Research On Image Quality Assessment System Based On MATLAB GUI

With the speedy development of the Internet technology, digital image is widely used in many fields of human life. The image quality is good or bad will directly affect accuracy of acquired information. So, the image quality assessment has become a basic problem. Since 1980’s, the research about image quality assessment has never been interrupted. From the full reference assessment to reduced reference assessment, and then to no-reference process of research, has countless image quality evaluation algorithm came into being. The full reference assessment which is need to have complete information about the original image as a reference, reduced reference assessment which is need only part of the original image information as a reference, and no-reference which is not need original image information as a reference. Each algorithm has its own unique way of thinking, as well as feature extraction methods. Therefore, how to find the optimal image quality assessment algorithm suitable for a certain kind of distorted image quickly and intuitively from numerous image quality assessment algorithm is very valuable.MATLAB has a powerful matrix computing capabilities, rich library, convenient toolbox, as well as visual GUI interface platform, etc. It is more convenient to deal with images, and is very popular among researchers. Therefore, in order to comparative analysis of advantages and disadvantages of numerous image quality evaluation algorithm and its applicable range of distortion types, the article designs an image quality assessment system based on the visual GUI interface platform provided by MATLAB. The specific contents are as follows:(1) firstly, The thesis deeply discussed the research background and significance of the visual image quality assessment system based on MATLAB. Secondly, Analysis of the current domestic and international research results and development on the Visual image quality assessment system, evaluate the performance of quality assessment algorithm and image quality assessment. Finally, the author mainly introduced the research on the full reference and no-reference assessment.(2) Simply introduced the correlative theory and techniques of MATLAB GUI visualization platform. It provides the System programming for system programming.(3) Detailed introduced the implementation of classical image quality assessment, provides the basic arithmetic for the design of image quality assessment system.(4) Writing the callback function for every model in the interface and menu, and then test and analyze every model.The system main implement some function in image quality assessment, with being convenient to assessment algorithm. For example, encapsulation, comparative analysis and evaluate the performance of image quality assessment algorithm, and leave the interface for system of secondary development. The image file format include of BMP, JPEG,GIF, PNG etc are all supported to use in this system; while also support to read, write, display and quality assessment the image type include of gray image and color image. In addition, there are a lot of functions in this system, for example, image processing module includes adding noise, gray, distortion and JPEG compression; open file module includes open the image or file, save the image or file, exit system etc; assessment algorithms library module includes add a new algorithm, modify, delete and selection algorithm; algorithm assessment module include evaluate the performance of algorithm; assessment algorithm analysis module includes comparative analysis of full reference and no-reference assessment algorithm; other modules includes show the current date/time of windows system, simple maintenance of the system. Is used in the experimental part of this paper, have five kinds of(SSIM, MSSIM, MSE, PSNR, Universal) full reference and six kinds of(HFSSIM, HSPSNR, NIQE, DIIVINE, JPEG, new SVD) no-reference assessment algorithms.