数学形态学图象处理算法应用研究
|
摘要
随着对海洋的开发和勘探需求的增长，用于民用和军用等各领域的水下目标探测技术得到了突飞猛进的发展。其中，图象信息作为主要信息来源成为水下探测技术研究的焦点之一。成像声纳由于具有更大的成像范围和抗干扰能力，成为水下目标搜索和识别的主要工具。
本论文的工作是智能水下机器人“声视觉系统”中的一部分，进行水声图象的预处理和识别算法的研究。在众多的算法中，基于频域分割原理的线性处理方法在去除噪声的同时会模糊图象的边缘，很难满足针对水声图象特点的“保边去噪”的要求；而一些非线性方法又缺乏统一的理论基础，使其在系统整体的应用中受到限制。
本文针对水声图象预处理和识别的特殊性，把数学形态学的形态变换应用到“声视觉系统”的预处理和识别中，实现了以数学形态学为理论基础的非线性处理方法，使得图象的去噪、边缘提取和物体识别等处理较传统方法优越和快速。
文中介绍了二值形态学基本变换与其改进算法，灰度形态变换与彩色形态变换，并详细分析了它们的性质以及用于图象预处理和分析识别时的滤波性能。另外对一些经典的图象处理和分析算法进行了充分的仿真研究和比较，得出了有用的结论，同时给出了大量水声图象的处理结果。最后实现了一些常用线性和非线性算法，与数学形态学算法一起形成一个适用于水声图象处理和分析的软件。
With the growth of the requirements for developing and exploration of the resources in the ocean, kinds of technology to detect objects underwater have been developed rapidly. As the most important information sources, video information became one of the focuses in the research of underwater objects' detection. The advanced imaging sonar has been the primary tool to do objects' detecting and recognizing underwater because of its wider imaging range and stronger anti-jamming capability.
In this paper, the work is a part of acoustic vision in the Intelligent Autonomous Underwater Vehicle. The aim of the task is researching the algorithms for the preprocessing and recognition of underwater acoustic images. Even if numerous methods, but most of them are linear processing which have the side effect of bluring edges, so can not meet the case of detecting edges and suppressing noise. And the use of some nonlinear methods we have had are also limited for the lack of theoretical basis.
With the characteristics of the task of preprocessing and recognition of underwater acoustic images, the nonlinear methods based on the mathematical morphology was proposed here. The performance of it is better than the conventional methods for the case of suppressing noise, detecting edges and recognition.
In this paper, we introduced the binary morphology and its improvement, gray-scale and color morphology and analyzed their characteristics in filtering particularly. Then we simulated and compared the classical image processing and recognition algorithms adequately, obtain the useful conclusions and many processing results.
Finally, we constructed a software which included some linear and nonlinear algorithms along with the mathematical morphology for the preprocessing and analyzing in underwater acoustic images.
本论文的工作是智能水下机器人“声视觉系统”中的一部分，进行水声图象的预处理和识别算法的研究。在众多的算法中，基于频域分割原理的线性处理方法在去除噪声的同时会模糊图象的边缘，很难满足针对水声图象特点的“保边去噪”的要求；而一些非线性方法又缺乏统一的理论基础，使其在系统整体的应用中受到限制。
本文针对水声图象预处理和识别的特殊性，把数学形态学的形态变换应用到“声视觉系统”的预处理和识别中，实现了以数学形态学为理论基础的非线性处理方法，使得图象的去噪、边缘提取和物体识别等处理较传统方法优越和快速。
文中介绍了二值形态学基本变换与其改进算法，灰度形态变换与彩色形态变换，并详细分析了它们的性质以及用于图象预处理和分析识别时的滤波性能。另外对一些经典的图象处理和分析算法进行了充分的仿真研究和比较，得出了有用的结论，同时给出了大量水声图象的处理结果。最后实现了一些常用线性和非线性算法，与数学形态学算法一起形成一个适用于水声图象处理和分析的软件。
With the growth of the requirements for developing and exploration of the resources in the ocean, kinds of technology to detect objects underwater have been developed rapidly. As the most important information sources, video information became one of the focuses in the research of underwater objects' detection. The advanced imaging sonar has been the primary tool to do objects' detecting and recognizing underwater because of its wider imaging range and stronger anti-jamming capability.
In this paper, the work is a part of acoustic vision in the Intelligent Autonomous Underwater Vehicle. The aim of the task is researching the algorithms for the preprocessing and recognition of underwater acoustic images. Even if numerous methods, but most of them are linear processing which have the side effect of bluring edges, so can not meet the case of detecting edges and suppressing noise. And the use of some nonlinear methods we have had are also limited for the lack of theoretical basis.
With the characteristics of the task of preprocessing and recognition of underwater acoustic images, the nonlinear methods based on the mathematical morphology was proposed here. The performance of it is better than the conventional methods for the case of suppressing noise, detecting edges and recognition.
In this paper, we introduced the binary morphology and its improvement, gray-scale and color morphology and analyzed their characteristics in filtering particularly. Then we simulated and compared the classical image processing and recognition algorithms adequately, obtain the useful conclusions and many processing results.
Finally, we constructed a software which included some linear and nonlinear algorithms along with the mathematical morphology for the preprocessing and analyzing in underwater acoustic images.
引文
[1] 阮秋琦编著，数字图象处理学，电子工业出版社，2001
[2] 边肇祺，张学工等编著，模式识别，清华大学出版社，2000
[3] 陈贺新著，非线性滤波器与数字图象处理，国防工业出版社，1997
[4] 崔屹编著，图象处理与分析—数学形态学方法及应用，科学出版社， 2000
[5] 唐常青等，数学形态学方法及其应用，科学出版社，1990
[6] 龚炜，石青云，程民德著，数字空间中的数学形态学—理论与应用，科学出版社，1997
[7] 陈武凡主编，小波分析及其在图象处理中的应用，科学出版社，2002
[8] 章毓晋著，图象分割，科学出版社，2001，
[9] R．J．尤立克著，洪深译，水声原理，哈尔滨船舶工程学院出版社，1989
[10] 贾云得著，机器视觉，科学出版社，2000
[11] 孙兆林编著，Matlab 6.x图象处理，清华大学出版社，2002
[12] 张兆礼，赵春晖，梅晓丹著，现代图象处理技术及Matlab实现，人民邮电出版社，2001
[13] Microsoft Visual C++ 6.0技术内幕(第五版)，北京希望电子出版社，1999
[14] 何斌等，Visual C++数字图象处理，人民邮电出版社，2001
[15] 向世明编著，Visual C++数字图象与图形处理，电子工业出版社，2002
[16] 严学强著，数学形态学方法在水声图象预处理中的应用研究，哈尔滨船舶工程学院硕士学位论文，1995
[17] 刘志敏著，数学形态学在图象分析中的应用研究，上海交通大学硕士学位论文，1998
[18] 赵春晖著，数字形态滤波器理论及其算法研究，哈尔滨工业大学博士学位论文，1998
[19] 孙晓南著，水下机器人声视觉系统中的数字图象处理算法研究，哈尔
滨工程大学硕士学位论文，2000
[20] 卞红雨著，水下机器人声视觉系统的几项关键技术研究，哈尔滨工程大学博士学位论文，1999
[21] 李霞著，水下智能机器人声视觉成像与信息处理技术研究，哈尔滨工程大学硕士学位论文，2002
[22] 景晓军，李剑锋，熊玉庆，静止图象的一种自适应平滑滤波算法，通信学报，2002(10)
[23] 基于数学形态学的边缘检测和图象分割方法，自动化信息，2001(2)
[24] 戴青云，余英林，数学形态学在图象处理中的应用进展，控制理论与应用，2001(18)
[25] 刘直芳等，基于多尺度彩色形态矢量算子的边缘检测，中国图象图形学报，2002．Vol．7
[26] Milan Sonka, Vaclav Hlavac and Roger Boyle, Image Processing, Analysis, and Machine Vision Second Edition (英文版),人民邮电出版社, 2002
[27] Jon Bates，Tim Tompkin著，何健辉，董方鹏等译，实用Visual C++ 6.0教程，清华大学出版社，2000
[28] P. G. Auran and O. Silven. Underwater sonar range sensing and 3D image formation, Control Engineering Practice, 1996(3), Vol 4
[29] Nicholas Molton, Seophen Se, Michael Brady, David Lee ahd Penny Probert, Robotic sensing for the partially sighted, Robotics and Autonomous Systems, 1999(2), Vol 26
[30] Guy R. Cochrane and Kevin D. Lafferty, Use of acoustic classification of sidescan sonar data for mapping benthic habitat in the Northern Channel Islands, California, Continental Shelf Research, 2000(3), Vol 22
[31] C.J. Moran, Amorphological Transformation for sharpening edges of features before segmentation, Computer Vision, Graphics, and Image Processing, 1990, Vol 49
[32] Gady Agam, Regulated morphological operations, Pattern Recognition, 1999, Vol 32
[33] Pierre Soille, Short survey On morphological operators based on rank filters, Pattern Recognition, 2002, Vol 35
[34] Mike Nachtegael and Etienne E. Kerre, Connections between binary, gray-scale and fuzzy mathematical morphologies, Fuzzy Sets and Systems, 2001, Vol 124
[35] G. Louverdis, M. I. Vardavoulia, I. Andreadis, Ph. Tsalides, A new approach to morphological color image processing, Pattern Recognition, 2002, Vol 35
[36] Jos B. T. M. Roerdink, Group morphology, Pattern Recognition, 2000, Vol 33
[2] 边肇祺，张学工等编著，模式识别，清华大学出版社，2000
[3] 陈贺新著，非线性滤波器与数字图象处理，国防工业出版社，1997
[4] 崔屹编著，图象处理与分析—数学形态学方法及应用，科学出版社， 2000
[5] 唐常青等，数学形态学方法及其应用，科学出版社，1990
[6] 龚炜，石青云，程民德著，数字空间中的数学形态学—理论与应用，科学出版社，1997
[7] 陈武凡主编，小波分析及其在图象处理中的应用，科学出版社，2002
[8] 章毓晋著，图象分割，科学出版社，2001，
[9] R．J．尤立克著，洪深译，水声原理，哈尔滨船舶工程学院出版社，1989
[10] 贾云得著，机器视觉，科学出版社，2000
[11] 孙兆林编著，Matlab 6.x图象处理，清华大学出版社，2002
[12] 张兆礼，赵春晖，梅晓丹著，现代图象处理技术及Matlab实现，人民邮电出版社，2001
[13] Microsoft Visual C++ 6.0技术内幕(第五版)，北京希望电子出版社，1999
[14] 何斌等，Visual C++数字图象处理，人民邮电出版社，2001
[15] 向世明编著，Visual C++数字图象与图形处理，电子工业出版社，2002
[16] 严学强著，数学形态学方法在水声图象预处理中的应用研究，哈尔滨船舶工程学院硕士学位论文，1995
[17] 刘志敏著，数学形态学在图象分析中的应用研究，上海交通大学硕士学位论文，1998
[18] 赵春晖著，数字形态滤波器理论及其算法研究，哈尔滨工业大学博士学位论文，1998
[19] 孙晓南著，水下机器人声视觉系统中的数字图象处理算法研究，哈尔
滨工程大学硕士学位论文，2000
[20] 卞红雨著，水下机器人声视觉系统的几项关键技术研究，哈尔滨工程大学博士学位论文，1999
[21] 李霞著，水下智能机器人声视觉成像与信息处理技术研究，哈尔滨工程大学硕士学位论文，2002
[22] 景晓军，李剑锋，熊玉庆，静止图象的一种自适应平滑滤波算法，通信学报，2002(10)
[23] 基于数学形态学的边缘检测和图象分割方法，自动化信息，2001(2)
[24] 戴青云，余英林，数学形态学在图象处理中的应用进展，控制理论与应用，2001(18)
[25] 刘直芳等，基于多尺度彩色形态矢量算子的边缘检测，中国图象图形学报，2002．Vol．7
[26] Milan Sonka, Vaclav Hlavac and Roger Boyle, Image Processing, Analysis, and Machine Vision Second Edition (英文版),人民邮电出版社, 2002
[27] Jon Bates，Tim Tompkin著，何健辉，董方鹏等译，实用Visual C++ 6.0教程，清华大学出版社，2000
[28] P. G. Auran and O. Silven. Underwater sonar range sensing and 3D image formation, Control Engineering Practice, 1996(3), Vol 4
[29] Nicholas Molton, Seophen Se, Michael Brady, David Lee ahd Penny Probert, Robotic sensing for the partially sighted, Robotics and Autonomous Systems, 1999(2), Vol 26
[30] Guy R. Cochrane and Kevin D. Lafferty, Use of acoustic classification of sidescan sonar data for mapping benthic habitat in the Northern Channel Islands, California, Continental Shelf Research, 2000(3), Vol 22
[31] C.J. Moran, Amorphological Transformation for sharpening edges of features before segmentation, Computer Vision, Graphics, and Image Processing, 1990, Vol 49
[32] Gady Agam, Regulated morphological operations, Pattern Recognition, 1999, Vol 32
[33] Pierre Soille, Short survey On morphological operators based on rank filters, Pattern Recognition, 2002, Vol 35
[34] Mike Nachtegael and Etienne E. Kerre, Connections between binary, gray-scale and fuzzy mathematical morphologies, Fuzzy Sets and Systems, 2001, Vol 124
[35] G. Louverdis, M. I. Vardavoulia, I. Andreadis, Ph. Tsalides, A new approach to morphological color image processing, Pattern Recognition, 2002, Vol 35
[36] Jos B. T. M. Roerdink, Group morphology, Pattern Recognition, 2000, Vol 33