics of color categories.Especially for the color matching of sub-type charts,good distinguishability is the most basic requirement.Researchers have also proposed various definitions such as class visibility,visual salience,and minimum perceptual distance for quantitative measurement and analysis.optimization.The current automatic color matching method cannot meet the complex and changing needs of users.Whether it is the selection of the palette or the determination of the color distribution mapping,it is very difficult for users who have no experience in color design.In view of the above problems,this paper made the following research contents:(1)A method for automatic color matching of categorical visualization charts using reference image sets is proposed.Some artistic images,celebrity paintings,photographic works,and movie scenes often contain rich and aesthetic color schemes,which naturally have an overall harmonious and excellent aesthetic performance.Inspired by this,an image-guided color scheme is generated for a given classification visualization.A good color scheme should allow users to easily perceive the category structure and be similar to the reference image.The method of automatic color matching for categorical visual charts driven by image sets aims to transfer the color style in images to visual charts.To this end,a large number of colorful pictures are collected to form an image set.By modeling images and graphs into graph structures,the color matching problem is subtly transformed into a graph matching problem,and this method is extended to scenarios such as multi-style graphs,web page layouts,and infographic layouts.(2)Design and develop an interactive web system based on the method in this paper,test and evaluate the effectiveness of the method.The candidate sets of matching results are displayed for different inputs,and a scoring function is specially designed to sort them.In this function,factors such as similarity with the original image,category distinguishability,and color harmony are considered.Users can interactively select pictures that meet their preferences and needs,and the system will automatically generate and display a color scheme based on the selected pictures,allowing users to choose vivid natural images instead of dull parameters to constrain the results.Because it is difficult to define the true value of the color matching problem,a series of user subjectivity surveys are designed for the method proposed in this paper,and the experimental results prove the effectiveness of the method.