一种适用于航空影像的无参考模糊探测方法
|
摘要
针对侧风、强风、湍流等飞行环境容易造成航空影像运动模糊,严重影响航空影像质量,同时航空影像数据量大,手动挑选模糊影像费时费力的问题,为了提高航摄内业人员的工作效率,该文研究一种适用于航空影像的自动模糊探测方法,以主流的无参考再模糊算法Reblur和无参考结构清晰度算法NRSS为基础,结合航空影像具有丰富地物的特点,对影像进行分块处理,计算所有字块的Reblur和NRSS模糊探测值,最后得到整幅影像的模糊探测值。其中,再模糊算法通过计算待测影像和参考影像的水平和垂直运动方向上的灰度变化来评价图像模糊度;NRSS算法在结构相似度SSIM算法基础上加入梯度信息提取和高斯滤波等改进,通过计算结构相似度评价图像模糊度。实验结果表明,该文研究的无参考模糊影像探测方法适用于航空影像数据,其评价结果与人眼主观评价结果具有较高的一致性,能够准确地缩小模糊影像的查找范围,极大地提高了航摄内业效率。
Aiming at the problem that cross-wind,strong wind,turbulent and other flight environments are likely to cause motion blur in aerial images,which seriously affects aerial image quality.Due to the large amount of aerial image data,manual selection of blurred images is a time-consuming and laborious task.In order to improve the working efficiency of aerial photogrammetry,this paper studies an automatic blur detection method suitable for aerial images.The methods is based on the mainstream Reblur and non-Reference Structural Sharpness(NRSS)algorithm,combined with the characteristics of aerial imagery,making the image segmented,and then calculating the Reblur and NRSS detection values of all blocks.Finally,the blur detection value of the entire image is obtained.The Reblur algorithm evaluates the blurred image by calculating the gradation changes in the horizontal and vertical directions of the image to be tested and the reference image.The NRSS algorithm adds gradient information extraction and Gaussian filter to the structural similarity(SSIM)algorithm.The blurred image is evaluated by calculating the structural similarity.The experimental results show that the non-reference blur detection method studied in this paper is suitable for aerial image data,and its evaluation results are highly consistent with the subjective evaluation results of human eyes,which can accurately narrow the search range of blurred images and greatly improve the efficiency of aerial photogrammetry.
Aiming at the problem that cross-wind,strong wind,turbulent and other flight environments are likely to cause motion blur in aerial images,which seriously affects aerial image quality.Due to the large amount of aerial image data,manual selection of blurred images is a time-consuming and laborious task.In order to improve the working efficiency of aerial photogrammetry,this paper studies an automatic blur detection method suitable for aerial images.The methods is based on the mainstream Reblur and non-Reference Structural Sharpness(NRSS)algorithm,combined with the characteristics of aerial imagery,making the image segmented,and then calculating the Reblur and NRSS detection values of all blocks.Finally,the blur detection value of the entire image is obtained.The Reblur algorithm evaluates the blurred image by calculating the gradation changes in the horizontal and vertical directions of the image to be tested and the reference image.The NRSS algorithm adds gradient information extraction and Gaussian filter to the structural similarity(SSIM)algorithm.The blurred image is evaluated by calculating the structural similarity.The experimental results show that the non-reference blur detection method studied in this paper is suitable for aerial image data,and its evaluation results are highly consistent with the subjective evaluation results of human eyes,which can accurately narrow the search range of blurred images and greatly improve the efficiency of aerial photogrammetry.
引文
[1]BOON T K,HAIDI I.Exploration of current trend on blur detection method utilized in digital image processing[J].Journal of Industrial and Information,2013,1(3):143-147.
[2]FRDRIQUE CRT-ROFFET,DOLMIERE T,LADRET P,et al.The blur effect:perception and estimation with a new No-reference perceptual blur metric[J].Proceedings of SPIE-The International Society for Optical Engineering,2007,12:64920I-64920I-11.
[3]ZHOU WANG,ALAN C BOVIK,ET AL.Image quality assessment:from error visibility to structural similarity[J].IEEE Transactions on Image Processing,2004,13(4):600-612.
[4]谢小甫,周进,吴钦章.一种针对图像模糊的无参考质量评价指标[J].计算机应用,2010,30(4):921-925.(XIE Xiaopu,ZHOU Jin,WU Qinzhang.A non-reference quality evaluation index for image blur[J].Journal of Computer Applications,2010,30(4):921-925.)
[5]杨春玲,陈冠豪,谢胜利.基于梯度信息的图像质量评价方法的研究[J].电子学报,2007,35(7):1313-1317.(YANG Chunling,CHEN Guanhao,XIE Shengli.Study on image quality evaluation method based on gradient information[J].Acta Electronica Sinica,2007,35(7):1313-1317.)
[6]付强,孙秀霞,彭轲,等.一种无参考运动模糊图像的质量评价方法[J].激光与红外,2015,45(6):710-714.(FU Qiang,SUN Xiuxia,PENG Ke,et al.A quality evaluation method for unreferenced motion blurred images[J].Laser&Infrared,2015,45(6):710-714.)
[7]SIEBERTH T,WACKROW R,CHANDLER J H.Automatic isolation of blurred images from UAVimage sequences[J].ISPRS-International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2013,XL-1/W2(1):361-366.
[8]LIU H,WANG J,REDI J,et al.An efficient no-reference metric for perceived blur[C]∥European Workshop on Visual Information Processing.[S.l.]:IEEE,2011.
[9]TEO T A,ZHAN K Z.Integration of image-derived and pos-derived features for image blur detection[J].ISPRS-International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2016,XLI-B1:1051-1055.
[10]MARIA JOO M.VASCONCELOS,LUS ROSA-DO.No-reference blur assessment of dermatological images acquired via mobile devices[J].Lecture Notes in Computer Science,2014,8509:350-357.
[11]张满.无参考图像质量评价及其应用[D]:西安:西安电子科技大学,2018:2-7.(ZHANG Man.No-reference image quality assessment and the application[D]:Xi’an:Xidian University,2018:2-7.)
[12]张偌雅,李珍珍.数字图像质量评价综述[J].现代计算机,2017,29(19):1007-1013.(ZHANG Nuoya,LIZhenzhen.Summary of digital image quality evaluation[J].Modern Computer,2017,29(19):1007-1013.)
[13]GAO X,WEN L,TAO D,et al.Image quality assessment based on multiscale geometric analysis[J].IEEE Transactions on Image Processing,2009,18(7):1409-1423.
[14]SAZZAD Z M P,KAWAYOKE Y,HORITA Y.No reference image quality assessment for JPEG2000based on spatial features[J].Signal Processing Image Communication,2008,23(4):257-268.
[2]FRDRIQUE CRT-ROFFET,DOLMIERE T,LADRET P,et al.The blur effect:perception and estimation with a new No-reference perceptual blur metric[J].Proceedings of SPIE-The International Society for Optical Engineering,2007,12:64920I-64920I-11.
[3]ZHOU WANG,ALAN C BOVIK,ET AL.Image quality assessment:from error visibility to structural similarity[J].IEEE Transactions on Image Processing,2004,13(4):600-612.
[4]谢小甫,周进,吴钦章.一种针对图像模糊的无参考质量评价指标[J].计算机应用,2010,30(4):921-925.(XIE Xiaopu,ZHOU Jin,WU Qinzhang.A non-reference quality evaluation index for image blur[J].Journal of Computer Applications,2010,30(4):921-925.)
[5]杨春玲,陈冠豪,谢胜利.基于梯度信息的图像质量评价方法的研究[J].电子学报,2007,35(7):1313-1317.(YANG Chunling,CHEN Guanhao,XIE Shengli.Study on image quality evaluation method based on gradient information[J].Acta Electronica Sinica,2007,35(7):1313-1317.)
[6]付强,孙秀霞,彭轲,等.一种无参考运动模糊图像的质量评价方法[J].激光与红外,2015,45(6):710-714.(FU Qiang,SUN Xiuxia,PENG Ke,et al.A quality evaluation method for unreferenced motion blurred images[J].Laser&Infrared,2015,45(6):710-714.)
[7]SIEBERTH T,WACKROW R,CHANDLER J H.Automatic isolation of blurred images from UAVimage sequences[J].ISPRS-International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2013,XL-1/W2(1):361-366.
[8]LIU H,WANG J,REDI J,et al.An efficient no-reference metric for perceived blur[C]∥European Workshop on Visual Information Processing.[S.l.]:IEEE,2011.
[9]TEO T A,ZHAN K Z.Integration of image-derived and pos-derived features for image blur detection[J].ISPRS-International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2016,XLI-B1:1051-1055.
[10]MARIA JOO M.VASCONCELOS,LUS ROSA-DO.No-reference blur assessment of dermatological images acquired via mobile devices[J].Lecture Notes in Computer Science,2014,8509:350-357.
[11]张满.无参考图像质量评价及其应用[D]:西安:西安电子科技大学,2018:2-7.(ZHANG Man.No-reference image quality assessment and the application[D]:Xi’an:Xidian University,2018:2-7.)
[12]张偌雅,李珍珍.数字图像质量评价综述[J].现代计算机,2017,29(19):1007-1013.(ZHANG Nuoya,LIZhenzhen.Summary of digital image quality evaluation[J].Modern Computer,2017,29(19):1007-1013.)
[13]GAO X,WEN L,TAO D,et al.Image quality assessment based on multiscale geometric analysis[J].IEEE Transactions on Image Processing,2009,18(7):1409-1423.
[14]SAZZAD Z M P,KAWAYOKE Y,HORITA Y.No reference image quality assessment for JPEG2000based on spatial features[J].Signal Processing Image Communication,2008,23(4):257-268.