离焦图像清晰区域分割方法研究
|
摘要
本文主要讨论了离焦模糊图像清晰区域的分割及应用。本文首先从光学成像原理上分析了离焦模糊的形成原因,并建立了基于区域的离焦模糊模型。由于图像离焦模糊主要反映在图像边缘特征上,所以本文在经典“sobel”边缘提取方法的基础上,提出了基于边缘最大梯度的边缘细化方法。接着,本文先是使用边缘最大梯度和过渡区宽度这两个特征对图像的细节区域进行分类;再依据就近原则,使用竞争型神经网络对图像的平坦区域进行分类;然后根据图像的锐利分布情况和地域性原则,实现了图像的清晰区域分割和排序。最后,本文把清晰区域分割方法应用到图像复原领域,实现了基于区域的图像复原;又应用到多焦点图像融合领域,提出了一种新的融合规则,实现了多焦点图像间的融合,得到了非常理想的融合效果。
The clear region segmentation of out-of-focus degraded image and its applications are mainly discussed in this paper. At first, the cause of out-of-focus blur is introduced, and a regions-based out-of-focus degraded model is established in this paper. Because out-of-focus blur is closely related to edge characters, a novel maximum gradient based edge thinning method is presented after detecting image edges using sobel operator.
Then a maximum gradient and the width of transition region based detail area classification method and a principle of proximity-based flat area classification method by using competed artificial neural network are proposed in this paper. Next the algorithm of clear regions segmentation based on sharpness distribution and‘region principle’is given.
At last, the clear region segmentation’s applications in other image processing fields are discussed. A regions-based image restoration algorithm and a new multi-focus images fusion algorithm are presented in this paper. The experiment results are very ideal.
The clear region segmentation of out-of-focus degraded image and its applications are mainly discussed in this paper. At first, the cause of out-of-focus blur is introduced, and a regions-based out-of-focus degraded model is established in this paper. Because out-of-focus blur is closely related to edge characters, a novel maximum gradient based edge thinning method is presented after detecting image edges using sobel operator.
Then a maximum gradient and the width of transition region based detail area classification method and a principle of proximity-based flat area classification method by using competed artificial neural network are proposed in this paper. Next the algorithm of clear regions segmentation based on sharpness distribution and‘region principle’is given.
At last, the clear region segmentation’s applications in other image processing fields are discussed. A regions-based image restoration algorithm and a new multi-focus images fusion algorithm are presented in this paper. The experiment results are very ideal.
引文
[1] J Z Wang, J Li, and G Wiederhold. Simplicity: Semantics-Sensitive Integrated Matching for Picture Libraries. Pattern Analysis and Machine Intelligence, IEEE Transactions. 2001, 9. 23(9). 947– 963.
[2] Salembier P, Marques F. Region-based representations of image and video: segmentation tools for multimedia services, Circuits and Systems for Video Technology, IEEE Transactions 1999, 12. 9(8). 1147– 1169.
[3] Jianping Fan, Xingquan Zhu, Lide Wu. Automatic model-based semantic object extraction algorithm, Circuits and Systems for Video Technology, IEEE Transactions 2001,10. 11(10), 1073– 1084.
[4] M I Sezan, G Pavlovc, A Ma Takalp, and A T Erdem. On modeling the focus blur in image restoration. Proc. IEEE Int. Conf Acoust. Speech, and Signal Proc. 1991, 5. 2485-2488.
[5] Bi Xiao jun, Wang Ting. A daptive Blind Image Restoration Algorithm of Degraded Image. Image and Signal Processing, 2008. CISP '08. Congress 2008, 5. 3. 536– 540.
[6] Dong Dong Cao, Ping Guo. Blind image restoration based on wavelet analysis. Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference 2005, 8. 8. 4977– 4982.
[7]那彦,焦李成.基于多分辨分析理论的图像融合方法.西安:西安电子科技大学出版社. 2007, 5. 1-182.
[8] Nayar S K, Watanabe M, Noguchi M. Real-Time Focus Range Sensor. Pattern Analysis and Machine Intelligence, IEEE Transactions. 1996, 12. 18(12). 1186– 1198.
[9] D. Kundur and D. Hatzinakos, Blind Image Deconvolutions, IEEE Signal Processing Magazine, 1996, 5. 43-63.
[10] D KUNDUR and D HATZINAKOS. blind image deconvolution revisited. Signal Processing Magazine, IEEE 1996, 11. 13(6). 61– 63.
[11] E Savakis and H J Trussell, On the Accuracy of PSF Representation in Image Restoration, IEEE Trans. Image Processing. 1993, 4. 2. 252-259.
[12] Shiqian Wu, Weisi Lin, Lijun Jiang. An Objective Out-of-Focus Blur Measurement, Information. Communications and Signal Processing, 2005 Fifth International Conference. 2005, 12. 334– 338.
[13] Kildal P S, Stamnes J J. A symptotic transition region theory for edge diffraction I. Tracing transition regions via reflectors Antennas and Propagation, IEEE Transactions 1990, 9. 38(9). 1350– 1358.
[14] Wallmark O, Harnefors L. Sensorless Control of Salient PMSM Drives in the Transition Region. Industrial Electronics, IEEE Transactions 2006, 6 53(4). 1179– 1187.
[15] Burt I J. Transition Region and Coronal Explorer Mission”Aerospace Applications Conference, 1996. Proceedings. IEEE.1996, 12. 2. 197– 212.
[16]李捷,唐星科,蒋延军.几种边缘检测算法的比较研究.信息技术.2007.6. (9). 106—108.
[17] Elmabrouk A, Aggoun A. Edge detection using local histogram analysis. Electronics Letters.1998, 6. 34(12), 1216– 1217.
[18] K Etemad, R Chellappa. A Neural Network Based Edge Detector. Proc IJCNN. 1993. 132-137
[19] Wang Rong ben, Yu Tian hong, Jin Li-sheng. Edge extraction method study based on maximum entropy for linear lane identifying and tracking. Intelligent Vehicles Symposium, 2005. Proceedings. IEEE 6-8 2005, 6. 849– 854.
[20] Xin Wang. Laplacian Operator-Based Edge Detectors. Pattern Analysis and Machine Intelligence, IEEE Transactions 2007, 5 29(5). 886– 890.
[21] Fangwei Zhao, deSilva C J S. Use of the Laplacian of Gaussian operator in prostate ultrasound image processing. Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE 1998, 11. 2. 812– 815.
[22] Sifeng Wang, Jingxiu Zhao. A self-adaptive edge detection method based on LoG algorithm. Audio, Language and Image Processing, 2008. ICALIP 2008. International Conference 2008, 7. 1009– 1011.
[23] Zhang Shuqing, Liu Songyan, Jin Mei. An improved Canny Arithmetic based on image pyramid theories. Automation and Logistics, 2008. ICAL 2008. IEEE International Conference 2008, 12. 2347– 2351.
[24] Xun Wang, Jian Qiu Jin. An Edge Detection Algorithm Based on Improved CANNY Operator. Intelligent Systems Design and Applications, 2007. ISDA 2007. Seventh International Conference 2007, 10. 623– 628.
[25]余洪山,王耀南.一种改进型Canny边缘检测算法. 2004, 4.计算机工程与应用. 27-29.
[26] Yishu Zhai, Xiao ming Liu. Adaptive Edge Detection Based on Multiscale WaveletFeatures. Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress 2006, 6 2. 10289– 10293.
[27] Beltran J R, Garcia Lucia J, Navarro J. Edge detection and classification using Mallat's wavelet. Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference. 1994, 11. 1. 293– 297.
[28] Dawei Qi, Yuanxiang Li, Lei Yu. The application of mathematical morphological optimization algorithm in edge detection of defected wood image. Automation and Logistics, 2008. ICAL 2008. IEEE International Conference on 2008, 9. 2271– 2276.
[29] Ongwattanakul S, Chewputtanagul P. Jackson D J. Scalable giga-pixels/s binary image morphological operations. Circuits and Systems, 2003. ISCAS '03. Proceedings of the 2003 International Symposium 2003, 5. 2. II-444 - II-447.
[30] El Khamy S E, Lotfy M E, Yamany N. A modified fuzzy Sobel edge detector. Radio Science Conference, 2000. 17th NRSC '2000. Seventeenth National. 2000, 12. C32/1 - C32/9.
[31] Kazakova N, Margala M. Durdle N G. Sobel edge detection processor for a real-time volume rendering system. Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium. 2004, 5. 2. II - 913-16.
[32] Cios K J, Sarieh A. An edge extraction technique for noisy images. Biomedical Engineering, IEEE Transactions. 1990, 5. 37(5). 520– 524.
[33]张建德,邵定宏.基于sobel细化算法的螺纹边缘检测.机械电子. 2007.5. 77-78
[34]林世毅,苏广州,陈东.基于二步法的边缘细化算法研究.仪器仪表学报2004, 8. 25(4). 682-684.
[35]楼顺天.基于MATLAB的系统分析与设计—神经网络. 3版.西安:西安电科技大学出版社, 2000.1-155.
[36]王学武,谭得健.神经网络得应用与发展趋势.计算机工程与应用. 2003, 3. 98-113.
[37] Weiyang Zhou. Verification of the nonparametric characteristics of back propagation neural networks for image classification. Geoscience and Remote Sensing, IEEE Transactions . 1999, 3. 37(2). 771– 779.
[38] Chuan Yu Chang, Shih Yu Fu. Image Classification using a Module RBF Neural Network. Innovative Computing, Information and Control, 2006. ICICIC '06. First International Conference. 2006, 8. 2. 270– 273.
[39] Machado R J, Neves P E C S A. Multi-model neural network for imageclassification. Cybernetic Vision, 1996. Proceedings., Second Workshop . 1996, 12. 57– 59.
[40] Ong E, Lin W, Lu Z. A no-reference quality metric for measuring image blur. Signal Processing and Its Applications, 2003. Proceedings. Seventh International Symposium. 2003, 7. 1. 469– 472.
[41] Fuzheng Yang, Shuai Wan, Yilin Chang. A novel objective no-reference metric for digital video quality assessment. Signal Processing Letters, IEEE. 2005, 10. 12(10). 685– 688.
[42] Hanghang Tong, Mingjing Li, Hong-Jiang Zhang. Learning No-Reference Quality Metric by Examples. Multimedia Modelling Conference, 2005. MMM 2005. Proceedings of the 11th International 2005, 1. 247– 254.
[43] Rafael C Gonzalez, Richard E Woods. Digital Image Processing Second Edition.北京:电子工业出版社.2002.7.141-193
[44] Shiqian Wu, Weisi Lin, Zhongkang Lu. Blind Blur Assessment for Vision-Based Applications. Multimedia and Expo, 2007 IEEE International Conference. 2007, 7. 1639– 1642.
[45] Xin Yu, Cairong Zou, Luxi Yang. Improved recursive inverse filtering method for blind image restoration. Signal Processing, 2002 6th International Conference. 2002, 8. 1. 37 - 40 .
[46] Berger T, Stromberg J O, Eltoft T. Adaptive regularized constrained least squares image restoration. Image Processing, IEEE Transactions. 1999, 9. 8(9). 1191 - 1203
[47] Graham L, Adhami R. Fusion of coincident images using spatial frequency discrimination by application of the wavelet transform. Time-Frequency and Time-Scale Analysis 1994. Proceedings of the IEEE-SP International Symposium 1994, 10. 326– 329.
[48]李建林,俞建成,孙胜利.基于梯度金字塔图像融合的研究.科学技术与工程. 2007,11. 7(22). 5818-5821.
[49]樊秀菊.脉冲耦合神经网络及基于此的图像分割与图像压缩. 2003 -西安电子科技大学:计算机应用技术.
[50] Yinghua Lu, Xue Feng, Jingbo Zhang. A Multi-focus Image Fusion Based on Wavelet and Region Detection. EUROCON, 2007. The International Conference on "Computer as a Tool" . 2007, 9. 294– 298.
[51] Yu Song, Mantian Li, Qingling Li. New Wavelet Based Multi-focus Image Fusion Scheme and Its Application on Optical Microscopy. Robotics and Biomimetics,2006.12.BIO '06. IEEE International Conference 2006, 12. 401– 405.
[2] Salembier P, Marques F. Region-based representations of image and video: segmentation tools for multimedia services, Circuits and Systems for Video Technology, IEEE Transactions 1999, 12. 9(8). 1147– 1169.
[3] Jianping Fan, Xingquan Zhu, Lide Wu. Automatic model-based semantic object extraction algorithm, Circuits and Systems for Video Technology, IEEE Transactions 2001,10. 11(10), 1073– 1084.
[4] M I Sezan, G Pavlovc, A Ma Takalp, and A T Erdem. On modeling the focus blur in image restoration. Proc. IEEE Int. Conf Acoust. Speech, and Signal Proc. 1991, 5. 2485-2488.
[5] Bi Xiao jun, Wang Ting. A daptive Blind Image Restoration Algorithm of Degraded Image. Image and Signal Processing, 2008. CISP '08. Congress 2008, 5. 3. 536– 540.
[6] Dong Dong Cao, Ping Guo. Blind image restoration based on wavelet analysis. Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference 2005, 8. 8. 4977– 4982.
[7]那彦,焦李成.基于多分辨分析理论的图像融合方法.西安:西安电子科技大学出版社. 2007, 5. 1-182.
[8] Nayar S K, Watanabe M, Noguchi M. Real-Time Focus Range Sensor. Pattern Analysis and Machine Intelligence, IEEE Transactions. 1996, 12. 18(12). 1186– 1198.
[9] D. Kundur and D. Hatzinakos, Blind Image Deconvolutions, IEEE Signal Processing Magazine, 1996, 5. 43-63.
[10] D KUNDUR and D HATZINAKOS. blind image deconvolution revisited. Signal Processing Magazine, IEEE 1996, 11. 13(6). 61– 63.
[11] E Savakis and H J Trussell, On the Accuracy of PSF Representation in Image Restoration, IEEE Trans. Image Processing. 1993, 4. 2. 252-259.
[12] Shiqian Wu, Weisi Lin, Lijun Jiang. An Objective Out-of-Focus Blur Measurement, Information. Communications and Signal Processing, 2005 Fifth International Conference. 2005, 12. 334– 338.
[13] Kildal P S, Stamnes J J. A symptotic transition region theory for edge diffraction I. Tracing transition regions via reflectors Antennas and Propagation, IEEE Transactions 1990, 9. 38(9). 1350– 1358.
[14] Wallmark O, Harnefors L. Sensorless Control of Salient PMSM Drives in the Transition Region. Industrial Electronics, IEEE Transactions 2006, 6 53(4). 1179– 1187.
[15] Burt I J. Transition Region and Coronal Explorer Mission”Aerospace Applications Conference, 1996. Proceedings. IEEE.1996, 12. 2. 197– 212.
[16]李捷,唐星科,蒋延军.几种边缘检测算法的比较研究.信息技术.2007.6. (9). 106—108.
[17] Elmabrouk A, Aggoun A. Edge detection using local histogram analysis. Electronics Letters.1998, 6. 34(12), 1216– 1217.
[18] K Etemad, R Chellappa. A Neural Network Based Edge Detector. Proc IJCNN. 1993. 132-137
[19] Wang Rong ben, Yu Tian hong, Jin Li-sheng. Edge extraction method study based on maximum entropy for linear lane identifying and tracking. Intelligent Vehicles Symposium, 2005. Proceedings. IEEE 6-8 2005, 6. 849– 854.
[20] Xin Wang. Laplacian Operator-Based Edge Detectors. Pattern Analysis and Machine Intelligence, IEEE Transactions 2007, 5 29(5). 886– 890.
[21] Fangwei Zhao, deSilva C J S. Use of the Laplacian of Gaussian operator in prostate ultrasound image processing. Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE 1998, 11. 2. 812– 815.
[22] Sifeng Wang, Jingxiu Zhao. A self-adaptive edge detection method based on LoG algorithm. Audio, Language and Image Processing, 2008. ICALIP 2008. International Conference 2008, 7. 1009– 1011.
[23] Zhang Shuqing, Liu Songyan, Jin Mei. An improved Canny Arithmetic based on image pyramid theories. Automation and Logistics, 2008. ICAL 2008. IEEE International Conference 2008, 12. 2347– 2351.
[24] Xun Wang, Jian Qiu Jin. An Edge Detection Algorithm Based on Improved CANNY Operator. Intelligent Systems Design and Applications, 2007. ISDA 2007. Seventh International Conference 2007, 10. 623– 628.
[25]余洪山,王耀南.一种改进型Canny边缘检测算法. 2004, 4.计算机工程与应用. 27-29.
[26] Yishu Zhai, Xiao ming Liu. Adaptive Edge Detection Based on Multiscale WaveletFeatures. Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress 2006, 6 2. 10289– 10293.
[27] Beltran J R, Garcia Lucia J, Navarro J. Edge detection and classification using Mallat's wavelet. Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference. 1994, 11. 1. 293– 297.
[28] Dawei Qi, Yuanxiang Li, Lei Yu. The application of mathematical morphological optimization algorithm in edge detection of defected wood image. Automation and Logistics, 2008. ICAL 2008. IEEE International Conference on 2008, 9. 2271– 2276.
[29] Ongwattanakul S, Chewputtanagul P. Jackson D J. Scalable giga-pixels/s binary image morphological operations. Circuits and Systems, 2003. ISCAS '03. Proceedings of the 2003 International Symposium 2003, 5. 2. II-444 - II-447.
[30] El Khamy S E, Lotfy M E, Yamany N. A modified fuzzy Sobel edge detector. Radio Science Conference, 2000. 17th NRSC '2000. Seventeenth National. 2000, 12. C32/1 - C32/9.
[31] Kazakova N, Margala M. Durdle N G. Sobel edge detection processor for a real-time volume rendering system. Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium. 2004, 5. 2. II - 913-16.
[32] Cios K J, Sarieh A. An edge extraction technique for noisy images. Biomedical Engineering, IEEE Transactions. 1990, 5. 37(5). 520– 524.
[33]张建德,邵定宏.基于sobel细化算法的螺纹边缘检测.机械电子. 2007.5. 77-78
[34]林世毅,苏广州,陈东.基于二步法的边缘细化算法研究.仪器仪表学报2004, 8. 25(4). 682-684.
[35]楼顺天.基于MATLAB的系统分析与设计—神经网络. 3版.西安:西安电科技大学出版社, 2000.1-155.
[36]王学武,谭得健.神经网络得应用与发展趋势.计算机工程与应用. 2003, 3. 98-113.
[37] Weiyang Zhou. Verification of the nonparametric characteristics of back propagation neural networks for image classification. Geoscience and Remote Sensing, IEEE Transactions . 1999, 3. 37(2). 771– 779.
[38] Chuan Yu Chang, Shih Yu Fu. Image Classification using a Module RBF Neural Network. Innovative Computing, Information and Control, 2006. ICICIC '06. First International Conference. 2006, 8. 2. 270– 273.
[39] Machado R J, Neves P E C S A. Multi-model neural network for imageclassification. Cybernetic Vision, 1996. Proceedings., Second Workshop . 1996, 12. 57– 59.
[40] Ong E, Lin W, Lu Z. A no-reference quality metric for measuring image blur. Signal Processing and Its Applications, 2003. Proceedings. Seventh International Symposium. 2003, 7. 1. 469– 472.
[41] Fuzheng Yang, Shuai Wan, Yilin Chang. A novel objective no-reference metric for digital video quality assessment. Signal Processing Letters, IEEE. 2005, 10. 12(10). 685– 688.
[42] Hanghang Tong, Mingjing Li, Hong-Jiang Zhang. Learning No-Reference Quality Metric by Examples. Multimedia Modelling Conference, 2005. MMM 2005. Proceedings of the 11th International 2005, 1. 247– 254.
[43] Rafael C Gonzalez, Richard E Woods. Digital Image Processing Second Edition.北京:电子工业出版社.2002.7.141-193
[44] Shiqian Wu, Weisi Lin, Zhongkang Lu. Blind Blur Assessment for Vision-Based Applications. Multimedia and Expo, 2007 IEEE International Conference. 2007, 7. 1639– 1642.
[45] Xin Yu, Cairong Zou, Luxi Yang. Improved recursive inverse filtering method for blind image restoration. Signal Processing, 2002 6th International Conference. 2002, 8. 1. 37 - 40 .
[46] Berger T, Stromberg J O, Eltoft T. Adaptive regularized constrained least squares image restoration. Image Processing, IEEE Transactions. 1999, 9. 8(9). 1191 - 1203
[47] Graham L, Adhami R. Fusion of coincident images using spatial frequency discrimination by application of the wavelet transform. Time-Frequency and Time-Scale Analysis 1994. Proceedings of the IEEE-SP International Symposium 1994, 10. 326– 329.
[48]李建林,俞建成,孙胜利.基于梯度金字塔图像融合的研究.科学技术与工程. 2007,11. 7(22). 5818-5821.
[49]樊秀菊.脉冲耦合神经网络及基于此的图像分割与图像压缩. 2003 -西安电子科技大学:计算机应用技术.
[50] Yinghua Lu, Xue Feng, Jingbo Zhang. A Multi-focus Image Fusion Based on Wavelet and Region Detection. EUROCON, 2007. The International Conference on "Computer as a Tool" . 2007, 9. 294– 298.
[51] Yu Song, Mantian Li, Qingling Li. New Wavelet Based Multi-focus Image Fusion Scheme and Its Application on Optical Microscopy. Robotics and Biomimetics,2006.12.BIO '06. IEEE International Conference 2006, 12. 401– 405.