Design And Implementation Of The High Definition Image Generation System Based On Super-resolution Technology

With the arrival of the era of cloud computing and big data,as well as the rapid growth of information quantity,digital signal processing technology has become an indispensable part of all walks of life.For example,the use of spark,Hadoop and other big data framework for large-scale data processing,the research about the learning-based algorithms to collect and calculate the data,etc.The field of digital image processing has also experienced the change from the traditional algorithm to the learning-based algorithm.The super-resolution technology,which is a branch of the field of digital image processing,also reflects this trend.Based on the study of various algorithms for super-resolution technology,this research has designed a high definition image generation system based on super-resolution technology.The calculation of this system relies mainly on visual studio 2015 programming environment,while the mixed programming of MATLAB and C++ takes subsidiary function.MFC and the Windows Ribbon designing interface are put to use.The system realizes several auxiliary function modules and three super-resolution algorithm modules,which include NEDI interpolation algorithm,linear neighbor embedding algorithm and deep convolutional neural networks algorithm.Moreover,to take into account the practical application and the experimental study,the system designs the part of evaluation and analysis as an optional module.When the system runs in professional mode,we can get information about required time of algorithm,and do statistical analysis through the root mean square error(RMSE)and the peak signal-to-noise ratio(PSNR)as performance evaluation indexes.The system intuitively shows the advantages and disadvantages of algorithms under different categories of input image,conditions and parameters,so that the direction of improvement becomes clear.This paper discusses the principle of the three super-resolution algorithms in the system and other related technologies with an analysis of the native and foreign research from the perspective of the significance of the study on image super-resolution technology.At the same time,as a research and practice application system,the system follows the software design and implementation process strictly,which,as well as the test analysis and test results,will be detailed in this paper.According to the test results,the performance merits and defects of various algorithms under different conditions are to be found.