Research On Multi-description Image Coding Based On Human Visual Characteristics

In recent years,with the development of information technology,multimedia information has made our life rich and colorful.However,packet loss occurs when the information is transmitted on heterogeneous networks,and a variety of image coding methods are proposed to solve the problem.Due to the human eye is the final receptor of visual,how to solve the problems of the data packet loss,and human visual characteristic has become a research focus in the current image coding.In this article,we focus on multiple description image coding,mainly studied multiple description image coding of adaptive quantization step,minimum perceptible distortion of multiple description image coding and adaptive prediction model of multiple description image coding,etc.The main innovative results are as follows:(1)A perceptual multiple description coding with randomly offset quantizers method is proposed in this paper.Due to the human eye is not sensitive to high frequency information,sensitive to low frequency information,therefore this article changes quantization step according to the frequency features of human eye length,which can effectively balance the relationship between the image coding efficiency and reconstruction quality.The different quantization steps are called adaptive quantization step.The experimental results show that the scheme can improve the image reconstruction quality under the premise of ensuring the efficiency of image coding.(2)A JND-based multiple description image coding method is proposed in this paper.As the human eye is the final acceptor of vision,this paper applies the human visual system(HVS)to multi-description image coding.In order to quantitatively measure the HVS perception mechanism and awareness,this paper proposes a Just Noticeable Distortion(JND)model,and applies it in multiple description image coding.The experimental results show that this scheme can further improve the efficiency of image coding by processing DCT coefficient using JND model.(3)A adaptive reconstruction based multiple description coding method is proposed in this paper.The scheme can improve the quality of image reconstruction.Traditional multi-description image coding has either no predictive model or only one predictive model.And this paper proposes two prediction models: mode 1 and mode 2.The optimal reconstruction value of the source is obtained through adaptive selection prediction model.The experimental results show that the proposed scheme is better than the existing ones.