基于IZAS图像放大处法的SONY47万像素数字视频展台系统
|
摘要
本文给出了一种数字视频展台的基于IZAS图像放大算法的设计方案。视频展台可将文档、图片、实物等信息转换成图像信号输出在投影机、监视器等显示设备上,广泛应用于教学、会议、产品展示等场合。
传统的模拟展台图像不够清晰、画面抖动、图像质量受环境影响较大。为此,本文研究视频展台的图像处理方法和FPGA实现方案,并给出了基于FPGA的实际应用系统,有效的解决目前存在的问题。
本文的主要内容为:
(1)研究用FPGA实现图像实时放大的处理方法,选择能满足视频展台要求的缩放算法为研究对象,利用给定的放大比例寻找一个简单、易于实现且有较高精度的核函数。对核函数连续模型重新采样得到放大后的像素值。采用滤波器结构进行计算仅使用FPGA内部的3个双端口RAM缓冲图像数据,将原图像有效分辨率为738×575放大到1024×768。
(2)给出一种基于色空变换的色彩调制方法,将图像传感器输出的PAL制数字YUV(4:2:2)格式信号换成RGB(5:6:5)格式。
(3)设计一种针对视频展台的更为简洁的隔行转逐行扫描的实现方案,仅使用一块SDRAM作为帧缓存,利用场间插值算法,并将帧率由25Hz提升到60Hz。
(4)FPGA完成了光学镜头中步进电机的驱动;同时实现了PS/2鼠标的驱动及其显示、图像冻结、负片显示、图像＼文本选择、黑白＼彩色图像言袷涑龅裙δ堋?
上述实现的功能在ISE8.1开发环境下,采用VHDL语言描述,并在Xilinx(赛灵思)公司Spartan3E系列FPGA XC3S250E上实现。
The paper brought forward a new design of digital video presenter on baseof FPGA. The video presenter converts manuscript, picture, object andinformationin in other forms into image signal and outputs to projecter, monitorand other display equipment. Therefore it owns such a large utilization in fieldsof education, conference, product presention and so on. Whileconventioanl Analog video prenster is not satisfied for its unclear and non-stableand glint picture, furthermore its image is remarkbaly subject to the enviroment.In this thesis, the reserch focused on image processing technique and the FPGArealization along with an application system based on FPGA and thus solvingthe problems.
Firstly, the technique of image real-time magnification by FPGA is adopted in this research. The zoom arithmetic which could fulfill the demands of the video presenter is chosen to be the research objects. By using the assured zoom scale search a simple,higher precision core function of easily implementing. Re-sampling to its continuity model can obtain the zoomed pixel value and then the original image will be effectively magnified from 738x575 to 1024x768, just using 3 double port RAM embedded in the FPGA as the image signal buffers.
Secondly, The PAL format YUV(4:2:2) signal from image sensor, a kind of measure of colour space transformation is used in, then is converted into RGB(5:6:5) format.
Thirdly, the paper also carried out a kind of more compact scheme for video presenter which could transform interleved scan into linered scan, just by using one SDRAM as frame buffers, the image signal is de-interlaced with the field interpolation algorithm and the frame rate is speed up to 60Hz from 25Hz.
At the same time, in FPGA, realizes the driving to optic lens'step motor and PS/2 mouse, so the mouse cursor can be displayed on the screen. It also realizes the function of image freezing, negative image display, color/black switching and image/text switching. The whole design is written in VHDL, and it is fulfilled in the Xilinx Spartan3E series FPGA XC3S250E with the developing environment of ISE8.1.
传统的模拟展台图像不够清晰、画面抖动、图像质量受环境影响较大。为此,本文研究视频展台的图像处理方法和FPGA实现方案,并给出了基于FPGA的实际应用系统,有效的解决目前存在的问题。
本文的主要内容为:
(1)研究用FPGA实现图像实时放大的处理方法,选择能满足视频展台要求的缩放算法为研究对象,利用给定的放大比例寻找一个简单、易于实现且有较高精度的核函数。对核函数连续模型重新采样得到放大后的像素值。采用滤波器结构进行计算仅使用FPGA内部的3个双端口RAM缓冲图像数据,将原图像有效分辨率为738×575放大到1024×768。
(2)给出一种基于色空变换的色彩调制方法,将图像传感器输出的PAL制数字YUV(4:2:2)格式信号换成RGB(5:6:5)格式。
(3)设计一种针对视频展台的更为简洁的隔行转逐行扫描的实现方案,仅使用一块SDRAM作为帧缓存,利用场间插值算法,并将帧率由25Hz提升到60Hz。
(4)FPGA完成了光学镜头中步进电机的驱动;同时实现了PS/2鼠标的驱动及其显示、图像冻结、负片显示、图像＼文本选择、黑白＼彩色图像言袷涑龅裙δ堋?
上述实现的功能在ISE8.1开发环境下,采用VHDL语言描述,并在Xilinx(赛灵思)公司Spartan3E系列FPGA XC3S250E上实现。
The paper brought forward a new design of digital video presenter on baseof FPGA. The video presenter converts manuscript, picture, object andinformationin in other forms into image signal and outputs to projecter, monitorand other display equipment. Therefore it owns such a large utilization in fieldsof education, conference, product presention and so on. Whileconventioanl Analog video prenster is not satisfied for its unclear and non-stableand glint picture, furthermore its image is remarkbaly subject to the enviroment.In this thesis, the reserch focused on image processing technique and the FPGArealization along with an application system based on FPGA and thus solvingthe problems.
Firstly, the technique of image real-time magnification by FPGA is adopted in this research. The zoom arithmetic which could fulfill the demands of the video presenter is chosen to be the research objects. By using the assured zoom scale search a simple,higher precision core function of easily implementing. Re-sampling to its continuity model can obtain the zoomed pixel value and then the original image will be effectively magnified from 738x575 to 1024x768, just using 3 double port RAM embedded in the FPGA as the image signal buffers.
Secondly, The PAL format YUV(4:2:2) signal from image sensor, a kind of measure of colour space transformation is used in, then is converted into RGB(5:6:5) format.
Thirdly, the paper also carried out a kind of more compact scheme for video presenter which could transform interleved scan into linered scan, just by using one SDRAM as frame buffers, the image signal is de-interlaced with the field interpolation algorithm and the frame rate is speed up to 60Hz from 25Hz.
At the same time, in FPGA, realizes the driving to optic lens'step motor and PS/2 mouse, so the mouse cursor can be displayed on the screen. It also realizes the function of image freezing, negative image display, color/black switching and image/text switching. The whole design is written in VHDL, and it is fulfilled in the Xilinx Spartan3E series FPGA XC3S250E with the developing environment of ISE8.1.
引文
[1]周杰,刘旭,刘向东.采用SAA7110和AL250集成芯片的TV/VGA转换电路.电子技术与应用,2005,27(4):66-68
[2]周卫,王永华.视频展台采用模拟视频转换器提升图像质量的一种方案.广东自动化与信息工程,2004,25(4):37-39
[3]徐欣,于红旗,易凡等.基于的嵌入式系统设计.北京:机械工业出版社,2005
[4]Sony.Digital signal processor for color CCD cameral LR38603A specifications [EB/OL].www.sharp.com,2003
[5]曾繁泰,王强,盛娜.EDA工程的理论与实践-SOC系统芯片的设计方法.北京:电子工业出版社,2004
[6]Bob Zeidman著,赵宏图译.基于FPGA&CPLD的数字IC设计方法.北京:北京航空航天大学出版社,2004
[7]王诚,薛小刚,钟信潮.FPGA/CPLD设计工具-Xilinx ISE5.X使用详解.北京:人民邮电出版社,2003
[8]XILINX.Spartan-3E FPGA Family:Complete Data Sheet[EB/OL].www.xilinx.com,2006
[9]Hynix.HY57V281620HCST datasheet[EB/OL].www.hynix.com,2002
[10]Betas R S,Alman D H,Reniff L.Visual determination of supra threshold color difference tolerance using probit analysis.Color Research and Application,1991,16(5):297-316
[11]刘健,吴玉广.用FPGA实现色空间R′G′B′到YCbCr的转换.现代显示,2004,(1):56-58
[12]Zhou Chang-guo,Kohu Leslie,Rice Daniel,etal.M.PEG video decoding with the ultraSPARC visusl instruction set[J].IEEE Computer Society International Conference,1995,P470-475
[13]卢毅,赖杰.VI-IDL与数字电路设计.科学出版社,2002
[14]孙航.Xilinx可编程逻辑器件的高级应用与设计技巧.北京:电子工业出版社, 2004
[15]俞斯乐,侯正信,冯启明等.电视原理.北京:国防工业出版社,2000
[16]赵坚勇.电视原理与系统.西安:西安电子科技大学出版社,2005
[17]裴昌幸,刘乃安.电视原理与现代电视系统.西安:西安电子科技大学出版社,1997
[18]盛磊.面向特种LCD图像处理方法与FPGA实现研究[博士学位论文].合肥:合肥工业大学,2005
[19]Soon-Kak Kwon,Kang-Soo Seo,Jae-Kyoon Kim.A motion-adaptive de-interlacing method,Consumer Electronics.IEEE Transaction on,Volume 38,Issue 3,Aug 1992 Page(145-150)
[20]Sugiyama K,Nakamura H.A method of de-interlacing with motion compensated interpolation.Consumer Electronis,1999,ICCE.International Conference on,22-24 June 1999 Page:210-211
[21]张倩.视频显示格式转换电路的设计与验证[硕士学位论文].北京:清华大学,2004
[22]杭平华.数字电视自适应内插电路的设计与实现.南京邮电学院学报,1996,16(4):29-33
[23]周昆正.基于FPGA的SDRAM控制器的设计[J].现代电子技术,2003(13):63-65
[24]段然,樊哓桠,张盛兵等.基于状态机的SDRAM控制器的设计与实现.计算机工程与应用,2005(17):110-112
[25]付树军,阮秋琦,王文洽.一种基于双曲正切函数的自适应图像放大算法.计算机工程与应用,2005(3):12-13
[26]付树军,阮秋琦,王文洽.基于变分的保持特征的有噪图像放大算法.计算机工程与应用,2005(14):9-11
[27]Micheal Unser,Akram Aldroubi,Murray Eden.Enlargement or Reduction of Digital Images with Minimum Loss of Information[J].IEEE Transaction on Image Processing,1995,4(5):247-258
[28]Kenneth R Castleman.Digital image processing[M].Englewood Cliffs,NJ:Prentice-Hall International Inc,1996
[29]Jechang Jeong,Byeungwoo Jeon.A Multiplierless Letter-Box Converter for Displaying 16:9 Images on a 4:3 Screen.IEEE Transactions on Circuits and Systems for Video Technology,1995,5(4):363-366
[30]贾中云.一种利用采样率转换方法的灰度图像放大算法.杭州师范学院学报(自然科学版),2005,14(5):360-363
[31]马天骏.图像放大技术研究[硕士学位论文].西安:西安电子科技大学,2004
[32]张晓玲,沈兰荪,Lam Kin-Man.一种基于分形码和模型约束的图像放大算法.电子学报,2006(3):433-436
[33]R R Schultz,R L Stevenson.A bayesian approach to image expansion for improved definition[J].IEEE Trans Image Processing,1994,3(3):233-242
[34]S K Mitra,C A Murthy,M K Kundu.A technique for image magnification using partitioned iterative function system[J].Pattern recognition,2000,33(7):1119-1133
[35]吴良武,欧宗瑛.保持轮廓清晰光滑的灰度图像放大算法[J].计算机辅助设计与图形学学报,2002,14(4):306-309
[36]李将云,杨勋年,汪国昭.图像缩放的分片连续算法[J].浙江大学学报(理学版),2002,29(5):530-534
[37]倪伟,郭宝龙.一种基于分形和小波变换的图像放大算法.仪器仪表学报,2004,25(4):352-354
[38]朱长青,王倩,陈虹等.基于多进制小波变换的图像放大算法.中国图像图形学报,2002,7(3):261-266
[39]郑新景,蒋洪.PS/2键盘鼠标协议.电子科技,2001(17):26-28
[40]刘迎春,牟金秀,毕宏彦.一种简单串行鼠标控制的单片机实现.单片机与嵌入式系统应用,2003,(10):61-63
[41]白立芬,徐毓娴,于水等.基于图像处理的显微镜自动调焦方法研究[J].仪器仪表学报,1999,12(6):612-614
[42]刘义鹏.显微镜自动对焦系统设计与应用[硕士学位论文].杭州:浙江大学,2005
[43]冯华君.反射能量法测聚焦系统及其重叠设计法.光电工程,1998,25(2):48-53
[44]李奇,冯华君.数字图像清晰度评价函数研究.光子学报,2002,31(6):736-738
[2]周卫,王永华.视频展台采用模拟视频转换器提升图像质量的一种方案.广东自动化与信息工程,2004,25(4):37-39
[3]徐欣,于红旗,易凡等.基于的嵌入式系统设计.北京:机械工业出版社,2005
[4]Sony.Digital signal processor for color CCD cameral LR38603A specifications [EB/OL].www.sharp.com,2003
[5]曾繁泰,王强,盛娜.EDA工程的理论与实践-SOC系统芯片的设计方法.北京:电子工业出版社,2004
[6]Bob Zeidman著,赵宏图译.基于FPGA&CPLD的数字IC设计方法.北京:北京航空航天大学出版社,2004
[7]王诚,薛小刚,钟信潮.FPGA/CPLD设计工具-Xilinx ISE5.X使用详解.北京:人民邮电出版社,2003
[8]XILINX.Spartan-3E FPGA Family:Complete Data Sheet[EB/OL].www.xilinx.com,2006
[9]Hynix.HY57V281620HCST datasheet[EB/OL].www.hynix.com,2002
[10]Betas R S,Alman D H,Reniff L.Visual determination of supra threshold color difference tolerance using probit analysis.Color Research and Application,1991,16(5):297-316
[11]刘健,吴玉广.用FPGA实现色空间R′G′B′到YCbCr的转换.现代显示,2004,(1):56-58
[12]Zhou Chang-guo,Kohu Leslie,Rice Daniel,etal.M.PEG video decoding with the ultraSPARC visusl instruction set[J].IEEE Computer Society International Conference,1995,P470-475
[13]卢毅,赖杰.VI-IDL与数字电路设计.科学出版社,2002
[14]孙航.Xilinx可编程逻辑器件的高级应用与设计技巧.北京:电子工业出版社, 2004
[15]俞斯乐,侯正信,冯启明等.电视原理.北京:国防工业出版社,2000
[16]赵坚勇.电视原理与系统.西安:西安电子科技大学出版社,2005
[17]裴昌幸,刘乃安.电视原理与现代电视系统.西安:西安电子科技大学出版社,1997
[18]盛磊.面向特种LCD图像处理方法与FPGA实现研究[博士学位论文].合肥:合肥工业大学,2005
[19]Soon-Kak Kwon,Kang-Soo Seo,Jae-Kyoon Kim.A motion-adaptive de-interlacing method,Consumer Electronics.IEEE Transaction on,Volume 38,Issue 3,Aug 1992 Page(145-150)
[20]Sugiyama K,Nakamura H.A method of de-interlacing with motion compensated interpolation.Consumer Electronis,1999,ICCE.International Conference on,22-24 June 1999 Page:210-211
[21]张倩.视频显示格式转换电路的设计与验证[硕士学位论文].北京:清华大学,2004
[22]杭平华.数字电视自适应内插电路的设计与实现.南京邮电学院学报,1996,16(4):29-33
[23]周昆正.基于FPGA的SDRAM控制器的设计[J].现代电子技术,2003(13):63-65
[24]段然,樊哓桠,张盛兵等.基于状态机的SDRAM控制器的设计与实现.计算机工程与应用,2005(17):110-112
[25]付树军,阮秋琦,王文洽.一种基于双曲正切函数的自适应图像放大算法.计算机工程与应用,2005(3):12-13
[26]付树军,阮秋琦,王文洽.基于变分的保持特征的有噪图像放大算法.计算机工程与应用,2005(14):9-11
[27]Micheal Unser,Akram Aldroubi,Murray Eden.Enlargement or Reduction of Digital Images with Minimum Loss of Information[J].IEEE Transaction on Image Processing,1995,4(5):247-258
[28]Kenneth R Castleman.Digital image processing[M].Englewood Cliffs,NJ:Prentice-Hall International Inc,1996
[29]Jechang Jeong,Byeungwoo Jeon.A Multiplierless Letter-Box Converter for Displaying 16:9 Images on a 4:3 Screen.IEEE Transactions on Circuits and Systems for Video Technology,1995,5(4):363-366
[30]贾中云.一种利用采样率转换方法的灰度图像放大算法.杭州师范学院学报(自然科学版),2005,14(5):360-363
[31]马天骏.图像放大技术研究[硕士学位论文].西安:西安电子科技大学,2004
[32]张晓玲,沈兰荪,Lam Kin-Man.一种基于分形码和模型约束的图像放大算法.电子学报,2006(3):433-436
[33]R R Schultz,R L Stevenson.A bayesian approach to image expansion for improved definition[J].IEEE Trans Image Processing,1994,3(3):233-242
[34]S K Mitra,C A Murthy,M K Kundu.A technique for image magnification using partitioned iterative function system[J].Pattern recognition,2000,33(7):1119-1133
[35]吴良武,欧宗瑛.保持轮廓清晰光滑的灰度图像放大算法[J].计算机辅助设计与图形学学报,2002,14(4):306-309
[36]李将云,杨勋年,汪国昭.图像缩放的分片连续算法[J].浙江大学学报(理学版),2002,29(5):530-534
[37]倪伟,郭宝龙.一种基于分形和小波变换的图像放大算法.仪器仪表学报,2004,25(4):352-354
[38]朱长青,王倩,陈虹等.基于多进制小波变换的图像放大算法.中国图像图形学报,2002,7(3):261-266
[39]郑新景,蒋洪.PS/2键盘鼠标协议.电子科技,2001(17):26-28
[40]刘迎春,牟金秀,毕宏彦.一种简单串行鼠标控制的单片机实现.单片机与嵌入式系统应用,2003,(10):61-63
[41]白立芬,徐毓娴,于水等.基于图像处理的显微镜自动调焦方法研究[J].仪器仪表学报,1999,12(6):612-614
[42]刘义鹏.显微镜自动对焦系统设计与应用[硕士学位论文].杭州:浙江大学,2005
[43]冯华君.反射能量法测聚焦系统及其重叠设计法.光电工程,1998,25(2):48-53
[44]李奇,冯华君.数字图像清晰度评价函数研究.光子学报,2002,31(6):736-738