水稻表层根系图像分割算法研究
|
摘要
水稻根系形态特征的定量研究对于改进农田管理方式、水稻品种选育和遗传改良等具有重要意义。近年来,随着表型组技术迅速发展,利用图像处理技术对水稻根系生长情况进行测量和分析,同时配合施肥、灌溉、光照、温控等环境监控技术已成为水稻育种和功能基因组研究新型技术手段,而根系图像分割技术是进行后续表型组学分析的重要基础之一。由于生长在土壤中的水稻根系图像具有对比度低、信噪比低、纹理复杂的特点,分割十分困难。针对此问题,研究了主干-分支连接算法、基于形态特征的局部阈值分割算法和基于形态特征的自适应阈值分割算法,对生长在土壤中水稻根系图像进行分割处理和比较。实验结果表明,主干-分支连接算法虽然保留了大量细节,但是受噪声影响严重,其结构略显杂乱,毛刺现象严重;基于形态特征的局部阈值分割算法能保留更多根部的细节,但轮廓断裂的现象比较严重;自适应阈值分割算法分割的图像根系连续性较好,毛刺现象也得到了抑制,但是细小的须根无法保留。最终将两种算法结合起来,提出一种适用于水稻表层根系图像分割的综合算法,则可以获得较为理想的分割结果,为后续水稻根系性状提取奠定了重要基础。
The quantitative studies on morphological characteristics of rice root system are of great significance for improving farmland management,rice selective breeding and genetic improvement. In recent years,along with the rapid development of phenotype technology,image processing technology could be used to measure and analyze the growth of rice root system. At the same time,combined with fertilization,irrigation,light,temperature control and environmental monitoring technology,it has become a new technological means for rice breeding and functional genomics research. And the root system image segmentation technology is one of the key foundations for the subsequent phenotypic group analysis. The segmentation is very difficult,because the rice root system images have the characteristics of low contrast,low SNR and complex texture. Aiming at this problem,the paper firstly studied the main-branch connection algorithm,local threshold segmentation algorithm based on the morphological characteristics,and adaptive threshold segmentation based on morphological characteristics,and then performed segmentation and comparison on rice root system images. The results showed that although the main-branch connection algorithm retained a large number of root details,its structure was slightly chaotic and burr phenomenon was serious; the local threshold segmentation algorithm based on morphological features could retain more root details,but the contour fracture was more severe; adaptive threshold segmentation algorithm could improve image continuity, burrphenomenon was controlled,fine roots would not be preserved. These 2 algorithms were combined into a integrated algorithm suitable for image segmentation of rice surface root system. Thus,relatively ideal segmentation results could be obtained,which would provide important basis for follow-up studies on phenotypic traits of rice root system.
The quantitative studies on morphological characteristics of rice root system are of great significance for improving farmland management,rice selective breeding and genetic improvement. In recent years,along with the rapid development of phenotype technology,image processing technology could be used to measure and analyze the growth of rice root system. At the same time,combined with fertilization,irrigation,light,temperature control and environmental monitoring technology,it has become a new technological means for rice breeding and functional genomics research. And the root system image segmentation technology is one of the key foundations for the subsequent phenotypic group analysis. The segmentation is very difficult,because the rice root system images have the characteristics of low contrast,low SNR and complex texture. Aiming at this problem,the paper firstly studied the main-branch connection algorithm,local threshold segmentation algorithm based on the morphological characteristics,and adaptive threshold segmentation based on morphological characteristics,and then performed segmentation and comparison on rice root system images. The results showed that although the main-branch connection algorithm retained a large number of root details,its structure was slightly chaotic and burr phenomenon was serious; the local threshold segmentation algorithm based on morphological features could retain more root details,but the contour fracture was more severe; adaptive threshold segmentation algorithm could improve image continuity, burrphenomenon was controlled,fine roots would not be preserved. These 2 algorithms were combined into a integrated algorithm suitable for image segmentation of rice surface root system. Thus,relatively ideal segmentation results could be obtained,which would provide important basis for follow-up studies on phenotypic traits of rice root system.
引文
[1]吴迪,王梦坷,杨万能,等.水稻表层根系性状无损测量技术研究[J].中国农业科技导报,2017,19(7):66-77.Wu D,Wang M K,Yang W N,et al.. Research on nondestructive measurement of surface root traits of rice plants[J].J. Agric. Sci. Technol.,2017,19(7):66-77.
[2]郭海宇.植物根系图像的特征分析方法研究与实现[D].成都:电子科技大学,硕士学位论文,2013.Guo H Y. Research and implementation of characteristic analysis method using plant root images[D]. Chengdu:University of Electronic Science and Technology of China,Master Dissertation,2013.
[3]王会民,唐秀英,龙启樟,等.水稻根系遗传特征研究进展[J].分子植物育种,2016,14(4),910-917.Wang H M,Tang X Y,Long Q Z,et al.. Research progress on genetic characteristics of rice root[J]. Mol. Plant Breed.,2016,14(4),910-917.
[4]杨乐,唐建军,何火娇.水稻根系空间分布特性的三维建模及可视化研究[J].江西农业大学学报,2016,38(3):588-592.Yang L,Tang J J,He H J. Three dimensional modeling and visualization of spatial distribution of rice roots[J]. Acta Agric.Univ. Jiangxiensis,2016,38(3):588-592.
[5]徐其军,汤亮,顾东祥,等.基于形态参数的水稻根系三维建模及可视化[J].农业工程学报,2010,26(10):188-194.Xu Q J,Tang L,Gu D X,et al.. Architectural parameterbased three dimensional modeling and visualization of rice roots[J]. Trans. CSAE,2010,26(10):188-194.
[6]顾东祥,汤亮,曹卫星,等.基于图像分析方法的水稻根系形态特征指标的定量分析[J].作物学报,2010,36(5):810-817.Gu D X,Tang L,Cao Q X,et al.. Quantitative analysis on root morphological characteristics based on image analysis method in rice[J]. Acta Agron,Sin,,2010,36(5):810-817.
[7]梁永书,周军杰,南文斌,等.水稻根系研究进展[J].植物学报,2016,51(1):98-106.Liang Y S,Zhou J J,Nan W B,et al.. Progress in rice root system research[J]. Chin. Bull. Bot.,2016,51(1):98-106.
[8] Bates G H. A device for the observation of root growth in the soil[J]. Nature,1937,139(3527):966-967.
[9] Mairhofer S, Zappala S, Tracy S R, et al.. Roo Trak:Automated recovery of three-dimensional plant root architecture in soil from X-ray microcomputed tomography images using visual tracking[J]. Plant Physiol.,2012,158(2):561-569.
[10] Schulz H,Postma J A,van Dusschoten D,et al.. Plant root system analysis from MRI images,computer vision,imaging and computer graphics[J]. Theory Appl.,2013,359:411-425.
[11]冯太锐,苗玉彬,朱云开,等.叶菜根系图像的变分水平集分割算法[J].东华大学学报(自然科学版),2017,43(4):552-558,570.Feng T R,Miao Y B,Zhu Y K,et al.. Segmentation of leafy vegetable root images based on variational level set method[J].J. Donghua Univ.(Nat. Sci.),22017,43(4):552-558,570.
[12]惠放,马韫韬,朱晋宇,等.利用多视角图像法分析番茄幼苗根构型对氮水平的响应[J].植物营养与肥料学报,2016,22(5):1418-1424.Hui F,Ma W T,Zhu J Y,et al.. Analysis of nitrogen effects on root system architecture of tomato seedlings using the multiview images method[J]. J. Plant Nutr. Fertil.,2016,22(5):1418-1424.
[13]苗玉彬,翟鹏.基于改进FCM的葡萄根系图像分割算法[J].中国农机化学报,2014,35(6):69-71.Miao Y B,Zhai P. Grape root image segmentation algorithm based on improved FCM[J]. J. Chin. Agric. Mechanization,2014,35(6):69-71.
[14]贾庆宇,刘晶淼,谢艳兵.基于Matlab的玉米根系图像拼接、根系提取分析系统[J].中国农学通报,2014,30(3):98-102.Jia Q Y,Liu J M,Xie Y B. The corn roots image stitehing,extraetion and analysis system based on matlab software[J].Chin. Agric. Sci. Bull.,2014,30(3):98-102.
[15]刘柯楠,杨启良,郭浩,等.基于不同滴灌方式的苹果幼树根系生长的三维可视化模拟[J].中国生态农业学报,2013,21(6):744-751.Liu K N,Yang Q L,Guo H,et al.. Modeling and visualization of three-dimensional young apple tree root growth under different modes of drip irrigation[J]. Chin. J. Eco-Agric.,2013,21(6):744-751.
[16]肖潇,戈振扬,于英杰,等.层切压实土壤中绿豆根系的图像处理[A].见:中国农业工程学会2011年学术年会论文集[C].重庆,2011.Xiao X,Ge Z Y,Yu Y J,et al.. Image processing of layer cutting of compacted root-soil of mung bean[A]. In:Proceedings of 2011 Academic Annual Meeting of Chinese Society of Agricultural Engineering[C]. Chongqing,2011.
[17] Iyer-Pascuzzi A S,Symonova O,Mileyko Y,et al.. Imaging and analysis platform for automatic phenotyping and trait ranking of plant root systems[J]. Plant Physiol.,2010,152(3):1148-1157.
[18] Clark R T,Mac Curdy R B,Jung J K,et al.. Three-dimensional root phenotyping with a novel imaging and software platform[J]. Plant Physiol.,2011,156(2):455-465.
[19]梅川,吴桂良,杨媛,等.一种基于区域增长和结构识别的心血管X线造影图像分割方法[J].生物医学工程学杂志,2014,31(2):413-420.Mei C,Wu G L,Yang Y,et al.. An approach for segentation of x-ray angiographic image based on region-growing and structure inferring[J]. J. Biomed. Engin.,2014,31(2):413-420.
[20] Frangi A F,Niessen W J,Vincken K L,et al.. Multiscale vessel enhancement filtering[A]. In:Proceeding of Medical Image Computing and Computer-Assisted Interventation MICCAI'98 Lecture Notes in Computer Science[C]. MA:s.n.,1998,130-137.
[21]张缓缓,李仁忠,景军锋,等. Frangi滤波器和模糊C均值算法相结合的织物瑕疵检测[J].纺织学报,2015,36(9):120-124.Zhang Y Y,Li R Z,Jing J F,et al.. Fabric defect detection based on frangi filter and fuzzy c-means algorithm in combination[J]. J. Textile Res.,2015,36(9):120-124.
[22]夏振华,石玉,于盛林.基于Gabor滤波器的指纹图像增强[J].工程图学学报,2006,25(5):80-85.Xia Z H,Shi Y,Yu C Y. Fingerprint image enhancement based on gabor filters[J]. J. Engin. Graphics,2006,25(5):80-85.
[23]廖开阳,张学东,章明珠,等.基于Gabor滤波器的指纹图像快速增强[J].计算机工程与应用,2009,45(10):172-175,198.Liao C Y,Zhang X D,Zhang M Z,et al.. Fast fingerprint enhancement based on gabor filter[J]. Computer Engin.Appl.,2009,45(10):172-175,198.
[24] Ostu N,Nobuyuki O,Otsu N. A threshold selection method from gray-level histogram IEEE transactions on systems[J].IEEE Trans. Syst. Man Cybernetics,1979,9(1):62-66.
[25]李旭,赵文杰,杨凯达.基于小目标预提取的OTSU分割方法[J].红外技术,2013,35(8):492-496.Li X, Zhao W J, Yang K D. OTSU applied in image segmentation based on small targets pre-detection[J]. Infrared Technol.,2013,35(8):492-496.
[26]黄颖,许可.融合Ncut方法的脊椎图像OTSU分割算法[J].计算机工程与应用,2010,46(25):188-190.Huang Y,Xu K. Spine OTSU segmentation algorithm based on Ncut method[J]. Computer Engin. Appl.,2010,46(25):188-190.
[27] He K,Sun J,Tang X. Guided image filtering[J]. IEEE Trans. Pattern Anal. Machine Intell.,2013,35(6):1397-1409.
[28] Bradley D,Roth G. Adapting thresholding using the integral image[J]. J. Graphics Tools,2007,12(2):13-21.
[29] Powers D M W. Evaluation:From precision,recall and Ffactor to ROC,informedness,markedness Correlation[J]. J.Machine Learn. Technol.,2011,2:2229-3981.
[30] Xiong X,Duan L F,Liu L B,et al.. Panicle-SEG:A robust image segmentation method for rice panicles in the field based on deep learning and superpixel optimization[J]. Plant Methods. 2017 13 13(1):104.
[31] Zhang X L,Feng X Z,Xiao P F,et al.. Segmentation quality evaluation using region-based precision and recall measures for remote sensing images[J]. Isprs J. Photogrammetry Remote Sensing,2015,102:73-84.
[2]郭海宇.植物根系图像的特征分析方法研究与实现[D].成都:电子科技大学,硕士学位论文,2013.Guo H Y. Research and implementation of characteristic analysis method using plant root images[D]. Chengdu:University of Electronic Science and Technology of China,Master Dissertation,2013.
[3]王会民,唐秀英,龙启樟,等.水稻根系遗传特征研究进展[J].分子植物育种,2016,14(4),910-917.Wang H M,Tang X Y,Long Q Z,et al.. Research progress on genetic characteristics of rice root[J]. Mol. Plant Breed.,2016,14(4),910-917.
[4]杨乐,唐建军,何火娇.水稻根系空间分布特性的三维建模及可视化研究[J].江西农业大学学报,2016,38(3):588-592.Yang L,Tang J J,He H J. Three dimensional modeling and visualization of spatial distribution of rice roots[J]. Acta Agric.Univ. Jiangxiensis,2016,38(3):588-592.
[5]徐其军,汤亮,顾东祥,等.基于形态参数的水稻根系三维建模及可视化[J].农业工程学报,2010,26(10):188-194.Xu Q J,Tang L,Gu D X,et al.. Architectural parameterbased three dimensional modeling and visualization of rice roots[J]. Trans. CSAE,2010,26(10):188-194.
[6]顾东祥,汤亮,曹卫星,等.基于图像分析方法的水稻根系形态特征指标的定量分析[J].作物学报,2010,36(5):810-817.Gu D X,Tang L,Cao Q X,et al.. Quantitative analysis on root morphological characteristics based on image analysis method in rice[J]. Acta Agron,Sin,,2010,36(5):810-817.
[7]梁永书,周军杰,南文斌,等.水稻根系研究进展[J].植物学报,2016,51(1):98-106.Liang Y S,Zhou J J,Nan W B,et al.. Progress in rice root system research[J]. Chin. Bull. Bot.,2016,51(1):98-106.
[8] Bates G H. A device for the observation of root growth in the soil[J]. Nature,1937,139(3527):966-967.
[9] Mairhofer S, Zappala S, Tracy S R, et al.. Roo Trak:Automated recovery of three-dimensional plant root architecture in soil from X-ray microcomputed tomography images using visual tracking[J]. Plant Physiol.,2012,158(2):561-569.
[10] Schulz H,Postma J A,van Dusschoten D,et al.. Plant root system analysis from MRI images,computer vision,imaging and computer graphics[J]. Theory Appl.,2013,359:411-425.
[11]冯太锐,苗玉彬,朱云开,等.叶菜根系图像的变分水平集分割算法[J].东华大学学报(自然科学版),2017,43(4):552-558,570.Feng T R,Miao Y B,Zhu Y K,et al.. Segmentation of leafy vegetable root images based on variational level set method[J].J. Donghua Univ.(Nat. Sci.),22017,43(4):552-558,570.
[12]惠放,马韫韬,朱晋宇,等.利用多视角图像法分析番茄幼苗根构型对氮水平的响应[J].植物营养与肥料学报,2016,22(5):1418-1424.Hui F,Ma W T,Zhu J Y,et al.. Analysis of nitrogen effects on root system architecture of tomato seedlings using the multiview images method[J]. J. Plant Nutr. Fertil.,2016,22(5):1418-1424.
[13]苗玉彬,翟鹏.基于改进FCM的葡萄根系图像分割算法[J].中国农机化学报,2014,35(6):69-71.Miao Y B,Zhai P. Grape root image segmentation algorithm based on improved FCM[J]. J. Chin. Agric. Mechanization,2014,35(6):69-71.
[14]贾庆宇,刘晶淼,谢艳兵.基于Matlab的玉米根系图像拼接、根系提取分析系统[J].中国农学通报,2014,30(3):98-102.Jia Q Y,Liu J M,Xie Y B. The corn roots image stitehing,extraetion and analysis system based on matlab software[J].Chin. Agric. Sci. Bull.,2014,30(3):98-102.
[15]刘柯楠,杨启良,郭浩,等.基于不同滴灌方式的苹果幼树根系生长的三维可视化模拟[J].中国生态农业学报,2013,21(6):744-751.Liu K N,Yang Q L,Guo H,et al.. Modeling and visualization of three-dimensional young apple tree root growth under different modes of drip irrigation[J]. Chin. J. Eco-Agric.,2013,21(6):744-751.
[16]肖潇,戈振扬,于英杰,等.层切压实土壤中绿豆根系的图像处理[A].见:中国农业工程学会2011年学术年会论文集[C].重庆,2011.Xiao X,Ge Z Y,Yu Y J,et al.. Image processing of layer cutting of compacted root-soil of mung bean[A]. In:Proceedings of 2011 Academic Annual Meeting of Chinese Society of Agricultural Engineering[C]. Chongqing,2011.
[17] Iyer-Pascuzzi A S,Symonova O,Mileyko Y,et al.. Imaging and analysis platform for automatic phenotyping and trait ranking of plant root systems[J]. Plant Physiol.,2010,152(3):1148-1157.
[18] Clark R T,Mac Curdy R B,Jung J K,et al.. Three-dimensional root phenotyping with a novel imaging and software platform[J]. Plant Physiol.,2011,156(2):455-465.
[19]梅川,吴桂良,杨媛,等.一种基于区域增长和结构识别的心血管X线造影图像分割方法[J].生物医学工程学杂志,2014,31(2):413-420.Mei C,Wu G L,Yang Y,et al.. An approach for segentation of x-ray angiographic image based on region-growing and structure inferring[J]. J. Biomed. Engin.,2014,31(2):413-420.
[20] Frangi A F,Niessen W J,Vincken K L,et al.. Multiscale vessel enhancement filtering[A]. In:Proceeding of Medical Image Computing and Computer-Assisted Interventation MICCAI'98 Lecture Notes in Computer Science[C]. MA:s.n.,1998,130-137.
[21]张缓缓,李仁忠,景军锋,等. Frangi滤波器和模糊C均值算法相结合的织物瑕疵检测[J].纺织学报,2015,36(9):120-124.Zhang Y Y,Li R Z,Jing J F,et al.. Fabric defect detection based on frangi filter and fuzzy c-means algorithm in combination[J]. J. Textile Res.,2015,36(9):120-124.
[22]夏振华,石玉,于盛林.基于Gabor滤波器的指纹图像增强[J].工程图学学报,2006,25(5):80-85.Xia Z H,Shi Y,Yu C Y. Fingerprint image enhancement based on gabor filters[J]. J. Engin. Graphics,2006,25(5):80-85.
[23]廖开阳,张学东,章明珠,等.基于Gabor滤波器的指纹图像快速增强[J].计算机工程与应用,2009,45(10):172-175,198.Liao C Y,Zhang X D,Zhang M Z,et al.. Fast fingerprint enhancement based on gabor filter[J]. Computer Engin.Appl.,2009,45(10):172-175,198.
[24] Ostu N,Nobuyuki O,Otsu N. A threshold selection method from gray-level histogram IEEE transactions on systems[J].IEEE Trans. Syst. Man Cybernetics,1979,9(1):62-66.
[25]李旭,赵文杰,杨凯达.基于小目标预提取的OTSU分割方法[J].红外技术,2013,35(8):492-496.Li X, Zhao W J, Yang K D. OTSU applied in image segmentation based on small targets pre-detection[J]. Infrared Technol.,2013,35(8):492-496.
[26]黄颖,许可.融合Ncut方法的脊椎图像OTSU分割算法[J].计算机工程与应用,2010,46(25):188-190.Huang Y,Xu K. Spine OTSU segmentation algorithm based on Ncut method[J]. Computer Engin. Appl.,2010,46(25):188-190.
[27] He K,Sun J,Tang X. Guided image filtering[J]. IEEE Trans. Pattern Anal. Machine Intell.,2013,35(6):1397-1409.
[28] Bradley D,Roth G. Adapting thresholding using the integral image[J]. J. Graphics Tools,2007,12(2):13-21.
[29] Powers D M W. Evaluation:From precision,recall and Ffactor to ROC,informedness,markedness Correlation[J]. J.Machine Learn. Technol.,2011,2:2229-3981.
[30] Xiong X,Duan L F,Liu L B,et al.. Panicle-SEG:A robust image segmentation method for rice panicles in the field based on deep learning and superpixel optimization[J]. Plant Methods. 2017 13 13(1):104.
[31] Zhang X L,Feng X Z,Xiao P F,et al.. Segmentation quality evaluation using region-based precision and recall measures for remote sensing images[J]. Isprs J. Photogrammetry Remote Sensing,2015,102:73-84.