激光微纳表面处理技术中二维数字图像半色调方法研究
|
摘要
长期以来,高质量和高速度制版一直是我国凹印行业发展的瓶颈,也是凹印制版领域亟待解决的关键技术。获得高质量的输出图像是提高凹版制版技术水平的有效途径。激光在这一领域的应用有如下特点：加工效率高、聚焦精确和具备数字调制的优势,因此直接激光微构造技术已经取代了传统凹版制版技术。激光凹版制版的雕刻方式是一种模仿打印输出设备的扫描式雕刻法。数字半色调是计算机输入／输出的重要支撑技术,解决了仅具备二值再现能力的设备无法直接输出多灰度级图像的矛盾。对于多灰度图像,在输出前必须通过数字图像半色调技术将之转换为适合激光输出的黑白二值图像,所以多灰度图像的数字半色调算法对激光图像输出效果起着尤为关键的作用。
本文以凹版印版滚筒网穴为研究对象,针对激光雕刻凹版技术的实际打样应用需求,利用MATLAB7.0.1仿真软件、德国SCHEPERS激光雕刻机、MZD0745C连续变倍单筒视频显微摄取仪、SH-DXA凹版打样机和Epson Perfection V300扫描仪等测试仪器,重点研究适应于激光雕刻多灰度图像的数字半色调技术以及建立激光雕刻凹版网穴的质量检测和分选系统。
首先针对激光雕刻凹版制版过程中输出图像质量不稳定和制版速度受到图像处理时间快慢的限制,提出了多灰度图像数字半色调算法,新算法引用和应用了K-means聚类法、加权最小平方法、Harr二维离散小波变换和改进的直接二值搜索算法等技术手段。针对图像平滑区域质量的提高,提出基于K-means聚类的分区域数字半色调技术提高了半色调图像质量,降低了计算复杂度,减少了计算所需时间。但K-means聚类算法对初始中心点的选取敏感,算法易陷入局部最优解,因此应用改进的K-means聚类算法进行图像分割,使分割结果更加准确,采用加权最小平方法进行迭代优化,使迭代速度得到进一步的提高,加快了图像半色调化的进程。针对应用K-means聚类法及改进的方法只考虑了图像平滑区域,而未考虑图像边缘信息,因此采用小波域多尺度信息融合的方法以获得图像边缘信息,使半色调图像质量和制版速度得到一定的提高。
然后针对凹版印版滚筒上激光雕刻多灰度图像的质量取决于滚筒上网穴的质量,应用机器视觉和神经网络方法建立凹版网穴的质量检测和分选系统,对激光雕刻凹版网穴质量进行了正确的判定。最后将激光雕刻的凹版印版在凹版打样机上进行打样,对样张进行扫描后,将印刷扫描样张与本文数字半色调仿真实验图像对比,应用基于人类视觉系统半色调图像质量评价方法对扫描样张和不同半色调方法仿真实验结果进行质量评价。
研究结果表明：基于K-means聚类法的分区域策略使迭代优化发生在区域内部,如果在某个区域出现两次迭代误差相差不大,就停止了此区域的迭代。在算法中采用人类视觉模型和二值输出设备模型,所构造的半色调算法与基于模型的最小平方算法相比,随着聚类分区的增加,图像平滑且清晰度增加,尤其是在图像细节部位。与基于模型的最小平方算法比较,均方误差降低了0.0129～0.2102,权重信噪比增加了1.4938～2.2072,峰值信噪比增加了0.53～3.17。模拟实验结果验证了算法的有效性。应用改进的K-menas聚类法进行灰度图像分割,通过分析各区域像素均值和方差发现随机误差项具有异方差性,应用加权最小平方法构建新的能量函数,采用各个分割区域像素方差倒数做为权重因子,进行半色调图像的转换。算法4次迭代计算以后,收敛误差降到0.20以下,具有较快的收敛速度。对实验图像进行频域分析,分析结果表明随着聚类分区的增加,频谱图上“亮点”逐渐趋向于分布均匀,所提出算法得到的半色调人眼视觉效果优于基于模型的最小平方算法,灰度值分布均匀,图像平滑过渡。在多尺度信息融合的半色调方法中,混合误差测度函数的建立是利用多尺度小波系数融合建立的边界误差测度函数和改进的K-means聚类法建立的区域误差测度函数。应用直接二值搜索方法最小化初始图像和半色调图像的误差,基于模型的最小平方算法在迭代15次时收敛误差降低到0.01以下,而多尺度信息融合算法只需迭代8次,算法具有较快的收敛速度。通过分析半色调噪声频谱图发现,半色调噪声谱仅分布在中、高频段,无很强的周期分量,且局部的频谱分量从低到高逐步增加,人眼视觉不敏感。
利用网穴摄取仪提取激光雕刻凹版网穴图像,当阈值为16时坎尼算子提取的边缘效果最好。根据本文研究的图像网穴雕刻工艺确定标准通沟值和暗调值,当通沟值和暗调值同时在标准范围内时,输出1该网穴判定为合格,输出0判定为不合格。应用神经网络建立凹版网穴的质量分选系统,基于模型最小平方法雕刻网穴分选结果处于不合格类中,应用本文3种半色调方法雕刻网穴分选结果均处于合格类中,得到的网穴合格率均在98%以上。
应用了基于人类视觉系统的质量评价方法,其评价结果既能描述边缘细节的保持情况,又能反应半色调纹理抑制水平。对同一种扫描图像进行不同半色调方法处理后,结果表明,多尺度信息融合方法和加权最小平方法梯度幅值和与纹理熵均低于基于模型半色调方法,角二阶矩均高于基于模型最小平方法。
For a long time, high-quality and high-speed plate-setter have been a bottleneck during the development of gravure industry gravure engraving field of key technologies to be solved. Access to high-quality output image is an effective way to improve the level of intaglio plate technology. Laser carving technology has the advantages of higher resolution, focusing precision and digital modulation. Direct laser micro-structure technology has replaced the traditional intaglio plate technology. It has been widely used in laser engraving gravure plate and laser plate processing. Laser engraving carving is a kind of imitation print output device scanning engraving method. Gray scale image halftone algorithm plays a crucial role on laser image output effect beyond of the laser wavelength, pulse frequency, quality of laser beam and processing properties of materials and other factors. Digital image halftone is a continuous tone image such as laser engraving machine, digital printing machine, laser printers and other equipment on the development and value of two in the human visual system to generate a continuous tone image illusion of key technology. In the process of production, it gets more and more extensive application. At present, digital halftone technology is ubiquitous from the home office small desktop inkjet, laser printer, laser engraving machine to large publishing and printing system.
In this thesis, the laser engraving the gravure cylinder was studied. The actual proofing applications demand for the technology of laser engraving intaglio. Digital halftone technology as well as the establishment of laser engraving gravure quality of cells and focusing on the detection and sorting system to adapt to the laser engraving gray-scale images using MATLAB7.0.1simulation software, German SCHEPERS laser engraving machine, MZD0745C zoom monocular video microscope intake instrument and SH-DXA gravure proffer and Epson Perfection V300.
Firstly, the laser engraving gravure plate-making process output image quality is unstable and plate-making speed by the speed of the image processing time limit, gray-scale images digital halftone algorithm, new the algorithm references and application of the K-means clustering method, the weighted least squares method, the Harr two-dimensional discrete wavelet transform and improved direct binary search algorithm techniques. For the improvement of the quality of the image smoothing area is proposed based on K-means clustering sub-regional digital halftone technology to improve the quality of halftone image, reducing the computational complexity, reduces the computation time required. K-means clustering algorithm on the selection of the initial centers is very sensitive, the algorithm is easy to fall into local optimal solution, so the improved K-means clustering algorithm for image segmentation, the segmented image is more accurate, using the weighted least squares the method of iterative optimization, the iteration speed has been further improved to accelerate the speed of image halftone. Only consider the application of K-means clustering method and the improved method image smoothing area, without considering the image edge information, so the use of the wavelet domain multi-scale information fusion method to get the image edge information, the halftone image quality and speed of plate making a certain improvement was obtained.
The results show that the sub-region strategy based on the K-means clustering method of iterative optimization occurs within the region, which the area twice iterative error less iteration then this area stops. Halftoning algorithms constructed with the least squares algorithm based on the model, compared with the increase in the clustering partitions, image smoothing, and the sharpness is increased, especially in the image detail parts. Compared with the least squares model-based algorithm, the mean square error is reduced0.0129to0.2102, the signal-to-noise ratio of the weight is increased from1.4938to2.2072, the peak signal-to-noise ratio is increased from0.53to3.17. The simulation results verify the effectiveness of the algorithm. The application improved K-means clustering gray scale image segmentation by random error term heteroscedasticity, which application of weighted least squares method to build a new energy function, each divided region pixel variance countdown, found by analyzing the pixel mean and variance of the regional as the weight factor, and conversion of the half-tone image. After the calculation of the4iterations of the algorithm, the convergence error drops below0.20, and has a faster convergence rate. Experimental image frequency domain analysis and the results show that with increasing clustering partition, the "bright spots" on the spectrogram gradually tend to be evenly distributed, the proposed algorithm halftone human visual effect is better than the least squares model-based algorithm gray scale values evenly distributed, the image smooth transition. Multiscale information fusion halftone method, mixed error measure function of the establishment of multiscale wavelet coefficients fusion to establish the boundary error measure function and improved K-means clustering method to establish the area error metric function. The applications direct binary search method to minimize the error of the initial image and halftone images, the least squares model-based algorithm in iteration15when convergence error is reduced to less than0.01, multiscale information fusion algorithms simply iterate8times, the algorithm has faster convergence rate. By analyzing the halftone noise spectrum Figure found the halftone noise spectrum is only distributed in the high frequency band, no strong periodic component and the local spectral components from low to high and gradually increase the human visual system is not sensitive.
The image of cells extracts the edge effect using canny operator when the best threshold equal to16. According to the image cells of this study engraving process to determine standards through trench values and dark tone value, when the value of the pass ditch and dark tone values in the standard range, the output1of the network points determination the qualified output deemed to have failed. Neural network to establish the quality of gravure network points sorting system, based on the model of the least squares method engraving network points separation results in a the unqualified class, the application of this article half toning method engraving network points of election results are in a qualified class, get the network points pass rates of98%or more.
Application of evaluation method based on the quality of the human visual system, the evaluation results can describe the hold of the edge detail, but also the reaction halftone texture suppression level. During the processing of the image halftoning, the results show that the multiscale information fusion method and weighted least squares gradient amplitude and texture entropy, angular second moment were higher than those based on model-based half toning method were lower than model the least squares method.
本文以凹版印版滚筒网穴为研究对象,针对激光雕刻凹版技术的实际打样应用需求,利用MATLAB7.0.1仿真软件、德国SCHEPERS激光雕刻机、MZD0745C连续变倍单筒视频显微摄取仪、SH-DXA凹版打样机和Epson Perfection V300扫描仪等测试仪器,重点研究适应于激光雕刻多灰度图像的数字半色调技术以及建立激光雕刻凹版网穴的质量检测和分选系统。
首先针对激光雕刻凹版制版过程中输出图像质量不稳定和制版速度受到图像处理时间快慢的限制,提出了多灰度图像数字半色调算法,新算法引用和应用了K-means聚类法、加权最小平方法、Harr二维离散小波变换和改进的直接二值搜索算法等技术手段。针对图像平滑区域质量的提高,提出基于K-means聚类的分区域数字半色调技术提高了半色调图像质量,降低了计算复杂度,减少了计算所需时间。但K-means聚类算法对初始中心点的选取敏感,算法易陷入局部最优解,因此应用改进的K-means聚类算法进行图像分割,使分割结果更加准确,采用加权最小平方法进行迭代优化,使迭代速度得到进一步的提高,加快了图像半色调化的进程。针对应用K-means聚类法及改进的方法只考虑了图像平滑区域,而未考虑图像边缘信息,因此采用小波域多尺度信息融合的方法以获得图像边缘信息,使半色调图像质量和制版速度得到一定的提高。
然后针对凹版印版滚筒上激光雕刻多灰度图像的质量取决于滚筒上网穴的质量,应用机器视觉和神经网络方法建立凹版网穴的质量检测和分选系统,对激光雕刻凹版网穴质量进行了正确的判定。最后将激光雕刻的凹版印版在凹版打样机上进行打样,对样张进行扫描后,将印刷扫描样张与本文数字半色调仿真实验图像对比,应用基于人类视觉系统半色调图像质量评价方法对扫描样张和不同半色调方法仿真实验结果进行质量评价。
研究结果表明：基于K-means聚类法的分区域策略使迭代优化发生在区域内部,如果在某个区域出现两次迭代误差相差不大,就停止了此区域的迭代。在算法中采用人类视觉模型和二值输出设备模型,所构造的半色调算法与基于模型的最小平方算法相比,随着聚类分区的增加,图像平滑且清晰度增加,尤其是在图像细节部位。与基于模型的最小平方算法比较,均方误差降低了0.0129～0.2102,权重信噪比增加了1.4938～2.2072,峰值信噪比增加了0.53～3.17。模拟实验结果验证了算法的有效性。应用改进的K-menas聚类法进行灰度图像分割,通过分析各区域像素均值和方差发现随机误差项具有异方差性,应用加权最小平方法构建新的能量函数,采用各个分割区域像素方差倒数做为权重因子,进行半色调图像的转换。算法4次迭代计算以后,收敛误差降到0.20以下,具有较快的收敛速度。对实验图像进行频域分析,分析结果表明随着聚类分区的增加,频谱图上“亮点”逐渐趋向于分布均匀,所提出算法得到的半色调人眼视觉效果优于基于模型的最小平方算法,灰度值分布均匀,图像平滑过渡。在多尺度信息融合的半色调方法中,混合误差测度函数的建立是利用多尺度小波系数融合建立的边界误差测度函数和改进的K-means聚类法建立的区域误差测度函数。应用直接二值搜索方法最小化初始图像和半色调图像的误差,基于模型的最小平方算法在迭代15次时收敛误差降低到0.01以下,而多尺度信息融合算法只需迭代8次,算法具有较快的收敛速度。通过分析半色调噪声频谱图发现,半色调噪声谱仅分布在中、高频段,无很强的周期分量,且局部的频谱分量从低到高逐步增加,人眼视觉不敏感。
利用网穴摄取仪提取激光雕刻凹版网穴图像,当阈值为16时坎尼算子提取的边缘效果最好。根据本文研究的图像网穴雕刻工艺确定标准通沟值和暗调值,当通沟值和暗调值同时在标准范围内时,输出1该网穴判定为合格,输出0判定为不合格。应用神经网络建立凹版网穴的质量分选系统,基于模型最小平方法雕刻网穴分选结果处于不合格类中,应用本文3种半色调方法雕刻网穴分选结果均处于合格类中,得到的网穴合格率均在98%以上。
应用了基于人类视觉系统的质量评价方法,其评价结果既能描述边缘细节的保持情况,又能反应半色调纹理抑制水平。对同一种扫描图像进行不同半色调方法处理后,结果表明,多尺度信息融合方法和加权最小平方法梯度幅值和与纹理熵均低于基于模型半色调方法,角二阶矩均高于基于模型最小平方法。
For a long time, high-quality and high-speed plate-setter have been a bottleneck during the development of gravure industry gravure engraving field of key technologies to be solved. Access to high-quality output image is an effective way to improve the level of intaglio plate technology. Laser carving technology has the advantages of higher resolution, focusing precision and digital modulation. Direct laser micro-structure technology has replaced the traditional intaglio plate technology. It has been widely used in laser engraving gravure plate and laser plate processing. Laser engraving carving is a kind of imitation print output device scanning engraving method. Gray scale image halftone algorithm plays a crucial role on laser image output effect beyond of the laser wavelength, pulse frequency, quality of laser beam and processing properties of materials and other factors. Digital image halftone is a continuous tone image such as laser engraving machine, digital printing machine, laser printers and other equipment on the development and value of two in the human visual system to generate a continuous tone image illusion of key technology. In the process of production, it gets more and more extensive application. At present, digital halftone technology is ubiquitous from the home office small desktop inkjet, laser printer, laser engraving machine to large publishing and printing system.
In this thesis, the laser engraving the gravure cylinder was studied. The actual proofing applications demand for the technology of laser engraving intaglio. Digital halftone technology as well as the establishment of laser engraving gravure quality of cells and focusing on the detection and sorting system to adapt to the laser engraving gray-scale images using MATLAB7.0.1simulation software, German SCHEPERS laser engraving machine, MZD0745C zoom monocular video microscope intake instrument and SH-DXA gravure proffer and Epson Perfection V300.
Firstly, the laser engraving gravure plate-making process output image quality is unstable and plate-making speed by the speed of the image processing time limit, gray-scale images digital halftone algorithm, new the algorithm references and application of the K-means clustering method, the weighted least squares method, the Harr two-dimensional discrete wavelet transform and improved direct binary search algorithm techniques. For the improvement of the quality of the image smoothing area is proposed based on K-means clustering sub-regional digital halftone technology to improve the quality of halftone image, reducing the computational complexity, reduces the computation time required. K-means clustering algorithm on the selection of the initial centers is very sensitive, the algorithm is easy to fall into local optimal solution, so the improved K-means clustering algorithm for image segmentation, the segmented image is more accurate, using the weighted least squares the method of iterative optimization, the iteration speed has been further improved to accelerate the speed of image halftone. Only consider the application of K-means clustering method and the improved method image smoothing area, without considering the image edge information, so the use of the wavelet domain multi-scale information fusion method to get the image edge information, the halftone image quality and speed of plate making a certain improvement was obtained.
The results show that the sub-region strategy based on the K-means clustering method of iterative optimization occurs within the region, which the area twice iterative error less iteration then this area stops. Halftoning algorithms constructed with the least squares algorithm based on the model, compared with the increase in the clustering partitions, image smoothing, and the sharpness is increased, especially in the image detail parts. Compared with the least squares model-based algorithm, the mean square error is reduced0.0129to0.2102, the signal-to-noise ratio of the weight is increased from1.4938to2.2072, the peak signal-to-noise ratio is increased from0.53to3.17. The simulation results verify the effectiveness of the algorithm. The application improved K-means clustering gray scale image segmentation by random error term heteroscedasticity, which application of weighted least squares method to build a new energy function, each divided region pixel variance countdown, found by analyzing the pixel mean and variance of the regional as the weight factor, and conversion of the half-tone image. After the calculation of the4iterations of the algorithm, the convergence error drops below0.20, and has a faster convergence rate. Experimental image frequency domain analysis and the results show that with increasing clustering partition, the "bright spots" on the spectrogram gradually tend to be evenly distributed, the proposed algorithm halftone human visual effect is better than the least squares model-based algorithm gray scale values evenly distributed, the image smooth transition. Multiscale information fusion halftone method, mixed error measure function of the establishment of multiscale wavelet coefficients fusion to establish the boundary error measure function and improved K-means clustering method to establish the area error metric function. The applications direct binary search method to minimize the error of the initial image and halftone images, the least squares model-based algorithm in iteration15when convergence error is reduced to less than0.01, multiscale information fusion algorithms simply iterate8times, the algorithm has faster convergence rate. By analyzing the halftone noise spectrum Figure found the halftone noise spectrum is only distributed in the high frequency band, no strong periodic component and the local spectral components from low to high and gradually increase the human visual system is not sensitive.
The image of cells extracts the edge effect using canny operator when the best threshold equal to16. According to the image cells of this study engraving process to determine standards through trench values and dark tone value, when the value of the pass ditch and dark tone values in the standard range, the output1of the network points determination the qualified output deemed to have failed. Neural network to establish the quality of gravure network points sorting system, based on the model of the least squares method engraving network points separation results in a the unqualified class, the application of this article half toning method engraving network points of election results are in a qualified class, get the network points pass rates of98%or more.
Application of evaluation method based on the quality of the human visual system, the evaluation results can describe the hold of the edge detail, but also the reaction halftone texture suppression level. During the processing of the image halftoning, the results show that the multiscale information fusion method and weighted least squares gradient amplitude and texture entropy, angular second moment were higher than those based on model-based half toning method were lower than model the least squares method.
引文
[1]易丽芳,张逸新.激光雕刻凹版制版技术,包装工程,2006,27(2)：82-84
[2]G.Hennig, K.H.Selbm, A.B.Rock. Laser Engraving in Gravure Industry, Proc. of SPIE,2001:615-621
[3]金杨.从德鲁巴看激光雕刻凹版及柔版制版技术的发展,印刷技术,2000,(10)：78-81
[4]C.A.Bouman. Digital Halftoning, Digital Image Processing,2010(11):253-261
[5]R.A.Ulichney. Digital Halftoning and the Physical ReconstructionFunction[Ph.D] , Massachusetts Institute of Technology,1986
[6]B.Oztan, G.Sharma. Continuous Phase-Modulated Halftones, IEEE Transactions on Image Processing,2009,18(12):2718-2734
[7]P.W.Wong. Adaptive Error Diffusion and Its Application in Multiresolution Rendering. IEEE Transactions on Image Processing,1996,5(7):1184-1196
[8]H.Tronteli, J.Farrell, J.Wiseman etal. Optimal Halftoning Algorithm Depends on Printing Resolution, In Proc.SID Int.Symp.,Dig.Tech.Papers,1992:749-752
[9]T.D.Kite, B.L.Evans, A.C.Bovik. Modeling and Quality Assessment of Halftoning by Error Diffusion. IEEE Transactions on Image Processing,2000,9(5):909-922
[10]D.L.Lau, G.R.Arce.N.C.Gallagher. Digital Color Halftoning with Generalized Error Diffusion and Multichannel Green-Noise Masks. IEEE Transactions on Image Processing,2000,9(5):923-935
[11]N.D.Venkata, B.L.Evans. Adaptive Threshold Modulation for Error Diffusion Halftoning. IEEE Transactions on Image Processing,2001,10(1):104-116
[12]P.Li, J.P.AIlebach. Tone-dependent Error Diffusion. IEEE Transactions on Image Processing,2004,13(2):201-215
[13]Z.He. Hierarchical Error Diffusion. IEEE Transactions on Image Processing, 2009,18(7):1524-1535
[14]Z.He. Memory Efficient Hierarchical Error Diffusion, Color Imaging XV: Displaying, Hardcopy and Applications,2010,7(5):1-9
[15]王志国,刘纯亮,梁志虎.改善等离子体显示器动态假轮廓的显示灰度级调整方法.西安交通大学学报,2009,43(10)：89-93
[16]Yan Zhou, Chun Chen, Qiang Wang etal. Jumping Scanning Path Error Diffusion: A Novel Halftoning Algorithm Improving Mid-tone Quality.2011 Workshop on Digital Media and Digital Content Management,2011:215-222
[17]H.Nakai,K.Nakano. Cluster-Dot Halftoning Based on the Error Diffusion with no Directional Characteristic. IMECS 2010:1436-1441
[18]J.M.Guo, Y.F.Liu. Improved Dot Diffusion by Diffused Matrix and Class Matrix Co-Optimization. IEEE Transactions on Image Processing,2009,18(8)1804-1816
[19]I.Katsavounidis, C.J.Kuo. A Multiscale Error Diffusion Technique for Digital Halftoning. IEEE Transactions on Image Processing,1997,6(3)483-490
[20]Y.H.Fung, Y.H.Chan. Embedding Halftones of Different Resolutions in a Full-Scale Halftone. IEEE Signal Processing Letters,2006,13(3):153-156
[21]Y.H.Fung, K.C.Lui,Y.H.Chan. Low-Complexity High-Performance Multiscale Error Diffusion Technique for Digital Halftoning. Journal of Electronic Imaging,2007,16(1):35-44
[22]Y.H.Fung, Y.H.Chan. Green Noise Digital Halftoning With Multiscale Error Diffusion. IEEE Transactions on Image Processing,2010,19(7):1808-1823
[23]A.Chatterjee, K.C. Paul, Grayscale Digital Halftoning using Binary PSO.2011 International Conference on Recent Trends in Information Systems. IEEE Computer Society,2011:218-222
[24]P.W.Wong. Entropy-Constrained Halftoning Using Multipath Tree Coding. IEEE Transactions on Image Processing,1997,6(11):1567-1579
[25]K.Chandu, M.Stanich, B.Tragery etal. A GPU Implementation of Color Digital Halftoning using the Direct Binary Search Algorithm, In Proc IEEE, 2012:185-189
[26]M.Mese, P.P.Vaidyanathan. Optimized Halftoning Using Dot Diffusion and Methods for Inverse Halftoning. IEEE Transactions on Image Processing,2000, 9(4):691-709
[27]D.J.Lieberman, J.P.Allebach. A Dual Interpretation for Direct Binary Search and Its Implications for Tone Reproduction and Texture Quality. IEEE Transactions on Image Processing,2000,9(11):1950-1963
[28]Y.Zhou, C.Chen. Jumping Scanning Path Error Diffusion:A Novel Halftoning Algorithm Improving Mid-tone Quality. IEEE 2011 Workshop on Digital Media and Digital Content Management,2011:215-222
[29]G.Sarailidis, I.Katsavounidis. A Multiscale Error Diffusion Technique for Digital Multitoning. IEEE Transactions on Image Processing,2012, 21(5):2693-2705
[30]D.Kacker, T.Camis, J.P.Allebach. Electrophotographic Process Embedded in Direct Binary Search. IEEE Transactions on Image Processing,2002,11(3):243-257
[31]S.H.Kim, J.P.Allebach. Impact of HVS Models on Model-Based Halftoning.IEEE Transactions on Image Processing,2002,11(3):258-269
[32]M.Mese, P.P.Vaidyanathan. Tree-Structured Method for LUT Inverse Halftoning and for Image Halftoning. IEEE Transactions on Image Processing,2002,11(6): 644-655
[33]F.A.Baqai, J.P.Allebach. Halftoning via Direct Binary Search Using Analytical and Stochastic Printer Models. IEEE Transactions on Image Processing,2003, 12(1):1-15
[34]A.U.Agar, J.P.Allebach. Model-Based Color Halftoning Using Direct Binary Search. IEEE Transactions on Image Processing,2005,14(12):1945-1959
[35]G.Y.Lin, J.P.Allebach. Generating Stochastic Dispersed and Periodic Clustered Textures Using a Composite Hybrid Screed. IEEE Transactions on Image Processing,2006,15(12):3746-3758
[36]B. Trager, C.W.Wu, M.Stanich etal. GPU-enabled Parallel Processing for Image Halftoning Applications. IEEE Transactions on Image Processing,2011,13(4): 1528-1531
[37]T.Asano, P.Brass,S.Sasahara. Disc Covering Problem with Application to Digital Halftoning. Theory of Computing Systems,2010,46(2):157-173
[38]C.Lee,J.P.Allebach. The Hybrid Screen-Improving the Breed. IEEE Transactions on Image Processing,2010,19(2):435-450
[39]T.N.Pappas, J.P.Allebach, D.L.Neuhoff. Model-Based Digital Halftoning. IEEE Transactions Signal Processing Magazine,2003(7):14-27
[40]M.P. Eckert, A.P. Bradley. Perceptual Quality Metrics Applied to Still Image Compression. Signal Processing,1998,70:177-200
[41]T.N. Pappas, R.J. Safranek. Perceptual Criteria for Image Quality Evaluation. In Handbook of Image and Video Processing, New York:Academic,2001:669-684
[42]J. B. Mulligan, A. J. Ahumada. Principled Halftoning Based on Models of Human Vision. Proc. SPIE, Human Vision, Visual Proc, and Digital Display Ⅲ, 1992,16:109-121
[43]C.F.Hall, E.L.Hall. A Nonlinear Model for the Spatial Characteristics of the Human Visual System. IEEE Trans. Syst., Man, Cybern,1977, SMC-7:162-170
[44]F. W. Campbell, J. J. Kulikowski, J. Levinson. The Effect of Orientation on the Visual Resolution of Gratings. J. Physiol.,1966,187:427-436
[45]S. Daly. Subroutine for the Generation of A Two Dimensional Human Visual Contrast Sensitivity Function. Tech. Rep.233203Y, Eastman Kodak, Rochester, NY,1987.
[46]B.A. Wandell. Foundations of Vision. Sunderland, MA:Sinauer,1995
[47]A.N. Netravali, B.G. Haskell, Eds. Digital Pictures:Representation and Compression, New York:Plenum,1988
[48]A.B.Watson. The Cortex Transform:Rapid Computation of Simulated Neural Images. Compute.Vis., Graph., Image Process.,1987,39:311-327
[49]S. Daly. The Visible Differences Predictor:An Algorithm for the Assessment of Image Fidelity. In Digital Images and Human Vision, A.B.Watson, Ed.Cambridge, MA:MIT Press,1993:179-206
[50]D.J. Heeger, P.C. Teo. A Model of Perceptual Image Fidelity. In Proc.ICIP-95, Washington, DC, Oct.1995,11:343-345
[51]J. Lubin. The Use of Psychophysical Data and Models in the Analysis of Display System Performance. In Digital Images and Human Vision, A.B. Watson, Ed. Cambridge, MA:MIT Press,1993
[52]P.G. Roetling, T.M. Holladay. Tone Reproduction and Screen Design for Pictorial Electrographic Printing. J. Appli. Phot. Eng.,1979,15(4):179-182
[53]T.N. Pappas, D.L. Neuhoff. Model-based Halftoning. In Proc. SPIE, Human Vision, Visual Proc., and Digital Display Ⅱ, San Jose, CA, Feb.1991,1453:244-255.
[54]T. N. Pappas, D. L. Neuhoff. Printer Models and Error Diffusion. IEEE Transactions. Image Processing,1995,4:66-80
[55]T.N. Pappas, C.-K. Dong, D.L. Neuhoff. Measurement of Printer Parameters for Model-based Halftoning. J. Electron.Imaging,1993,2:193-204
[56]S.G. Wang, K.T. Knox, N. George. Novel Centering Method for Overlapping Correction in Halftoning. In Proc. IS & T's 47th Annu.Conf.,Rochester,NY,1994: 482-486
[57]L.B. Schien. Electrophotography and Development Physics.2nd ed.Morgan Hill, CA:Laplacian,1996
[58]D. Kacker, T. Camis, J.P. Allebach. Electrophotographic Process Embedded in Direct Binary Search. IEEE Transactions Image Processing,2002,11:243-257
[59]J. Yen, M. Carlsson, M. Chang, J.M. Carcia etal. Constraint Solving for Inkjet Print Mask Design. J. Imaging Sci. Technology,2000,44:391-397
[60]D.L.Lau, G.R.Arce, N.C Gallagher. Green Noise Digital Halftoning. Proceedings of IEEE,1998,86(12):2424-2444
[61]夏琴香,王甲子等.激光凹印制版设备及其关键技术分析.包装工程,2008,29(7)：37-40
[62]邓普军,金杨.凹版印刷网点增大讨论.包装工程,2006,(12)：19-21
[63]R.Bronstein. Automatic Inspection of Pringting Plates or Cylinders. The United States, NO.6031932[P].2000
[64]M. Mese, V.Dyana. Recent Advances in Digital Halftoning and Inverse Halftoning Methods. IEEE Transactions on Circuits and Systems:Fundamental Theory and Applications,2002,49(6):790-805
[65]T.N.Pappas, D.L.Neuhoff. Least Squares Model-Based Halftoning. IEEE Transactions Image Processing,1999,8(8):1102-1116
[66]J.L.Mannos, D.J. Sakrison. The Effects of A Visual Fidelity Criterion on the Encoding of Images. IEEE Transactions Information. Theory,1974, IT-20:525-536
[67]M.P.Segundo, L.Silva, O.R.P.Bellon etal. Automatic Face Segmentation and Facial Landmark Detection in Range Images. IEEE Transactions on Systems, Man, and Cybernetics, Part B:Cybernetics,2010,40:1319-1330
[68](美)韩家炜(Han, J.)等著,范明等译.数据挖掘：概念与技术(第3版).北京：机械工业出版社,2012
[69]S.N.Sulaiman, N.A.Mat Isa. Adaptive Fuzzy K-means Clustering Algorithm for Image Segmentation. IEEE Transactions on Consumer Electronics,2010, 56:2661-2668
[70]S.N.Sulaiman, N.A.Mat Isa. Denoising-based Clustering Algorithms for Segmentation of Low Level Salt-and-Pepper Noise Corrupted Images. IEEE Transactions on Consumer Electronics,2010,56:2702-2710
[71]D.J.Lieberman, J.P. Allebach. Digital Halftoning Using the Direct Binary Search Algorithm. In Proc.1ST Int. Conf. High Technology, Chiba, Japan, Sept.1996: 114-124
[72]Z.Wang, A.C.Bovik. A Universal Image Quality Index. IEEE Signal Processing Letters,2001,9(3):81-84
[73]D.L Lau, G.R.Arce. Modern Digital Halftoning. Marcel Dekker, New York NY, USA 2001
[74]T.D.Kite, B.L.Evans, A.C.Bovik. Modeling and Quality Assessment of Halftoning by Error Diffusion. IEEE Transactions Image Processing,2009 (5):909-922
[75]Z. Wang.L. Lu. Video quality assessment based on structural distortion measurement. Signal Processing:Image Communication,2004,19(2):121-132
[76]F. U. Siddiqui,N.A. M. Isa.Enhanced Moving K-Means (EMKM) Algorithm for Image Segmentation. IEEE Transactions on Consumer Electronics,2011,57(2): 1322-1332
[77]F. W.Campbell,R.H.S.Carpenter.J.Z.Levinson. Visibility of Aperiodic Patterns Compared with that of Sinusoidal Gratings.Journal of Physical,1969,204:283-298
[78]Mannos.J.L,Sakrison.D.J. The Effects of a Visual Fidelity Criterion on the Encoding of Images. IEEE Transaction on Image Processing,1974,20(5):525-536
[79]S. Daly. Subroutine for the Generation of a two Dimensional Human Visual Contrast Sensitivity Function Eastman Kodak.Rochester, NY,Tech. Rep.233203Y 1987.
[80]R.Nasanen.Visibility of Halftone Dot Textures,IEEE Trans.Syst.Man. Cyb,1984, 14(6):920-924
[81]Sang Ho Kim,Jan P.Allebach. Impact of HVS Models on Model-based Halftoning [C]. Proceeding of SPIE,2001:422-437
[82]Alford.J.R,Mitsa.T. Qualitative Comparison of Visual Model in an Iterative Halftoning Proceedure[C]. International Conference on Image Processing,1995: 339-342
[83]Ursula U. Miiller. Weighted Least Squares Estimators in Possibly Miss-pecified Nonlinear Regression. Metrika,2007,66:39-59
[84]邸继征.小波分析原理.北京：科学出版社,2010
[85]蒋英春.小波分析基本原理.天津：天津大学出版社,2012
[86]Xu Yansun. Weaver J B. Healyd M. Wavelet Transform Domain Filters A Spatially Selective Noise Filtration Technique. IEEE Transactions on Image Processing,1994,3(2):747-758
[87]Jaffard S.Beyond Besov Spaces.Oscillation Spaces.Constructive Approximation, 2005,21(1):29-61
[88]Shin G S,Kang M G. Wavelet-based Denoising Considering Inter-scale and Intrascale Dependences. Optical Engineering,2005,44(6):1-9
[89]Aristidis Likas,Nikos Vlassis,Jakob etal. The Global K-means Clustering Algorithm. Pattern Recognition,2003,36:451-461
[90](美)桑卡等著,艾海舟等译.图像处理分析与机器视觉.北京：清华大学出版社,2011
[91]赵鹏.机器视觉理论及应用.北京：电子工业出版社,2011
[92]朱振伟.一阶边缘检测算法的研究.现代电子技术,2009,(24)89-90
[93]Nobuyuki Yoshizawa,Yutaka Shimada. Magnetic Field Sensing by a Piezoelectric Magnetostrictive Resonator. IEEE Trans on Magn,2005, 41(11):4359-4361
[94]Warren S. McCulloch,Walter Pitts. A Logical Calculus of the Ideas Immanent in Nervous activity,1943
[95]Donald Hebb.The Organization of Behavior. A Neuropsychological Theory,1949
[96]杨守义,罗伟雄.一种基于高阶统计量的图像质量客观评价方法.北京理工大学学报,2001,21(5)：610-613
[2]G.Hennig, K.H.Selbm, A.B.Rock. Laser Engraving in Gravure Industry, Proc. of SPIE,2001:615-621
[3]金杨.从德鲁巴看激光雕刻凹版及柔版制版技术的发展,印刷技术,2000,(10)：78-81
[4]C.A.Bouman. Digital Halftoning, Digital Image Processing,2010(11):253-261
[5]R.A.Ulichney. Digital Halftoning and the Physical ReconstructionFunction[Ph.D] , Massachusetts Institute of Technology,1986
[6]B.Oztan, G.Sharma. Continuous Phase-Modulated Halftones, IEEE Transactions on Image Processing,2009,18(12):2718-2734
[7]P.W.Wong. Adaptive Error Diffusion and Its Application in Multiresolution Rendering. IEEE Transactions on Image Processing,1996,5(7):1184-1196
[8]H.Tronteli, J.Farrell, J.Wiseman etal. Optimal Halftoning Algorithm Depends on Printing Resolution, In Proc.SID Int.Symp.,Dig.Tech.Papers,1992:749-752
[9]T.D.Kite, B.L.Evans, A.C.Bovik. Modeling and Quality Assessment of Halftoning by Error Diffusion. IEEE Transactions on Image Processing,2000,9(5):909-922
[10]D.L.Lau, G.R.Arce.N.C.Gallagher. Digital Color Halftoning with Generalized Error Diffusion and Multichannel Green-Noise Masks. IEEE Transactions on Image Processing,2000,9(5):923-935
[11]N.D.Venkata, B.L.Evans. Adaptive Threshold Modulation for Error Diffusion Halftoning. IEEE Transactions on Image Processing,2001,10(1):104-116
[12]P.Li, J.P.AIlebach. Tone-dependent Error Diffusion. IEEE Transactions on Image Processing,2004,13(2):201-215
[13]Z.He. Hierarchical Error Diffusion. IEEE Transactions on Image Processing, 2009,18(7):1524-1535
[14]Z.He. Memory Efficient Hierarchical Error Diffusion, Color Imaging XV: Displaying, Hardcopy and Applications,2010,7(5):1-9
[15]王志国,刘纯亮,梁志虎.改善等离子体显示器动态假轮廓的显示灰度级调整方法.西安交通大学学报,2009,43(10)：89-93
[16]Yan Zhou, Chun Chen, Qiang Wang etal. Jumping Scanning Path Error Diffusion: A Novel Halftoning Algorithm Improving Mid-tone Quality.2011 Workshop on Digital Media and Digital Content Management,2011:215-222
[17]H.Nakai,K.Nakano. Cluster-Dot Halftoning Based on the Error Diffusion with no Directional Characteristic. IMECS 2010:1436-1441
[18]J.M.Guo, Y.F.Liu. Improved Dot Diffusion by Diffused Matrix and Class Matrix Co-Optimization. IEEE Transactions on Image Processing,2009,18(8)1804-1816
[19]I.Katsavounidis, C.J.Kuo. A Multiscale Error Diffusion Technique for Digital Halftoning. IEEE Transactions on Image Processing,1997,6(3)483-490
[20]Y.H.Fung, Y.H.Chan. Embedding Halftones of Different Resolutions in a Full-Scale Halftone. IEEE Signal Processing Letters,2006,13(3):153-156
[21]Y.H.Fung, K.C.Lui,Y.H.Chan. Low-Complexity High-Performance Multiscale Error Diffusion Technique for Digital Halftoning. Journal of Electronic Imaging,2007,16(1):35-44
[22]Y.H.Fung, Y.H.Chan. Green Noise Digital Halftoning With Multiscale Error Diffusion. IEEE Transactions on Image Processing,2010,19(7):1808-1823
[23]A.Chatterjee, K.C. Paul, Grayscale Digital Halftoning using Binary PSO.2011 International Conference on Recent Trends in Information Systems. IEEE Computer Society,2011:218-222
[24]P.W.Wong. Entropy-Constrained Halftoning Using Multipath Tree Coding. IEEE Transactions on Image Processing,1997,6(11):1567-1579
[25]K.Chandu, M.Stanich, B.Tragery etal. A GPU Implementation of Color Digital Halftoning using the Direct Binary Search Algorithm, In Proc IEEE, 2012:185-189
[26]M.Mese, P.P.Vaidyanathan. Optimized Halftoning Using Dot Diffusion and Methods for Inverse Halftoning. IEEE Transactions on Image Processing,2000, 9(4):691-709
[27]D.J.Lieberman, J.P.Allebach. A Dual Interpretation for Direct Binary Search and Its Implications for Tone Reproduction and Texture Quality. IEEE Transactions on Image Processing,2000,9(11):1950-1963
[28]Y.Zhou, C.Chen. Jumping Scanning Path Error Diffusion:A Novel Halftoning Algorithm Improving Mid-tone Quality. IEEE 2011 Workshop on Digital Media and Digital Content Management,2011:215-222
[29]G.Sarailidis, I.Katsavounidis. A Multiscale Error Diffusion Technique for Digital Multitoning. IEEE Transactions on Image Processing,2012, 21(5):2693-2705
[30]D.Kacker, T.Camis, J.P.Allebach. Electrophotographic Process Embedded in Direct Binary Search. IEEE Transactions on Image Processing,2002,11(3):243-257
[31]S.H.Kim, J.P.Allebach. Impact of HVS Models on Model-Based Halftoning.IEEE Transactions on Image Processing,2002,11(3):258-269
[32]M.Mese, P.P.Vaidyanathan. Tree-Structured Method for LUT Inverse Halftoning and for Image Halftoning. IEEE Transactions on Image Processing,2002,11(6): 644-655
[33]F.A.Baqai, J.P.Allebach. Halftoning via Direct Binary Search Using Analytical and Stochastic Printer Models. IEEE Transactions on Image Processing,2003, 12(1):1-15
[34]A.U.Agar, J.P.Allebach. Model-Based Color Halftoning Using Direct Binary Search. IEEE Transactions on Image Processing,2005,14(12):1945-1959
[35]G.Y.Lin, J.P.Allebach. Generating Stochastic Dispersed and Periodic Clustered Textures Using a Composite Hybrid Screed. IEEE Transactions on Image Processing,2006,15(12):3746-3758
[36]B. Trager, C.W.Wu, M.Stanich etal. GPU-enabled Parallel Processing for Image Halftoning Applications. IEEE Transactions on Image Processing,2011,13(4): 1528-1531
[37]T.Asano, P.Brass,S.Sasahara. Disc Covering Problem with Application to Digital Halftoning. Theory of Computing Systems,2010,46(2):157-173
[38]C.Lee,J.P.Allebach. The Hybrid Screen-Improving the Breed. IEEE Transactions on Image Processing,2010,19(2):435-450
[39]T.N.Pappas, J.P.Allebach, D.L.Neuhoff. Model-Based Digital Halftoning. IEEE Transactions Signal Processing Magazine,2003(7):14-27
[40]M.P. Eckert, A.P. Bradley. Perceptual Quality Metrics Applied to Still Image Compression. Signal Processing,1998,70:177-200
[41]T.N. Pappas, R.J. Safranek. Perceptual Criteria for Image Quality Evaluation. In Handbook of Image and Video Processing, New York:Academic,2001:669-684
[42]J. B. Mulligan, A. J. Ahumada. Principled Halftoning Based on Models of Human Vision. Proc. SPIE, Human Vision, Visual Proc, and Digital Display Ⅲ, 1992,16:109-121
[43]C.F.Hall, E.L.Hall. A Nonlinear Model for the Spatial Characteristics of the Human Visual System. IEEE Trans. Syst., Man, Cybern,1977, SMC-7:162-170
[44]F. W. Campbell, J. J. Kulikowski, J. Levinson. The Effect of Orientation on the Visual Resolution of Gratings. J. Physiol.,1966,187:427-436
[45]S. Daly. Subroutine for the Generation of A Two Dimensional Human Visual Contrast Sensitivity Function. Tech. Rep.233203Y, Eastman Kodak, Rochester, NY,1987.
[46]B.A. Wandell. Foundations of Vision. Sunderland, MA:Sinauer,1995
[47]A.N. Netravali, B.G. Haskell, Eds. Digital Pictures:Representation and Compression, New York:Plenum,1988
[48]A.B.Watson. The Cortex Transform:Rapid Computation of Simulated Neural Images. Compute.Vis., Graph., Image Process.,1987,39:311-327
[49]S. Daly. The Visible Differences Predictor:An Algorithm for the Assessment of Image Fidelity. In Digital Images and Human Vision, A.B.Watson, Ed.Cambridge, MA:MIT Press,1993:179-206
[50]D.J. Heeger, P.C. Teo. A Model of Perceptual Image Fidelity. In Proc.ICIP-95, Washington, DC, Oct.1995,11:343-345
[51]J. Lubin. The Use of Psychophysical Data and Models in the Analysis of Display System Performance. In Digital Images and Human Vision, A.B. Watson, Ed. Cambridge, MA:MIT Press,1993
[52]P.G. Roetling, T.M. Holladay. Tone Reproduction and Screen Design for Pictorial Electrographic Printing. J. Appli. Phot. Eng.,1979,15(4):179-182
[53]T.N. Pappas, D.L. Neuhoff. Model-based Halftoning. In Proc. SPIE, Human Vision, Visual Proc., and Digital Display Ⅱ, San Jose, CA, Feb.1991,1453:244-255.
[54]T. N. Pappas, D. L. Neuhoff. Printer Models and Error Diffusion. IEEE Transactions. Image Processing,1995,4:66-80
[55]T.N. Pappas, C.-K. Dong, D.L. Neuhoff. Measurement of Printer Parameters for Model-based Halftoning. J. Electron.Imaging,1993,2:193-204
[56]S.G. Wang, K.T. Knox, N. George. Novel Centering Method for Overlapping Correction in Halftoning. In Proc. IS & T's 47th Annu.Conf.,Rochester,NY,1994: 482-486
[57]L.B. Schien. Electrophotography and Development Physics.2nd ed.Morgan Hill, CA:Laplacian,1996
[58]D. Kacker, T. Camis, J.P. Allebach. Electrophotographic Process Embedded in Direct Binary Search. IEEE Transactions Image Processing,2002,11:243-257
[59]J. Yen, M. Carlsson, M. Chang, J.M. Carcia etal. Constraint Solving for Inkjet Print Mask Design. J. Imaging Sci. Technology,2000,44:391-397
[60]D.L.Lau, G.R.Arce, N.C Gallagher. Green Noise Digital Halftoning. Proceedings of IEEE,1998,86(12):2424-2444
[61]夏琴香,王甲子等.激光凹印制版设备及其关键技术分析.包装工程,2008,29(7)：37-40
[62]邓普军,金杨.凹版印刷网点增大讨论.包装工程,2006,(12)：19-21
[63]R.Bronstein. Automatic Inspection of Pringting Plates or Cylinders. The United States, NO.6031932[P].2000
[64]M. Mese, V.Dyana. Recent Advances in Digital Halftoning and Inverse Halftoning Methods. IEEE Transactions on Circuits and Systems:Fundamental Theory and Applications,2002,49(6):790-805
[65]T.N.Pappas, D.L.Neuhoff. Least Squares Model-Based Halftoning. IEEE Transactions Image Processing,1999,8(8):1102-1116
[66]J.L.Mannos, D.J. Sakrison. The Effects of A Visual Fidelity Criterion on the Encoding of Images. IEEE Transactions Information. Theory,1974, IT-20:525-536
[67]M.P.Segundo, L.Silva, O.R.P.Bellon etal. Automatic Face Segmentation and Facial Landmark Detection in Range Images. IEEE Transactions on Systems, Man, and Cybernetics, Part B:Cybernetics,2010,40:1319-1330
[68](美)韩家炜(Han, J.)等著,范明等译.数据挖掘：概念与技术(第3版).北京：机械工业出版社,2012
[69]S.N.Sulaiman, N.A.Mat Isa. Adaptive Fuzzy K-means Clustering Algorithm for Image Segmentation. IEEE Transactions on Consumer Electronics,2010, 56:2661-2668
[70]S.N.Sulaiman, N.A.Mat Isa. Denoising-based Clustering Algorithms for Segmentation of Low Level Salt-and-Pepper Noise Corrupted Images. IEEE Transactions on Consumer Electronics,2010,56:2702-2710
[71]D.J.Lieberman, J.P. Allebach. Digital Halftoning Using the Direct Binary Search Algorithm. In Proc.1ST Int. Conf. High Technology, Chiba, Japan, Sept.1996: 114-124
[72]Z.Wang, A.C.Bovik. A Universal Image Quality Index. IEEE Signal Processing Letters,2001,9(3):81-84
[73]D.L Lau, G.R.Arce. Modern Digital Halftoning. Marcel Dekker, New York NY, USA 2001
[74]T.D.Kite, B.L.Evans, A.C.Bovik. Modeling and Quality Assessment of Halftoning by Error Diffusion. IEEE Transactions Image Processing,2009 (5):909-922
[75]Z. Wang.L. Lu. Video quality assessment based on structural distortion measurement. Signal Processing:Image Communication,2004,19(2):121-132
[76]F. U. Siddiqui,N.A. M. Isa.Enhanced Moving K-Means (EMKM) Algorithm for Image Segmentation. IEEE Transactions on Consumer Electronics,2011,57(2): 1322-1332
[77]F. W.Campbell,R.H.S.Carpenter.J.Z.Levinson. Visibility of Aperiodic Patterns Compared with that of Sinusoidal Gratings.Journal of Physical,1969,204:283-298
[78]Mannos.J.L,Sakrison.D.J. The Effects of a Visual Fidelity Criterion on the Encoding of Images. IEEE Transaction on Image Processing,1974,20(5):525-536
[79]S. Daly. Subroutine for the Generation of a two Dimensional Human Visual Contrast Sensitivity Function Eastman Kodak.Rochester, NY,Tech. Rep.233203Y 1987.
[80]R.Nasanen.Visibility of Halftone Dot Textures,IEEE Trans.Syst.Man. Cyb,1984, 14(6):920-924
[81]Sang Ho Kim,Jan P.Allebach. Impact of HVS Models on Model-based Halftoning [C]. Proceeding of SPIE,2001:422-437
[82]Alford.J.R,Mitsa.T. Qualitative Comparison of Visual Model in an Iterative Halftoning Proceedure[C]. International Conference on Image Processing,1995: 339-342
[83]Ursula U. Miiller. Weighted Least Squares Estimators in Possibly Miss-pecified Nonlinear Regression. Metrika,2007,66:39-59
[84]邸继征.小波分析原理.北京：科学出版社,2010
[85]蒋英春.小波分析基本原理.天津：天津大学出版社,2012
[86]Xu Yansun. Weaver J B. Healyd M. Wavelet Transform Domain Filters A Spatially Selective Noise Filtration Technique. IEEE Transactions on Image Processing,1994,3(2):747-758
[87]Jaffard S.Beyond Besov Spaces.Oscillation Spaces.Constructive Approximation, 2005,21(1):29-61
[88]Shin G S,Kang M G. Wavelet-based Denoising Considering Inter-scale and Intrascale Dependences. Optical Engineering,2005,44(6):1-9
[89]Aristidis Likas,Nikos Vlassis,Jakob etal. The Global K-means Clustering Algorithm. Pattern Recognition,2003,36:451-461
[90](美)桑卡等著,艾海舟等译.图像处理分析与机器视觉.北京：清华大学出版社,2011
[91]赵鹏.机器视觉理论及应用.北京：电子工业出版社,2011
[92]朱振伟.一阶边缘检测算法的研究.现代电子技术,2009,(24)89-90
[93]Nobuyuki Yoshizawa,Yutaka Shimada. Magnetic Field Sensing by a Piezoelectric Magnetostrictive Resonator. IEEE Trans on Magn,2005, 41(11):4359-4361
[94]Warren S. McCulloch,Walter Pitts. A Logical Calculus of the Ideas Immanent in Nervous activity,1943
[95]Donald Hebb.The Organization of Behavior. A Neuropsychological Theory,1949
[96]杨守义,罗伟雄.一种基于高阶统计量的图像质量客观评价方法.北京理工大学学报,2001,21(5)：610-613