期刊文献+

多光源图像细化和细节增强的协同图像处理算法研究 被引量:7

Multi-light source image defined and enhanced image processing algorithms
原文传递
导出
摘要 多光源图像合成处理过程中,为了增强图像的轮廓和表面细节,同时使合成后的图像边缘无伪迹、保真度高,提出了一种多光源图像细化与细节增强的协同处理算法。该算法通过梯度域法构建辅助层,采用二次过滤法提取出细节层,提出了新的阴影检测算法用来去除细节层的伪迹,同时使该细节层包含了输入图像的全部信息;使用一幅输入图像构造一个基础层并进行暗区亮度增强处理;细节层和基础层进行复合处理后,重现了暗影区丢失的细节,同时增强了现有的细节。该算法与其他算法进行图像处理对比实验,结果表明,该算法用于实现多光源图像细化和细节增强是可行的、有效的,采用这种方法合成的图像看起来更自然,相对于输入图像,输出图像保持了较高的保真度。该算法具有交互性,用户可以手动或自动调整合成图像的效果。 In order to enhance the image contour and surface details, and make the image edge without artifact after synthe-sis, high fidelity, the authors propose a multi light source image processing algorithm with details enhancement. The algo-rithm, through the gradient domain method to construct assist, secondary filtering method, is used to extract details layer, and a new shadow detection algorithm is used to remove detail layer artifact, at the same time make the detail layer contain all the information in the input image; Using a picture of the input image constructs a base layer and dark space brightness enhancement processing; After detail layer and base layer composite processing, the shadow area missing details are recon- structed. We enhance the existing details at the same time. The results of algorithm and other algorithms for image process-ing experiments show that the algorithm is used to realize the source image and detail enhancement is feasible and effective; using the method of synthetic image looks more natural; relative to the input image, the output image remains high fidelity. The algorithm is interactive; the user can adjust the effect of the composite image manually or automatically.
作者 张德发 肜丽
出处 《重庆邮电大学学报(自然科学版)》 CSCD 北大核心 2014年第2期260-264,共5页 Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)
基金 工业高速缝制装备关键机构智能仿真及可缝性能平台研究(2012C21045) 浙江省科技厅公益项目(2012C20145)~~
关键词 协同图像处理 图像细化 细节增强 阴影检测 collaborative image processing detail refinement detail enhancement shadow detection
  • 相关文献

参考文献15

  • 1LERTRATFANAPANICH S, BOSE N K. High Resolution Image Formation From Low Resolution Frames Using Delaunay Triangulation [ J ]. IEEE Transactions on Image Processing,2002,11 ( 12 ) : 1427-1441.
  • 2DEBEVEC P E, MALIK J. Recovering high dynamic range radiance maps from photographs [ C]//ACM SIG- GRAPH 2008 classes. NewYork, USA : ACM, 2008 : 31.
  • 3AKERS D, LOSASSO F, KLINGNER J, et al. Conve- ying shape and features with image-based relighting [ C]//Proceedings of the 14th IEEE Visuahzation 2003 (VIS03). [ s. 1. ] : IEEE Computer Society, 2003 : 46.
  • 4RASKAR R, ILIE A, YU J. Image fusion for context en- hancement and video surrealism [ C ]//ACM SIGGRAPH 2005 Courses. NewYork, USA : ACM, 2005 : 4.
  • 5朱立新,王平安,夏德深.基于梯度场均衡化的图像对比度增强[J].计算机辅助设计与图形学学报,2007,19(12):1546-1552. 被引量:37
  • 6关雪梅.边缘检测算法在图像处理中的应用研究[J].信息与电脑(理论版),2009(10):93-94. 被引量:2
  • 7TOET A. Multiscale contrast enhancement with applica- tions to image fusion[ J]. Optical Engineering, 1992, 31 (5) : 1026-1031.
  • 8GERONIMO J S, HARDIN D P, MASSOPUST P R. Fractal functions and wavelet expansions based on several scaling functions [ J ]. Journal of Approximation Theory, 1994, 78(3) : 373-401.
  • 9FATrAL R, AGRAWALA M, RUSINKIEWICZ S. Multi- scale shape and detail enhancement from multi-light image collections[J]. ACM Trans Graph, 2007, 26(3) : 51.
  • 10张笑微,王月琴.基于灰度图像的阴影检测算法[J].兵工自动化,2007,26(7):45-47. 被引量:4

二级参考文献39

  • 1肖梅,韩崇昭,张雷.交通监控系统中基于多源信息融合的运动阴影检测[J].西安交通大学学报,2005,39(10):1077-1080. 被引量:9
  • 2朱立新,欧阳晓丽,夏德深.基于伪线性方向扩散方程的指纹图像增强[J].模式识别与人工智能,2006,19(6):806-811. 被引量:5
  • 3阮秋琦.数字图像处理学[M].北京:电子工业出版社,2000..
  • 4Julio Cezar Silveira Jacques Jr,Claudio Rosito Jung,Soraia Raupp Musse.Background Subtraction and Shadow Detection in Grayscale Video Sequences[A].IEEE Computer Society[C],2005.189-196.
  • 5A J H.Hii,C E Hann,J G.Chase,E E W Van Houten.Fast Normalized Cross Correlation for Motion Tracking Using Basis Functions[J].Computer Methods and Programs in Biomedicine,2006,82 (2):144-156.
  • 6Du-Ming Tsai,Chien-Ta Lin.Fast Normalized Cross Correlation for Defect Detection[J].Pattern Recognition Letters,2003,24,(15):2625-2631.
  • 7Alessandro Bevilacqua.Optimizing Parameters of a Motion Detection System by Means of a Distributed Genetic Algorithm[A].Image and Vision Computing,2005,23 (9):815-829.
  • 8Jain A K. Fundamentals of digital image processing [M]. Englewood Cliffs, N J: Prentice-Hall, 1989.
  • 9Sid-Ahmed M A. Image processing [ M ]. New York: McGraw- Hill, 1995.
  • 10Rahman Z, Jobson D J, Woodell G. Retinex processing for automatic image enhancement [J]. Journal of Electronic Imaging, 2004, 13(1): 78-94.

共引文献48

同被引文献53

  • 1杨初平,翁嘉文,李海,谭穗妍.单幅条纹图相位解调的小波分析方法[J].光子学报,2012,41(10):1211-1216. 被引量:4
  • 2刘勃,周荷琴,魏铭旭.基于颜色和运动信息的夜间车辆检测方法[J].中国图象图形学报(A辑),2005,10(2):187-191. 被引量:32
  • 3杨再华,李玉和,李庆祥,郭阳宽.基于边缘特征提取的图像清晰度评价函数[J].计算机工程与应用,2005,41(10):35-36. 被引量:20
  • 4文学志,赵宏,王楠,袁淮.基于知识和外观方法相结合的后方车辆检测[J].东北大学学报(自然科学版),2007,28(3):333-336. 被引量:5
  • 5康金满,艾萨克·科恩,杰拉德.在强烈视差下检测和跟踪一个运动目标[J].第十届IEEE计算机视觉国际会议论文集,2005:178-195.
  • 6埃尔加马勒,哈伍德,戴维斯.基于背景想减的无参数模型[J].第6届欧洲计算机视觉会议,都柏林爱尔兰,2000:246-252.
  • 7埃尔加马勒,Duraiswami,哈伍德.采用非参数核密度估计的背景和前景型中的视觉监控[J].IEEE会议,2002,90(7):1151-1163.
  • 8GutchessD,trajkonic M,科恩索拉尔E.一个背景模型初始化算法.第八届视频监控[J].计算机视觉,温哥华2001年IEEE国际会议计算机视觉论文集733-740.
  • 9刘辉.基于低空平台的车辆检测与跟踪研究[D].吉林:吉林大学,2011.
  • 10Krymski A,Tu N.A 9-V/lux·s 5000-frames/s 512×512 CMOS sensor[J].IEEE Trans.Electron.Devices,2003,50(1): 136-143.

引证文献7

二级引证文献14

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部