In this paper,the technique of quasi_lossless compression based on the image restoration is presented.The technique of compression described in the paper includes three steps,namely bit compression,correlation removin...In this paper,the technique of quasi_lossless compression based on the image restoration is presented.The technique of compression described in the paper includes three steps,namely bit compression,correlation removing and image restoration based on the theory of modulation transfer function (MTF).The quasi_lossless compression comes to a high speed.The quality of the reconstruction image under restoration is up to par of the quasi_lossless with higher compression ratio.The experiments of the TM and SPOT images show that the technique is reasonable and applicable.展开更多
Discrete asine transform (DCT) is the key technique in JPEG and MPW, chch dds with tw bforkby block. HoWever, this methed is no sultabe for the blocks conaining many edges for high quality image reconstruc-tion in Par...Discrete asine transform (DCT) is the key technique in JPEG and MPW, chch dds with tw bforkby block. HoWever, this methed is no sultabe for the blocks conaining many edges for high quality image reconstruc-tion in Particular. An adaptive hybrid DPCM/DCT edng mehed is proposed to solve this problem. For each block,the ds dethetor botches to DPCM or gy ceder autoInaticthe depewhng upon quality requrement. The edge blocksare coded by DPCM coder that adaptively Selects a predictor from the given set, which results in minimum predictionerror, and the hadues obained are ced with fuce ed. For non-edg bforks, us, mlength nd vallabe lengthcoding(VLC) are applied. Experimental results showed the Proposed algorithm ouperforms baseline JPEG and JPEGlossless mode both on compression ratio and decoding run-time at the hit rates from 1 to 4 approximately.展开更多
In this paper,we present an edge detection scheme based on ghost imaging(GI)with a holistically-nested neural network.The so-called holistically-nested edge detection(HED)network is adopted to combine the fully convol...In this paper,we present an edge detection scheme based on ghost imaging(GI)with a holistically-nested neural network.The so-called holistically-nested edge detection(HED)network is adopted to combine the fully convolutional neural network(CNN)with deep supervision to learn image edges efectively.Simulated data are used to train the HED network,and the unknown object’s edge information is reconstructed from the experimental data.The experiment results show that,when the compression ratio(CR)is 12.5%,this scheme can obtain a high-quality edge information with a sub-Nyquist sampling ratio and has a better performance than those using speckle-shifting GI(SSGI),compressed ghost edge imaging(CGEI)and subpixel-shifted GI(SPSGI).Indeed,the proposed scheme can have a good signal-to-noise ratio performance even if the sub-Nyquist sampling ratio is greater than 5.45%.Since the HED network is trained by numerical simulations before the experiment,this proposed method provides a promising way for achieving edge detection with small measurement times and low time cost.展开更多
针对传统图像压缩比控制不精细及低维混沌系统保密性不高的问题,提出一种基于连续色调静态图像的无损或近无损压缩标准(JPEG-LS)压缩比控制的图像压缩加密算法。在深入分析JPEG-LS中失真控制参数Near对图像压缩比和重建质量的影响的基础...针对传统图像压缩比控制不精细及低维混沌系统保密性不高的问题,提出一种基于连续色调静态图像的无损或近无损压缩标准(JPEG-LS)压缩比控制的图像压缩加密算法。在深入分析JPEG-LS中失真控制参数Near对图像压缩比和重建质量的影响的基础上,首先,对光栅扫描的图像数据进行梯度处理;然后,比较梯度值与Near的大小关系以决定进入游程模式进行游长编码或常规模式进行Golomb编码;再次对三维Lorenz混沌系统生成的序列进行随机性处理,采用该序列作为密钥分别对游程模式、常规模式和全模式(游程和常规两种模式)下的压缩码流进行加密;最后,对Near进行实时动态调整,实现了对图像的压缩比精细控制且提高了保密性。仿真结果表明,所提算法能够实现良好的压缩比控制,且重建图像质量比线性压缩比控制算法提高了大约0.5 d B;同时算法安全性高,能够有效抵抗熵攻击、差分攻击、穷举攻击、统计攻击等多种攻击,且加密对压缩效率基本没有影响。展开更多
In this paper,a new predictive coding algorithm is presented for lossless image compression. This algorithm considers both the local edge and the variance ratio of pixel value in prediction process. It further reduces...In this paper,a new predictive coding algorithm is presented for lossless image compression. This algorithm considers both the local edge and the variance ratio of pixel value in prediction process. It further reduces the entropy of the predictive error image with error feedback technology. Simulation results show that the performance of this algorithm is better than not only the standard algorithm(LOCO_I)provided by JPEG_LS ,but also CALIC, which is the state-of-art in the literature of image compression.展开更多
文摘In this paper,the technique of quasi_lossless compression based on the image restoration is presented.The technique of compression described in the paper includes three steps,namely bit compression,correlation removing and image restoration based on the theory of modulation transfer function (MTF).The quasi_lossless compression comes to a high speed.The quality of the reconstruction image under restoration is up to par of the quasi_lossless with higher compression ratio.The experiments of the TM and SPOT images show that the technique is reasonable and applicable.
文摘Discrete asine transform (DCT) is the key technique in JPEG and MPW, chch dds with tw bforkby block. HoWever, this methed is no sultabe for the blocks conaining many edges for high quality image reconstruc-tion in Particular. An adaptive hybrid DPCM/DCT edng mehed is proposed to solve this problem. For each block,the ds dethetor botches to DPCM or gy ceder autoInaticthe depewhng upon quality requrement. The edge blocksare coded by DPCM coder that adaptively Selects a predictor from the given set, which results in minimum predictionerror, and the hadues obained are ced with fuce ed. For non-edg bforks, us, mlength nd vallabe lengthcoding(VLC) are applied. Experimental results showed the Proposed algorithm ouperforms baseline JPEG and JPEGlossless mode both on compression ratio and decoding run-time at the hit rates from 1 to 4 approximately.
基金supported by the National Natural Science Foundation of China(Grant Nos.61871234 and 62001249).
文摘In this paper,we present an edge detection scheme based on ghost imaging(GI)with a holistically-nested neural network.The so-called holistically-nested edge detection(HED)network is adopted to combine the fully convolutional neural network(CNN)with deep supervision to learn image edges efectively.Simulated data are used to train the HED network,and the unknown object’s edge information is reconstructed from the experimental data.The experiment results show that,when the compression ratio(CR)is 12.5%,this scheme can obtain a high-quality edge information with a sub-Nyquist sampling ratio and has a better performance than those using speckle-shifting GI(SSGI),compressed ghost edge imaging(CGEI)and subpixel-shifted GI(SPSGI).Indeed,the proposed scheme can have a good signal-to-noise ratio performance even if the sub-Nyquist sampling ratio is greater than 5.45%.Since the HED network is trained by numerical simulations before the experiment,this proposed method provides a promising way for achieving edge detection with small measurement times and low time cost.
文摘针对传统图像压缩比控制不精细及低维混沌系统保密性不高的问题,提出一种基于连续色调静态图像的无损或近无损压缩标准(JPEG-LS)压缩比控制的图像压缩加密算法。在深入分析JPEG-LS中失真控制参数Near对图像压缩比和重建质量的影响的基础上,首先,对光栅扫描的图像数据进行梯度处理;然后,比较梯度值与Near的大小关系以决定进入游程模式进行游长编码或常规模式进行Golomb编码;再次对三维Lorenz混沌系统生成的序列进行随机性处理,采用该序列作为密钥分别对游程模式、常规模式和全模式(游程和常规两种模式)下的压缩码流进行加密;最后,对Near进行实时动态调整,实现了对图像的压缩比精细控制且提高了保密性。仿真结果表明,所提算法能够实现良好的压缩比控制,且重建图像质量比线性压缩比控制算法提高了大约0.5 d B;同时算法安全性高,能够有效抵抗熵攻击、差分攻击、穷举攻击、统计攻击等多种攻击,且加密对压缩效率基本没有影响。
文摘In this paper,a new predictive coding algorithm is presented for lossless image compression. This algorithm considers both the local edge and the variance ratio of pixel value in prediction process. It further reduces the entropy of the predictive error image with error feedback technology. Simulation results show that the performance of this algorithm is better than not only the standard algorithm(LOCO_I)provided by JPEG_LS ,but also CALIC, which is the state-of-art in the literature of image compression.
