基于熵的深海资源图像处理算法研究与应用
|
摘要
图像处理是获取深海底结核覆盖率的关键技术,是采矿活动中指导采矿车按照最优的行走路线行走、优化采矿车的过程参数等作业需求的基础。开展深海资源图像处理技术的研究,对于提升深海采矿车的环境辨识能力有着重要的研究意义。针对真实环境下的深海图像进行图像增强、图像分割的算法研究,并计算出现有资源结核的覆盖率。最后通过使用DSP处理器来实现算法的移植,为系统升级预留充足的空间。
在图像处理的过程中充分利用深海图像的熵信息,首先在图像增强的研究中,针对传统模糊增强的分界点难以确定的问题,利用图像分割原理结合相对熵方法解决了分界点的确定。为了很好的提取出结核进而精确求取资源结核的覆盖率,通过分析一维最大熵的不足之处,采用了二维最大熵对图像进行分割的方法,并针对现有的深海资源图像因为光照不均而导致全局采用二位最大熵时存在误分割的问题,采用了分块的方法,对分块后的子图像采用二维最大熵进行分割,有效地减少了光照不均带来的影响,最后得到了较精确的覆盖率数据。
为了达到系统的实时性要求及方便深海资源采集系统图像处理的升级,以TM320DM642为核心处理器,在TI公司专用的图像处理系统开发平台上,成功地实现了算法从计算机到DSP系统的移植,为算法在实际系统中的应用奠定了基础。
Image processing is a critical technology for achieving the mining coverage in the deep-sea. It is the basis of moving the mining vehicle optimally and optimizing the process parameters. Researching on the deep-sea resource image processing is very important for improving the capacity of environment recognition about mine vehicle. This thesis is mainly concentrated on the processing algorithms of the deep-sea image, including image enhancement, image segmentation. And the coverage of mining is calculated by image processing. Using DSP (Digital Signal Processor), these algorithms can achieve the high speed running. And the algorithm procedure can be easily transplanted and upgraded.
The main work is that the entropy information is widely used in the processing of deep-sea image. The method of fuzzy enhancement on deep-sea image based on relative entropy is proposed. The problem of defining the cross point is well dealt through using the combination of the theory of image segmentation and relative entropy. In order to improve efficiency of segmentation and achieve accurate coverage, a method combing 2D maximum entropy is introduced. Owning to the uneven brightness in the deep-sea, the whole image of mining could not be segmented efficiently, lastly the image block is used. The method combing 2D maximum entropy is used in these blocks of the whole image. The method expresses a good effect to images segmentation.
In order to satisfy the real time request in the deep-sea mining, A DSP chip TMS320DM642 is used as core computer to design an image processing system. The paper completes the arithmetic transplant from PC to DSP and prepares well for application to practical systems.
在图像处理的过程中充分利用深海图像的熵信息,首先在图像增强的研究中,针对传统模糊增强的分界点难以确定的问题,利用图像分割原理结合相对熵方法解决了分界点的确定。为了很好的提取出结核进而精确求取资源结核的覆盖率,通过分析一维最大熵的不足之处,采用了二维最大熵对图像进行分割的方法,并针对现有的深海资源图像因为光照不均而导致全局采用二位最大熵时存在误分割的问题,采用了分块的方法,对分块后的子图像采用二维最大熵进行分割,有效地减少了光照不均带来的影响,最后得到了较精确的覆盖率数据。
为了达到系统的实时性要求及方便深海资源采集系统图像处理的升级,以TM320DM642为核心处理器,在TI公司专用的图像处理系统开发平台上,成功地实现了算法从计算机到DSP系统的移植,为算法在实际系统中的应用奠定了基础。
Image processing is a critical technology for achieving the mining coverage in the deep-sea. It is the basis of moving the mining vehicle optimally and optimizing the process parameters. Researching on the deep-sea resource image processing is very important for improving the capacity of environment recognition about mine vehicle. This thesis is mainly concentrated on the processing algorithms of the deep-sea image, including image enhancement, image segmentation. And the coverage of mining is calculated by image processing. Using DSP (Digital Signal Processor), these algorithms can achieve the high speed running. And the algorithm procedure can be easily transplanted and upgraded.
The main work is that the entropy information is widely used in the processing of deep-sea image. The method of fuzzy enhancement on deep-sea image based on relative entropy is proposed. The problem of defining the cross point is well dealt through using the combination of the theory of image segmentation and relative entropy. In order to improve efficiency of segmentation and achieve accurate coverage, a method combing 2D maximum entropy is introduced. Owning to the uneven brightness in the deep-sea, the whole image of mining could not be segmented efficiently, lastly the image block is used. The method combing 2D maximum entropy is used in these blocks of the whole image. The method expresses a good effect to images segmentation.
In order to satisfy the real time request in the deep-sea mining, A DSP chip TMS320DM642 is used as core computer to design an image processing system. The paper completes the arithmetic transplant from PC to DSP and prepares well for application to practical systems.
引文
[1]吴自军,周怀阳.刍议中国国际深海资源开发中长期发展战略.国际海底开发动态,2004,9(1):1-6
[2]中国大洋矿产资源研究开发协会.深海采矿政策研讨会论文集.北京:海洋出版社.1991,72-206
[3]陈峰.深海底采矿机器车运动建模与控制研究:[博士学位论文].长沙:中南大学,2002
[4]熊光辉,深海集矿机避障路径规划技术研究:[硕士学位论文].长沙:中南大学,2002
[5]肖体兵.深海采矿装置智能升沉补偿系统的研究:[博士学文论文].广州:广东工业大学,2004
[6]邱长军.钴结壳和基岩的特性和模拟研究:[博士学位论文].长沙:中南大学,2002
[7]吕华.海洋开发与管理的讨论.国土资源科技管理,1999,16(1):7-11
[8]谢龙水.我国深海采矿技术研究的阶段性成果及今后工作方向.湖南有色金属.1999,15(1):1-5
[9]P.Kaufman,J.P.Latimer,D.C.Tolefson.The Design and Operation of a Pacific Ocean Deep-Ocean Mining Test Ship:R/V Deep-sea Miner Ⅱ.Proceedings of the 17th Annual Offshore Technology Conference.Houston,Texas.1985,Vol.2,OTC4901:33-43
[10]S.K.Pal,R.A.King.Image enhancement using fuzzy sets.IEEE Trans.Electronics Letters.1980,16(10):376-378
[11]Y.Nie,K.E.Barner.Fuzzy transformation and its applications.2003 International Conference.On Image Processing.Sept 14-17,2003:890-893
[12]F.Farbiz,M.B.Menhaj,S.A.Motamedi et al.A new fuzzy logic filter for image enhancement.IEEE Trans.Systems,Man and Cybernetics.2000,30(1):110-119
[13]冷寒冰,拜丽萍,刘上乾.基于模糊域的红外图像自适应降噪增强技术.红外与激光工程,2003,32(5):521-522
[14]薛景浩,章毓晋.基于最大类间后验交叉熵的阈值化分割算法.中国图象图形学报,1999,10(6):111-114
[15]俞勇,施鹏飞,赵立初.基于最小能量的图像分割方法.红外与激光工程.1999,28(4):20-27
[16]任明武,杨静宇,孙涵.一种基于边缘模式的直方图构造新方法.计算机研 究与发展,2001,38(8):972-976
[17]程杰.一种基于直方图的分割方法.华中理工大学学报,1999,27(1):20-23
[18]乐宁,梁学军,翁世修.图像过渡区算法及其改进.红外与毫米波学报,2001,20(3):11-21
[19]任明武,杨静宇,孙涵.一种基于边缘模式的直方图构造新方法.计算机研究与发展,2001,38(8):972-976
[20]CHEN Zi-kuan,TAO Yang,CHEN Xin,et al.Wavelet Based Adaptive Thresholding Method for Image Segmentation.Optical Engineering,2001,40(5):868-874
[21]金立左,夏良正.图像分割的自适应模糊阈值法.中国图象图形学报,2000,5(1):44-47
[22]王保平,范九伦,谢维信.基于模糊熵的多值图像恢复方法.西安电子科技大学学报.2004,31(2):57-61
[23]李弼程,郭志刚,文超.图像的多层次模糊增强与边缘提取.模糊系统与数学,2000,14(4):77-83
[24]孔祥维,谢存,徐蔚然.基于多特征和模糊推理的边缘检测.电子学报.2000.28(6):36-39
[25]A.W.C.Liew,Hong Yan.An adaptive fuzzy clustering algorithm for medical image segmentation.Medical Imaging and Augmented Reality.June 10-12,2001:272-277
[26]薛忠,谢维信.模糊C均值聚类算法的一种初始化方法.系统工程与电子技术.1995,11:64-69
[27]J.K.Udupa,P.K.Saha.Fuzzy connectedness and image segmentation.Proceedings of the IEEE.2003,91(10):1649-1669
[28]K.K.Majumdar.Fuzzy fractals and fuzzy turbulence.IEEE Trans.Systems,Man and Cybernetics.Feb 1,2004:746-751
[29]谢维信.工程模糊数学.西安:西安电子科技大学出版社.1994
[30]王博,潘泉,张洪才等.图像平滑与边缘检测的模糊向量描述.小型微型计算机系统.1999.20(3):218-221
[31]Pal S K,King R A.On Edge Detection of X-ray Images using Fuzzy Sets.IEEE Trans.Patt.Analysis and Machine Intelligence.1983.5(1):69-77
[32]徐立中.数字图像的智能信息处理.北京:国防工业出版社.2004:4-11,22-61,87-91
[33]周泰文,王晓星.模糊数学基础简明教程.武汉:华中理工大学出版社.1993
[34]白景峰,赵学增,强锡富等.基于模糊梯度法的边缘检测方法.控制与决策.2001,16(3):351-353
[35]吴国雄,陈武凡.图像的模糊增强与聚类分割.小型微型计算机系统.1994,15(11):21-26
[36]Pal S K,King R A,Hashim A A.Automatic gray level thresholding through index of fuzziness and entropy.Pattern Recognition Letter,1983,1:141-146
[37]Pal S K,King R A.Image Enhancement Using Smoothing With Fuzzy Sets.IEEE Trans.System,Man,Cybern,1981,11(7):494-501
[38]Pal S K,Rosenfeld A.Image enhancement and thresholding by optimization of fuzzy compactness.Pattern Recognition Lett.1988,7:77-86
[39]章毓晋,图像工程(上册)—图像处理与分析.北京:清华大学出版社:1999
[40]Van den Bergh F.An analysis of particle swarm optimizers.South Africa,Department of Computer Science,University of Pretoria,2002
[41]Kapur J N,Sahoo P K,Wong A K C.A new method for gray-level picture thresholding using the entropy of the histogram.Computer Vision,Graphics and Image Processing,1985,29:273-285
[42]艾海舟,武勃.图像处理、分析与机器视觉.北京:人民邮电出版社,2003
[43]杨枝灵,王开.Visual C++数字图像获取处理及实践应用.人民邮电出版社,2003
[44]S.C.Sahasrabudhe,K.S.D.Gupta.A Valley-seeking Threshold Selection Technique,Computer Vision and Image Processing,(A.Rosenfeld,L.Shapiro,Eds),Academic Press.1992:55-65
[45]Otsu N.A threshold selection method from grey-level histograms.IEEE Trans.System.Man Cybemet,1979,SMC-9:62-66
[46]Kittiler J,Illingworth J.Minimum error thresholding.Pattem Recognition,1986,19(1):41-47
[47]Pun T.A new method for grey-level picture thresholding using the entropy of the histogram.Signal Process.1980,2(3):223-237
[48]Pal N R,Pal S K.A review on image segmentation techniques.Pattern Recognition,1993,26(9):1277-1294
[49]Yen J G,Chang F J,Chang S.A new criterion for automatic multilevel thresholding.IEEE Trans.on Image Processing,1995,4(3):233-260
[50]Sahoo P,Wilkins C,Yeager J.Threshold selection using Renyi's entropy.Pattern Recognition,1997,30(1):71-84
[51]Cheng H D,Chen Y H,Sun Y.A novel fuzzy entropy approach to image enhancement and thresholding.Signal Processing,1999,75(3):277-301
[52]周德龙,申石磊,蒲小勃等,基于灰度.梯度共生矩阵模型的最大熵阈值处理算法,小型微型计算机系统,2002,22(4):1-4
[53]北京瑞泰创新科技有限责任公司,ICETEK-DM642-IDK-M图像、语音和网络处理系统,硬件、软件使用说明书和实验指导.
[54]涂晓星.基于DSP的通用实时图像处理系统设计与研究.浙江大学,2004
[2]中国大洋矿产资源研究开发协会.深海采矿政策研讨会论文集.北京:海洋出版社.1991,72-206
[3]陈峰.深海底采矿机器车运动建模与控制研究:[博士学位论文].长沙:中南大学,2002
[4]熊光辉,深海集矿机避障路径规划技术研究:[硕士学位论文].长沙:中南大学,2002
[5]肖体兵.深海采矿装置智能升沉补偿系统的研究:[博士学文论文].广州:广东工业大学,2004
[6]邱长军.钴结壳和基岩的特性和模拟研究:[博士学位论文].长沙:中南大学,2002
[7]吕华.海洋开发与管理的讨论.国土资源科技管理,1999,16(1):7-11
[8]谢龙水.我国深海采矿技术研究的阶段性成果及今后工作方向.湖南有色金属.1999,15(1):1-5
[9]P.Kaufman,J.P.Latimer,D.C.Tolefson.The Design and Operation of a Pacific Ocean Deep-Ocean Mining Test Ship:R/V Deep-sea Miner Ⅱ.Proceedings of the 17th Annual Offshore Technology Conference.Houston,Texas.1985,Vol.2,OTC4901:33-43
[10]S.K.Pal,R.A.King.Image enhancement using fuzzy sets.IEEE Trans.Electronics Letters.1980,16(10):376-378
[11]Y.Nie,K.E.Barner.Fuzzy transformation and its applications.2003 International Conference.On Image Processing.Sept 14-17,2003:890-893
[12]F.Farbiz,M.B.Menhaj,S.A.Motamedi et al.A new fuzzy logic filter for image enhancement.IEEE Trans.Systems,Man and Cybernetics.2000,30(1):110-119
[13]冷寒冰,拜丽萍,刘上乾.基于模糊域的红外图像自适应降噪增强技术.红外与激光工程,2003,32(5):521-522
[14]薛景浩,章毓晋.基于最大类间后验交叉熵的阈值化分割算法.中国图象图形学报,1999,10(6):111-114
[15]俞勇,施鹏飞,赵立初.基于最小能量的图像分割方法.红外与激光工程.1999,28(4):20-27
[16]任明武,杨静宇,孙涵.一种基于边缘模式的直方图构造新方法.计算机研 究与发展,2001,38(8):972-976
[17]程杰.一种基于直方图的分割方法.华中理工大学学报,1999,27(1):20-23
[18]乐宁,梁学军,翁世修.图像过渡区算法及其改进.红外与毫米波学报,2001,20(3):11-21
[19]任明武,杨静宇,孙涵.一种基于边缘模式的直方图构造新方法.计算机研究与发展,2001,38(8):972-976
[20]CHEN Zi-kuan,TAO Yang,CHEN Xin,et al.Wavelet Based Adaptive Thresholding Method for Image Segmentation.Optical Engineering,2001,40(5):868-874
[21]金立左,夏良正.图像分割的自适应模糊阈值法.中国图象图形学报,2000,5(1):44-47
[22]王保平,范九伦,谢维信.基于模糊熵的多值图像恢复方法.西安电子科技大学学报.2004,31(2):57-61
[23]李弼程,郭志刚,文超.图像的多层次模糊增强与边缘提取.模糊系统与数学,2000,14(4):77-83
[24]孔祥维,谢存,徐蔚然.基于多特征和模糊推理的边缘检测.电子学报.2000.28(6):36-39
[25]A.W.C.Liew,Hong Yan.An adaptive fuzzy clustering algorithm for medical image segmentation.Medical Imaging and Augmented Reality.June 10-12,2001:272-277
[26]薛忠,谢维信.模糊C均值聚类算法的一种初始化方法.系统工程与电子技术.1995,11:64-69
[27]J.K.Udupa,P.K.Saha.Fuzzy connectedness and image segmentation.Proceedings of the IEEE.2003,91(10):1649-1669
[28]K.K.Majumdar.Fuzzy fractals and fuzzy turbulence.IEEE Trans.Systems,Man and Cybernetics.Feb 1,2004:746-751
[29]谢维信.工程模糊数学.西安:西安电子科技大学出版社.1994
[30]王博,潘泉,张洪才等.图像平滑与边缘检测的模糊向量描述.小型微型计算机系统.1999.20(3):218-221
[31]Pal S K,King R A.On Edge Detection of X-ray Images using Fuzzy Sets.IEEE Trans.Patt.Analysis and Machine Intelligence.1983.5(1):69-77
[32]徐立中.数字图像的智能信息处理.北京:国防工业出版社.2004:4-11,22-61,87-91
[33]周泰文,王晓星.模糊数学基础简明教程.武汉:华中理工大学出版社.1993
[34]白景峰,赵学增,强锡富等.基于模糊梯度法的边缘检测方法.控制与决策.2001,16(3):351-353
[35]吴国雄,陈武凡.图像的模糊增强与聚类分割.小型微型计算机系统.1994,15(11):21-26
[36]Pal S K,King R A,Hashim A A.Automatic gray level thresholding through index of fuzziness and entropy.Pattern Recognition Letter,1983,1:141-146
[37]Pal S K,King R A.Image Enhancement Using Smoothing With Fuzzy Sets.IEEE Trans.System,Man,Cybern,1981,11(7):494-501
[38]Pal S K,Rosenfeld A.Image enhancement and thresholding by optimization of fuzzy compactness.Pattern Recognition Lett.1988,7:77-86
[39]章毓晋,图像工程(上册)—图像处理与分析.北京:清华大学出版社:1999
[40]Van den Bergh F.An analysis of particle swarm optimizers.South Africa,Department of Computer Science,University of Pretoria,2002
[41]Kapur J N,Sahoo P K,Wong A K C.A new method for gray-level picture thresholding using the entropy of the histogram.Computer Vision,Graphics and Image Processing,1985,29:273-285
[42]艾海舟,武勃.图像处理、分析与机器视觉.北京:人民邮电出版社,2003
[43]杨枝灵,王开.Visual C++数字图像获取处理及实践应用.人民邮电出版社,2003
[44]S.C.Sahasrabudhe,K.S.D.Gupta.A Valley-seeking Threshold Selection Technique,Computer Vision and Image Processing,(A.Rosenfeld,L.Shapiro,Eds),Academic Press.1992:55-65
[45]Otsu N.A threshold selection method from grey-level histograms.IEEE Trans.System.Man Cybemet,1979,SMC-9:62-66
[46]Kittiler J,Illingworth J.Minimum error thresholding.Pattem Recognition,1986,19(1):41-47
[47]Pun T.A new method for grey-level picture thresholding using the entropy of the histogram.Signal Process.1980,2(3):223-237
[48]Pal N R,Pal S K.A review on image segmentation techniques.Pattern Recognition,1993,26(9):1277-1294
[49]Yen J G,Chang F J,Chang S.A new criterion for automatic multilevel thresholding.IEEE Trans.on Image Processing,1995,4(3):233-260
[50]Sahoo P,Wilkins C,Yeager J.Threshold selection using Renyi's entropy.Pattern Recognition,1997,30(1):71-84
[51]Cheng H D,Chen Y H,Sun Y.A novel fuzzy entropy approach to image enhancement and thresholding.Signal Processing,1999,75(3):277-301
[52]周德龙,申石磊,蒲小勃等,基于灰度.梯度共生矩阵模型的最大熵阈值处理算法,小型微型计算机系统,2002,22(4):1-4
[53]北京瑞泰创新科技有限责任公司,ICETEK-DM642-IDK-M图像、语音和网络处理系统,硬件、软件使用说明书和实验指导.
[54]涂晓星.基于DSP的通用实时图像处理系统设计与研究.浙江大学,2004