Research Of 3D Modeling Of Image And Color Image Coding Based On Correlation Model

The digital color image has been extensively used in each and every domain as one of most important information with the high development of information society. The digital color image is an indispensable information resource. Multicoloured realistic color image gives people more appropriate, more real and more pleasant information. The requirement of color images is increasing in rapid rate and egregious quantity in current society. How to transform and memorize digital color image quickly, real-timely, distinctly and in high efficiency is the key to accelerate the development of computer internetwork communication technology and multimedia technology.Research of 3D modeling and vectorization based on plane digital image (including color image) is very important to accelerate the development, popularization and application of modernization manufacture technology such as digital art carving, 3D spray painting, 3D laser processing and etc. And it is one of the key techniques of intelligentized and Agility Manufacturing.This paper studies problems such as color image coding method, color image coding quality evaluation and image (including color image) vectorization and 3D modeling. The main research content of this paper contains the following aspects:(1) The research history, current situation, development trend and existing problems about image (including color image) compression coding and modeling technology are reviewed. And the research ideas of this paper are presented.(2) The close correlation existing between three color components of RGB color model such as texture correlation, grads correlation, edge correlation, and pixel value and pixel distributing structure correlation is validated by deep analysis and a mass of experiments of color images of frequently-used color models and newly established color models. The correlation function models between three color components are established. Color image hybrid coding method based on color components correlation models using the correlation vectors or correlation coefficients combined with JPEG2000 is presented. In this method, one of the three color components is coded using JPEG2000 coding, and the coding for the other two components becomes the coding of a small quantity of correlation vectors or correlation coefficients. A method especially for color head image compression coding by establishing the color correlation template is presented. This method precisely extracts the facial feature region using a simple and suitable algorithm by feature analysis of color head image. The facial feature region and background region are segmented into 4 sub-regions and their correlation template or common template between components are established to realize grade coding of color head image.(3) An image coding method based on quadric Bezier curved surface is presented. This method syncretized computer graphics and digital image processing technology. By analyzing pixel value and pixel distributing structure of image and according to certain rules, this method selects, computes and compounds pixel value and pixel position parameters of image region block and use them as characteristic parameters for establishing curved surface model of image by quadric Bezier curved surface to realize image coding. And this method is used color image coding. At the same time, the image modeling method based on quadric Bezier curved surface and its application in digital carving technology are studied and discussed. Experiment results of coding methods in this paper and other existing coding methods are synthetically analyzed and compared. A color image coding quality evaluation method syncretized human eye visual characteristics is presented, and relative evaluation models is established.(4) A vectorization processing method especially for the handcraft drawing complex carving image is presented. This method extracts the edge outline of image region using a self-adaptive algorithm, and realizes segmented curve fitting and smooth connection using piecewise cubic Bezier curves. The method of integrative 3D modeling of complex carving surface based on handcraft drawing complex carving image or graphics is presented. The proposed method uses a scanning analysis algorithm, a normal distribution function and a distance function to model and create carving tracks. The methods of delamination, combination and interpolation during modeling process are described as well.The color image coding methods presented by this paper is of the feature for quick coding and decoding. The experimental results show that this method not only improves the coding efficient but also obtains high compression ratio and satisfactory signal to noise ratio and vision effect of the reconstructed image. It is especially suitable for the color image coding with medium or lower signal to noise ratio. The method of color head image coding by establishing color correlation template can give prominence to coding quality of facial feature region of color head image, and greatly improve the coding efficiency and compression ratio. The image vectorization method presented by this paper can realize efficient vectorization processing for handcraft drawing complex carving image, and it is of advantages such as lower algorithm complex degree and good effect of vectorization, which provides technic conditions for realizing digital plane art carving and the following 3D modeling. The integrative 3D modeling method of complex carving surface presented by this paper makes the modeling of complex carving surface more intelligent and more agile. It can accomplish 3D modeling of a complex carving surface in only one process, which provides an efficient way to realize integrative 3D modeling based on handcraft drawing complex carving image or graphics. And it is of wide practicability. The researching achievements have been used for production and obtained good benefits of sociality and economy.The method this paper is validated by a mass of experimental data, effect pictures, objective analysis, practice and application. The advantages and shortcomings of this method and its results are shown.