Research On Quality Evaluation Method Of Photoelectric Image

Image carrying history,it is the main source of information for human begings.The rapid development of science and technology,Image products are rich in people's life,and their quality is also more and more attention by scholars.Digital image in the acquisition,display,compression,transmission,storage,etc.at the same time,it will inevitably cause the distortion of the image and quality,These distortions will affect the quality of the image.This has brought great difficulty to the information acquisition or image processing,Therefore,how to evaluate the image quality has become a hot research topic in the field of image processing.This paper is aimed at the quality of photoelectric image,There will be a reference image quality evaluation method and no reference image quality evaluation method research content and research results are summarized as follows :In the process of image compression,image restoration and multimedia applications,it is an important task to evaluate the quality of the output image effectively and objectively.A good image quality evaluation model not only can provide high quality forecast accuracy,but also can be very fast.Because of the high speed network running under high video and image,the image quality measurement efficiency becomes very important.This paper provides a new effective image quality assessment model called gradient mode structure similarity error(GSSD),When the local structure of the distorted image is attenuated by different degrees,the image gradient is extremely sensitive to image distortion,This motivates us to study the role of global variables based on local quality images in order to predict global image quality.We find a combination of the standard deviation of the standard deviation of the reference and distorted images in the pixel gradient mode similarity(GMS)with a new gradient mode gradient image.Can predict the accuracy of perceived image quality.The result of GSSDalgorithm is faster than the state-of-the-art image quality assessment(IQA),and delivers highly competitive prediction accuracy.Quality evaluation for no reference image,In this study,a method based on local image gradient matrix singular value decomposition is discussed.It can well evaluate the content of the image at the same time by the fuzzy and noise distortion.For anisotropic block region ?It realizes the estimation of the structure of the image by the singular value decomposition.Experiments show that the results of this measurement are well correlated with the subjective evaluation.