Research On The Visualization Technology Based On Pixels And Its Measurement Model

With the rapid development of information technology, the amount of data to deal have been grown exponentially, visualization technology is playing an increasingly important role in the process of data analysis and mining. Meanwhile, the types of visualization technology are also increasing and it also expanded the scope of applications. A variety of visualization technology has its unique advantages and scope of applications, the promotion and in-depth application of visualization techniques need an effective evaluation, Current evaluation of visualization technology has serious limitations, It only focused on the evaluation of the specific visualization system, and often been influenced by human subjectivity, therefore it is usually ineffective.In order to accurately evaluate the effect on the visualization and guide its proper use. The deeply evaluation on itself and building a corresponding quantitative evaluation model is in deeply need.In this paper, We have classified the visualization techniques and seriously delivered on the visualization based on pixels. Then extract the data density, date dimensions, data span, data differ, and the total of visualization as a indicator and then established a quantitative evaluation model. Finally, I designed a software system to achieve this model. The research work of this paper is mainly reflected in the follows:1. This paper proposed new method to improve the traditional pixel technology of visualization, in the process of pixels arrangement, proposed the strategy of pixels arrangement by Peano curve trajectory. In the process of attributes arrangement, this new pixels technology Makes the dimensional attributes arranged in an orderly sequence, allowed to attributes arrange according to the similarity of the smallest similarity.2. This paper deeply study visualization technology based on pixels, then extract six indicators which make a serious influence on visual effect, according to its different importance, we use a chain ration method to give the different weights and then establish a quantitative evaluation model PVEM.3. This paper has designed and implemented PVEMSystem system, including the pixel visual module, visualization module and PVEM quantified evaluation authentication module. Among these modules, Pixel data visualization module is used to visualize the datasets. Quantitative evaluation module is used to analyze the impact of various factors on visualization, Authentication module is used to test the accuracy of quantitative evaluation module.The visualization technology based on pixel can support the visualization of massive datasets. Proper window shape attributes and pixels arrangement by the correct method can contribute to find the implicit reliance on the dataset and the clustering, correlation and other relations. Quantitative evaluation model is used to provide quantitative evaluation for each visualization process so as to guide the proper selection and practical application of specific visualization.