图像处理在高炉料面可视化监控中的理论应用
|
摘要
图像处理已在现如今的生产生活当中越来越多的被我们认识,其中对于图像的无缝拼接技术更是被广泛应用。图像拼接技术在太空探索,海底调查,医疗,气象,地质勘探,军事,视频压缩和存储,视频索引和检索,三维物体重建,等领域得到了广泛的应用。
在高炉冶炼过程中,控制炉内冶金材料料面的燃烧状态和分布情况对掌握和了解的气体流量控制材料表面的分布状况、指导正确的高炉操作是非常重要的,但通常电子监控设备一次成像的区域有限,很难帮助我们对整体料面情况做出判断。
本文提出了一种设计方案,以高炉料面监控的数字成像图片为基础,通过在炉喉部位开设风孔观测炉内实时情况。首先,先建立了监控系统的总体方案,尤其是由于风孔位置的选择可不同于以往在顶部的开设,使画面不再是顶部俯视料面,而是由侧上方直接观测,克服了由于俯视料面而无法获得具体料面堆叠状况的弊端；其次,再利用图片拼接技术完成整个料面实际的广视角全景图像,CCD驱动平台帮助摄像头获取图片资料,再利用计算机完成图像的无缝拼接。利用淡入淡出的融合算法,消除了重叠像素的亮度差异而造成的误判,以达到无缝拼接的根本目标。最终获得一副反映料面立体形状的广角全景图像,帮助人们对炉内的炉内生产状态及料面情况做出准确的判断。
Image processing in the production is now more and more known where the image of the seamless splicing technology is widely used. Image stitching technology have been widely used in space exploration, seabed surveys, medical, meteorological, geological exploration, military, video compression and storage, video indexing and retrieval, three-dimensional object reconstruction, and other fields.
In the blast furnace process, it is very important to get the surface of the furnace combustion of metallurgical materials, material status and the gas flow by control the distribution of surface materials to guide the correct operation of blast furnace. But usually an electronic monitoring area is limited, it is difficult to help us make the overall situation to determine situation of the surface material.
This paper presents a design, the surface charge monitor with high picture-based digital imaging, site set up by the furnace throat hole the wind real-time observation of the furnace. First of all, establish a monitoring system of the overall program, particularly because the wind hole may be different from the previous selection of the opening at the top, the screen is no longer at the top overlooking the material surface, but by the side of the top of the direct observation, to overcome the drawbacks of stacked material surface which caused by overlooking the specific situation view. Second, by re-use photo stitching technology, the wide viewing angle panoramic images of the material surface are finished actually. The driven platform of CCD camera is used to help get pictures, then we use the computer to complete the image without seam stitching. By fusion algorithm, we eliminate overlapping pixel intensity differences caused by miscarriage of justice, in order to achieve the fundamental goal of seamless. Ultimately, a three-dimensional shape of the surface material to reflect the wide-angle panoramic images to help people to make accurate judgments on the state of the furnace and furnace material side of production.
在高炉冶炼过程中,控制炉内冶金材料料面的燃烧状态和分布情况对掌握和了解的气体流量控制材料表面的分布状况、指导正确的高炉操作是非常重要的,但通常电子监控设备一次成像的区域有限,很难帮助我们对整体料面情况做出判断。
本文提出了一种设计方案,以高炉料面监控的数字成像图片为基础,通过在炉喉部位开设风孔观测炉内实时情况。首先,先建立了监控系统的总体方案,尤其是由于风孔位置的选择可不同于以往在顶部的开设,使画面不再是顶部俯视料面,而是由侧上方直接观测,克服了由于俯视料面而无法获得具体料面堆叠状况的弊端；其次,再利用图片拼接技术完成整个料面实际的广视角全景图像,CCD驱动平台帮助摄像头获取图片资料,再利用计算机完成图像的无缝拼接。利用淡入淡出的融合算法,消除了重叠像素的亮度差异而造成的误判,以达到无缝拼接的根本目标。最终获得一副反映料面立体形状的广角全景图像,帮助人们对炉内的炉内生产状态及料面情况做出准确的判断。
Image processing in the production is now more and more known where the image of the seamless splicing technology is widely used. Image stitching technology have been widely used in space exploration, seabed surveys, medical, meteorological, geological exploration, military, video compression and storage, video indexing and retrieval, three-dimensional object reconstruction, and other fields.
In the blast furnace process, it is very important to get the surface of the furnace combustion of metallurgical materials, material status and the gas flow by control the distribution of surface materials to guide the correct operation of blast furnace. But usually an electronic monitoring area is limited, it is difficult to help us make the overall situation to determine situation of the surface material.
This paper presents a design, the surface charge monitor with high picture-based digital imaging, site set up by the furnace throat hole the wind real-time observation of the furnace. First of all, establish a monitoring system of the overall program, particularly because the wind hole may be different from the previous selection of the opening at the top, the screen is no longer at the top overlooking the material surface, but by the side of the top of the direct observation, to overcome the drawbacks of stacked material surface which caused by overlooking the specific situation view. Second, by re-use photo stitching technology, the wide viewing angle panoramic images of the material surface are finished actually. The driven platform of CCD camera is used to help get pictures, then we use the computer to complete the image without seam stitching. By fusion algorithm, we eliminate overlapping pixel intensity differences caused by miscarriage of justice, in order to achieve the fundamental goal of seamless. Ultimately, a three-dimensional shape of the surface material to reflect the wide-angle panoramic images to help people to make accurate judgments on the state of the furnace and furnace material side of production.
引文
[1]艾海舟,武勃等译。Milan Sonka, Vaclav Hlavac, Roger Boyle著。图像处理、分析与机器视觉(第二版)[M]。人民邮电出版社,2003年9月：98-135
[2]潘立公,汤晓安.一种360度柱面全景图快速生成算法.研究与设计[J],2005,17(4)：19-35
[3]杨枝灵,王开.Visual C++数字幽像获取处理及实践应用[M].北京：人民邮电出版社,2003：100-360
[4]刘翔,曾芬芳.基于IBR技术的全景幽生成.华东船舶工业学院学报(自然科学版)[J].2005,15(1)：32-37
[5]沈建国,戴永刚.基于局部熵差的栅格地图拼接方法[J].华东师范大学学报(自然科学版).2002,(4)：48-53
[6]王昌军,高炉料面区域温度特征智能提取方法研究与应用,中南大学,2010,TF53
[7]陈先中,丁爱华,高炉雷达料面成像系统的设计与实现,冶金自动化[J],2009年3月,第33卷,第二期
[8]王水刚,热像仪在高炉料面温度检测中的应用,宝钢技术,2009年,第5期,FT068.2
[9]张碧梧,高炉料面分布的测量,冶金自动化[J],1985年第05期,
[10]曾秀丽；郝胜男;谭长江,高炉红外/雷达复合探测系统的设计与实现,制造业自动化[J],2010年第07期,
[11]尚明姝,解凯。一种基于特征的全自动图像拼接算法。微计算机应用[J],2006年11月,第2卷第6期
[12]张静,胡志萍,刘志泰,欧宗瑛。基于轮廓相位相关的图像自动拼接。大连理工大学学报[J],2005年1月
[13]崔汉国,张星,刘晓成。一种改进的全景图生成算法。海军工程大学学报[J],第18卷,第2期,2006年4月
[14]Richard Hartley, Aadrew Zisserman. Multiple View Geometry in compuer Vision[Second Edition]. Cambridge:The Press Syndicate The University of Cambridge. UK,2000[M].
[15]SZELISKI R. Video mosaics for virtual environments. IEEE Trans Comput Graphicsand Appl,1996,16(2):22-30[J]。
[16]蔺想红,王维盛。一种基于特征匹配的全景图自动拼接方法。西北师范大学学报(自然科学版),第41卷2005年第4期。[J]
[17]赵向阳,杜利民。一种全自动稳健的图像拼接融合算法。中国图象图形学报,第9卷,第4期,2004年4月[J]。
[18]文雅玫,李思昆。一种遥感图像高精度自动拼接算法。湘南学院学报,2006年10月,第27卷,第5期[J]。
[19]谭磊,薛彦斌。一种基于特征点的图像匹配算法。天津理工大学学报,第22卷,第6期,2006年12月[J]。
[20]何光渝,高永利。Visual Fortran常用数值算法集。科学出版社,2002年4月。[M]
[15]钟力,张茂军,孙立峰,李云浩。360度柱面全景图象生成算法及其实现。小型微型计算机系统,第20卷,第12期,1999年12月[J]。
[16]曹俊杰,苏志勋。全景图像拼接算法。大连理工大学学报,第43卷增刊1,2003年10月[J]。
[17]孙立峰,钟力。虚拟实景空间的实时漫游。中国图象图形学报,第4卷(A版)第6期,1999年6月[J]。
[18]张韵华,奚梅成,陈效群。数值计算方法和算法。科学出版社,2000年1月[M]。
[19]明日科技宋坤Visual C++视频技术方案宝典人民邮电出版社2008年02月[M]。
[20]L Millina, Q Bishop. Plenoptic Modeling:An image based rendering system[A]. SIGGRAPH.1995:39-46
[21]R. Szeliski, H. YShum. Creating full view panoramic image mosaics and Environmentmpas[A]. SIGGR. APH,1997:25 1.257
[22]Hongjie Xie et al.. Automatic image registration based on FFT algorithm and IDL/ENVI[C]. International Conference on Remote Sensing and GIS/ GPS。ICORG.2000,1:391-402
[23]R. Szeliski. Video mosaics for virtual environments[J]. IEEE ComputerGraphics andApplications,1996,16(2):22-30
[24]Thierry Blaszka. Rachid Deriche. Recovering and characterizing image features using an efficient model based approach[c]. Technical Report 2422。 1NRIA. November 1994
[25]Bao P., Xu D. Complex wavelet—based image mosaics using edge preserving visual perception modeling[J]. Computers and Graphics.1999。23(3):309-321
[26]Chen Shizhe, HU Tao, Liu Guodong el al.. Fast and precision image mosaic based on grating[J]. Optics and Precision Engineering.2006.2(14):242-246
[27]陈世哲.胡涛,刘Ⅰ珥栋等.基于光栅的快速精确图像拼接[J].光学精密工程.2006.2(14)：242-246
[28]Fan Q W. Wang X L。Hou L et al.. Speeding mosaic technology based on isometry serial image[J]. Journal of Sichuan University(Engineering science edition).2005.37(Ⅰ):39-142
[29]樊庆文,王小龙.侯力等.基r等距序列}薹Ⅰ像的伙速拼接技术FJ].四川大学学报(工程科学版).2005,1(36)：139-142
[30]Mei X M, Liu Z X, Wang H C et al.. Advanced Geometry (second edition)[M]. Beijing:Higher Education Press.2000
[31]梅向明.刘增贤,王汇淳等.高等几何(第二版)[M].北京：高等教育出版社.2000
[32]Ye Chuanqi, Miao Qiguang. Wang Baoshu. An image fusion algorithm using region segmentation and contourlet transform[J]. Acta Optica Sinica.2008.3(28):447-453
[33]叶传奇.苗启广,王’社树.基于Ⅸ域分割和Counterlet变换的图像融合算法[J].光学学报,2008.3(28)：447-453
[34]Liu C; uixi. I。iu Dongmei。Liu Fengpeng et al.. A Robust image registration algorithm based on feature points matching[J]. Acta Optica Sinica.2008.28(3):454-461
[35]刘贵磐。刘冬梅.刘风鹏等.一种稳健的特征点配准算法[J].光学学报[J]2008.28(3)：454-461
[36]Hu Shejiao, Ge Xiwang, Su I。ing. Panoramic image mosaic based on domain correlation [J]. J. Hejei University of Technology,2007。30(1):29-32
[37]胡社教.葛两旺,苏菱.基于相位相关的伞景图像拼接[J].合肥工业大学学报(自然科学版).2007。30(1)：29-32
[38]Li Wansong, Su Likun。Su Xianyu. Phase-measuring profilometry in big scale measurement. Acta Optica Sinica[J].2000,20(6):792-796
[39]李万松.苏礼坤.苏显渝.相位检测面形术在大尺度三维面形测量中的应用.光学学报[J].2000。20(6)：792-796
[40]Jing(renqiang, Zhang Tianxu, Yang Weidong el al.. Loca adaptive mosaic algorithm for serial images [J]. Infrared andLaser Engineering.2004。33(6): 622-625
[41]荆根强,张灭序.杨1!东等.局部自适应序列图像拼接[J].红外与激光工程.2004.33(2)：622-625
[42]Xie Kai·Gou Hengye, Tian Wen. A survey of image mosaics. Acta Electronica Sinica[J].2004。32(4):630-635
[43]解凯.郭恒业,田文.图像Mosaics技术综述.电子学报[J],2004,33(2)：630-635
[2]潘立公,汤晓安.一种360度柱面全景图快速生成算法.研究与设计[J],2005,17(4)：19-35
[3]杨枝灵,王开.Visual C++数字幽像获取处理及实践应用[M].北京：人民邮电出版社,2003：100-360
[4]刘翔,曾芬芳.基于IBR技术的全景幽生成.华东船舶工业学院学报(自然科学版)[J].2005,15(1)：32-37
[5]沈建国,戴永刚.基于局部熵差的栅格地图拼接方法[J].华东师范大学学报(自然科学版).2002,(4)：48-53
[6]王昌军,高炉料面区域温度特征智能提取方法研究与应用,中南大学,2010,TF53
[7]陈先中,丁爱华,高炉雷达料面成像系统的设计与实现,冶金自动化[J],2009年3月,第33卷,第二期
[8]王水刚,热像仪在高炉料面温度检测中的应用,宝钢技术,2009年,第5期,FT068.2
[9]张碧梧,高炉料面分布的测量,冶金自动化[J],1985年第05期,
[10]曾秀丽；郝胜男;谭长江,高炉红外/雷达复合探测系统的设计与实现,制造业自动化[J],2010年第07期,
[11]尚明姝,解凯。一种基于特征的全自动图像拼接算法。微计算机应用[J],2006年11月,第2卷第6期
[12]张静,胡志萍,刘志泰,欧宗瑛。基于轮廓相位相关的图像自动拼接。大连理工大学学报[J],2005年1月
[13]崔汉国,张星,刘晓成。一种改进的全景图生成算法。海军工程大学学报[J],第18卷,第2期,2006年4月
[14]Richard Hartley, Aadrew Zisserman. Multiple View Geometry in compuer Vision[Second Edition]. Cambridge:The Press Syndicate The University of Cambridge. UK,2000[M].
[15]SZELISKI R. Video mosaics for virtual environments. IEEE Trans Comput Graphicsand Appl,1996,16(2):22-30[J]。
[16]蔺想红,王维盛。一种基于特征匹配的全景图自动拼接方法。西北师范大学学报(自然科学版),第41卷2005年第4期。[J]
[17]赵向阳,杜利民。一种全自动稳健的图像拼接融合算法。中国图象图形学报,第9卷,第4期,2004年4月[J]。
[18]文雅玫,李思昆。一种遥感图像高精度自动拼接算法。湘南学院学报,2006年10月,第27卷,第5期[J]。
[19]谭磊,薛彦斌。一种基于特征点的图像匹配算法。天津理工大学学报,第22卷,第6期,2006年12月[J]。
[20]何光渝,高永利。Visual Fortran常用数值算法集。科学出版社,2002年4月。[M]
[15]钟力,张茂军,孙立峰,李云浩。360度柱面全景图象生成算法及其实现。小型微型计算机系统,第20卷,第12期,1999年12月[J]。
[16]曹俊杰,苏志勋。全景图像拼接算法。大连理工大学学报,第43卷增刊1,2003年10月[J]。
[17]孙立峰,钟力。虚拟实景空间的实时漫游。中国图象图形学报,第4卷(A版)第6期,1999年6月[J]。
[18]张韵华,奚梅成,陈效群。数值计算方法和算法。科学出版社,2000年1月[M]。
[19]明日科技宋坤Visual C++视频技术方案宝典人民邮电出版社2008年02月[M]。
[20]L Millina, Q Bishop. Plenoptic Modeling:An image based rendering system[A]. SIGGRAPH.1995:39-46
[21]R. Szeliski, H. YShum. Creating full view panoramic image mosaics and Environmentmpas[A]. SIGGR. APH,1997:25 1.257
[22]Hongjie Xie et al.. Automatic image registration based on FFT algorithm and IDL/ENVI[C]. International Conference on Remote Sensing and GIS/ GPS。ICORG.2000,1:391-402
[23]R. Szeliski. Video mosaics for virtual environments[J]. IEEE ComputerGraphics andApplications,1996,16(2):22-30
[24]Thierry Blaszka. Rachid Deriche. Recovering and characterizing image features using an efficient model based approach[c]. Technical Report 2422。 1NRIA. November 1994
[25]Bao P., Xu D. Complex wavelet—based image mosaics using edge preserving visual perception modeling[J]. Computers and Graphics.1999。23(3):309-321
[26]Chen Shizhe, HU Tao, Liu Guodong el al.. Fast and precision image mosaic based on grating[J]. Optics and Precision Engineering.2006.2(14):242-246
[27]陈世哲.胡涛,刘Ⅰ珥栋等.基于光栅的快速精确图像拼接[J].光学精密工程.2006.2(14)：242-246
[28]Fan Q W. Wang X L。Hou L et al.. Speeding mosaic technology based on isometry serial image[J]. Journal of Sichuan University(Engineering science edition).2005.37(Ⅰ):39-142
[29]樊庆文,王小龙.侯力等.基r等距序列}薹Ⅰ像的伙速拼接技术FJ].四川大学学报(工程科学版).2005,1(36)：139-142
[30]Mei X M, Liu Z X, Wang H C et al.. Advanced Geometry (second edition)[M]. Beijing:Higher Education Press.2000
[31]梅向明.刘增贤,王汇淳等.高等几何(第二版)[M].北京：高等教育出版社.2000
[32]Ye Chuanqi, Miao Qiguang. Wang Baoshu. An image fusion algorithm using region segmentation and contourlet transform[J]. Acta Optica Sinica.2008.3(28):447-453
[33]叶传奇.苗启广,王’社树.基于Ⅸ域分割和Counterlet变换的图像融合算法[J].光学学报,2008.3(28)：447-453
[34]Liu C; uixi. I。iu Dongmei。Liu Fengpeng et al.. A Robust image registration algorithm based on feature points matching[J]. Acta Optica Sinica.2008.28(3):454-461
[35]刘贵磐。刘冬梅.刘风鹏等.一种稳健的特征点配准算法[J].光学学报[J]2008.28(3)：454-461
[36]Hu Shejiao, Ge Xiwang, Su I。ing. Panoramic image mosaic based on domain correlation [J]. J. Hejei University of Technology,2007。30(1):29-32
[37]胡社教.葛两旺,苏菱.基于相位相关的伞景图像拼接[J].合肥工业大学学报(自然科学版).2007。30(1)：29-32
[38]Li Wansong, Su Likun。Su Xianyu. Phase-measuring profilometry in big scale measurement. Acta Optica Sinica[J].2000,20(6):792-796
[39]李万松.苏礼坤.苏显渝.相位检测面形术在大尺度三维面形测量中的应用.光学学报[J].2000。20(6)：792-796
[40]Jing(renqiang, Zhang Tianxu, Yang Weidong el al.. Loca adaptive mosaic algorithm for serial images [J]. Infrared andLaser Engineering.2004。33(6): 622-625
[41]荆根强,张灭序.杨1!东等.局部自适应序列图像拼接[J].红外与激光工程.2004.33(2)：622-625
[42]Xie Kai·Gou Hengye, Tian Wen. A survey of image mosaics. Acta Electronica Sinica[J].2004。32(4):630-635
[43]解凯.郭恒业,田文.图像Mosaics技术综述.电子学报[J],2004,33(2)：630-635