Nonnegative Matrix Factorization For Temporal Psycho Visual Modulation

With the rapid development of science and technology,more and more information display technologies are invented and studied by scientists.Many of these technologies have been used in our daily life.As the popularity of HD display devices,nowadays,pursuing high resolution or good quality can not satisfy the researchers,the studies in information display are on the way to implement new applications.The old display technology is just a conversion from analogy to digital,there are few studies on using the information processing technology to enhance the display ability.These current years,a display technology called Temporal Psycho Visual Modulation(TPVM)was proposed,it aims to generate multiple images on the single medium.Although other technologies,like time multiplexing,can realize this function,TPVM uses the advantage of the relationship between the input information reasonably.So,TPVM is conceived as an ingenious interplay of signal processing,optoelectronics,and psychophysics.According to the idea of TPVM,researchers have extended this technology into different applications,like the hot information security technology.This paper firstly introduces the principle of TPVM,analyze this technology in mathematical way,it leads the reader to find out that TPVM can be implemented by Nonnegative Matrix Factorization(NMF).For this purpose,this thesis describes the development of NMF,and depicts the process of the algorithms by different classifications.This is the fundamental knowledge to study TPVM.In the early studies on TPVM,researchers more focused on the images perceived through modulation devices.Actually,the functionality of glass free view,if attainable,is much desired or even necessary in many applications.Thus,this paper reorganizes the cost function,introduces a well-performed NMF algorithm,refined the iteration process,to propose a algorithm to implement glass free view in TPVM.This algorithm can ensure the realization of glass free view,meanwhile,nearly do no harm to the images perceived through modulation devices.As the popularity of color images and color videos,TPVM faces a challenge to process the color images or videos.The size of color image is three times than the size of its corresponding gray image.The huge size will slow down the TPVM to handle the color images or videos.According to the conversion between RGB formant and YUV format of color images,this thesis uses the advantage of the characteristic of YUV format to propose a algorithm to accelerate TPVM.This algorithm can save much run time when TPVM processes color images or videos.The idea of using the information redundancy in time domain has been extended into spatial domain.Given the feature of Human Visual System(HVS),Spatial Psycho Visual Modulation(SPVM)can use the information redundancy provided by high resolution display devices.This paper transplant this technology into android platform to implement a information hiding system,this system can hide the security information to prevent your privacy from invading.