数学形态学在果实蝇分类上的应用
|
摘要
【目的】探讨数学形态特征在果实蝇分类鉴定中的可行性、有效性及稳定性,为后续建立果实蝇分类模型的研究提供依据和理论基础。【方法】以我国南方果实蝇类的优势种-橘小实蝇(Bactrocera dorsalis Hendel)、南瓜实蝇(Bactrocera tau Walker)、瓜实蝇(Bactrocera cucurbitae)作为研究对象,在常见的数学形态特征基础上,以果实蝇胸背板作为特征区域,提取其局部条纹特征作为分类依据,包括中心条纹的相对长度R_x、相对宽度R_y、偏心率E、相对周长L、相对面积S、形状参数F共6项数学形态特征,提取每类果实蝇样本图像的特征数据并进行统计分析,包括总体分布型假设检验、方差齐性检验、差异显著性检验,最后进行基于数学形态特征的果实蝇分类试验。【结果】6项数学形态特征总体上在同类果实蝇中均服从正态分布;相对长度R_x、相对宽度R_y、相对周长L、相对面积S在3类果实蝇之间具有差异显著性,中心条纹偏心率E、形状参数F在南瓜实蝇和瓜实蝇之间具有差异显著性。基于数学形态特征的果实蝇分类试验中,单个类别的果实蝇分类正确率均达到90%以上,单次分类耗时约500 ms,正确率与实时性均可以满足实际工作需求。【结论】6项数学形态特征能够作为果实蝇分类研究的依据,且可靠性较高,正确率达到95.56%。
【Objective】 The feasibility,effectiveness and stability of using mathematical morphological characteristics in classification and identification of fruit flies were discussed to provide basis for constructing fruit fly classification models.【Method】 Three dominant fruit flies in southern China, including Bactrocera dorsalis Hendel,Bactrocera tau Walker and Bactrocera cucurbitae,were selected in this study.The images of adults were processed by computer vision technology and segmentation algorithm.Based on the mathematical morphology of common fruit fly,the feature parameters were extracted from the characteristic region with chest backboard as local area.The six parameters included central fringe relative length R_x,relative width R_y,eccentricity E,relative perimeter L,relative area S,and shape parameter F.The feature data were extracted from each sample image before being statistically analyzed.The statistical analysis including the general distribution pattern hypothesis test,the variance homogeneity test and the difference saliency test was conducted to generate classification results.【Result】 All the 6 characteristics generally followed the normal distribution for same fruit fly types.There were significant differences in R_x,R_y,L and S among different types of fruit flies.There were significant differences in E and F between Bactrocera tau Walker and Bactrocera cucurbitae.In the fruit fly classification experiment based on mathematical morphological characteristics,the accuracy rate of single fruit flies was over 90%,and the single classification took about 500 ms.Both the accuracy and real-time prediction met the needs of practical work.【Conclusion】 The 6 mathematical morphological features can be used as the basis of fruit fly classification with high reliability.According to the six mathematicol morphological characteristics,the correct classification rate of the total experimental samples is 95.56%.
【Objective】 The feasibility,effectiveness and stability of using mathematical morphological characteristics in classification and identification of fruit flies were discussed to provide basis for constructing fruit fly classification models.【Method】 Three dominant fruit flies in southern China, including Bactrocera dorsalis Hendel,Bactrocera tau Walker and Bactrocera cucurbitae,were selected in this study.The images of adults were processed by computer vision technology and segmentation algorithm.Based on the mathematical morphology of common fruit fly,the feature parameters were extracted from the characteristic region with chest backboard as local area.The six parameters included central fringe relative length R_x,relative width R_y,eccentricity E,relative perimeter L,relative area S,and shape parameter F.The feature data were extracted from each sample image before being statistically analyzed.The statistical analysis including the general distribution pattern hypothesis test,the variance homogeneity test and the difference saliency test was conducted to generate classification results.【Result】 All the 6 characteristics generally followed the normal distribution for same fruit fly types.There were significant differences in R_x,R_y,L and S among different types of fruit flies.There were significant differences in E and F between Bactrocera tau Walker and Bactrocera cucurbitae.In the fruit fly classification experiment based on mathematical morphological characteristics,the accuracy rate of single fruit flies was over 90%,and the single classification took about 500 ms.Both the accuracy and real-time prediction met the needs of practical work.【Conclusion】 The 6 mathematical morphological features can be used as the basis of fruit fly classification with high reliability.According to the six mathematicol morphological characteristics,the correct classification rate of the total experimental samples is 95.56%.
引文
[1] 黄海燕,陈媛,周善义.蚁属(膜翅目:蚁科)4种的形态测量学分析 [J].环境昆虫学报,2017,39(1):226-231.Huang H Y,Chen Y,Zhou S Y.Morphometric analysis on four species of Formica (Hymenoptera:Formicidae ) [J].Journal of Environmental Entomology,2017,39(1):226-231.
[2] 蔡小娜,黄大庄,沈佐锐,等.用于昆虫分类鉴定的几何形态计量学方法研究:相对扭曲分析 [J].生物数学学报,2016,31(2):254-262.Cai X N,Huang D Z,Shen Z R,et al.Study on geometric morphometric method for insect classification and identification [J].Journal of Biomathematics,2016,31(2):254-262.
[3] 张蕾,陈小琳,侯新文,等.实蝇科果实蝇属昆虫数字图像自动识别系统的构建和测试 [J].昆虫学报,2011,54(2):184-196.Zhang L,Chen X L,Hou X W,et al.Construction and testing of automated fruit fly identification systembactrocera macquart (Diptera:Tephritidae) [J].Acta Entomologica Sinica,2011,54(2):184-196.
[4] 彭莹琼,廖牧鑫,张永红,等.基于BP神经网络模型的果实蝇自动分类系统 [J].江西农业学报,2016,38(6):1205-1210.Peng Y Q,Liao M X,Zhang Y H,et al.A study on the automatic classification system for fruit flies based on BP neural network model [J].Acta Agriculturae Jiangxi,2016,38(6):1205-1210.
[5] 文韬,洪添胜,李震,等.基于机器视觉的橘小实蝇运动轨迹跟踪与数量检测 [J].农业工程学报,2011,27(10):137-141.Wen T,Hong T S,Li Z,et al.Statistics and tracking of Bactrocera dorsalis based on machine vision [J].Transactions of the CSAE,2011,27(10):137-141.
[6] 汪露,黄丽莉,杨慧勇,等.果实蝇属昆虫图像识别系统的开发与测试 [J].植物检疫,2013,27(5):29-35.Wang L,Huang L L,Yang H Y,et al.Developing and testing of image identification system for Bactrocera spp.[J].Plant Quarantine,2013,27(5):29-35 .
[7] 周志艳,罗锡文,张扬,等.农作物虫害的机器检测与监测技术研究进展 [J].昆虫学报,2010,53(1):98-109.Zhou Z Y,Luo X W,Zhang Y,et al.Machine-based technologies for detecting and monitoring insect pests of crops:a review [J].Acta Entomologica Sinica,2010,53(1):98-109.
[8] 杨红珍,张建伟,李湘涛,等.基于图像的昆虫远程自动识别系统的研究 [J].农业工程学报,2008,24(1):188-192.Yang H Z,Zhang J W,Li X T,et al.Remote automatic identification system based on insect image [J].Transactions of the CSAE,2008,24(1):188-192.
[9] 赵汗青,沈佐锐,于新文.数学形态学在昆虫分类学上的应用研究:Ⅰ.在目级阶元上的应用研究 [J].昆虫学报,2003,46(1):45-50.Zhao H Q,Shen Z R,Yu X W.Use of math-morphological features in insect taxonomy:Ⅰ.At the order level [J].Acta Entomologica Sinica,2003,46(1):45-50.
[10] 赵汗青,沈佐锐,于新文.数学形态学在昆虫分类学上的应用研究:Ⅱ.在总科阶元上的应用研究 [J].昆虫学报,2003,46(2):201-208.Zhao H Q,Shen Z R,Yu X W.Use of math-morphological features in insect taxonomy:Ⅱ.At superfamily level [J].Acta Entomologica Sinica,2003,46(2):201-208.
[11] 沈佐锐,赵汗青,于新文.数学形态学在昆虫分类学上的应用研究:Ⅲ.在科阶元上的应用研究 [J].昆虫学报,2003,46(3):339-344.Shen Z R,Zhao H Q,Yu X W.Use of math-morphological features in insect taxonomy:Ⅲ.At the family level [J].Acta Entomologica Sinica,2003,46(3):339-344.
[12] 李阳,沈佐锐,董学超,等.利用翅的数学形态特征对20种夜蛾进行分类鉴定的研究 [J].中国农业大学学报,2016,21(9):80-89.Li Y,Shen Z R,Dong X C,et al.Math-morphological characters of wings for classification and identification of twenty species [J].Journal of China Agricultural University,2016,21(9):80-89.
[13] 蔡小娜,韩旭,沈佐锐,等.基于蛾翅翅脉特征的夜蛾昆虫数字化分类研究(鳞翅目:夜蛾科) [J].环境昆虫学报,2016,38(2):348-353.Cai X N,Han X,Shen Z R,et al.Digital classification of noctuid moths (Lepidoptera:Noctuidae) base on wings vein characte [J].Journal of Environmental Entomology,2016,38(2):348-353.
[14] 娄定风,章桂明,焦懿,等.基于形状与纹理算法的通用昆虫图像模式识别研究 [J].植物检疫,2012,26(4):10-15.Lou D F,Zhang G M,Jiao Y,et al.Research on insect image recognition with a common algorithm based on shape and texture [J].Plant Quarantine,2012,26(4):10-15.
[15] 李正,倪远平,刘迪,等.实蝇图像识别中的形态特征提取研究 [J].计算机仿真,2011,28(7):254-257.Li Z,Ni Y P,Liu D,et al.Math-morphological feature extraction in classification of tephritidae [J].Computer Simulation,2011,28(7):254-257.
[16] 张慧,张玉波,王祝祝,等.蠓科昆虫网络检索鉴定系统的设计与实现 [J].中国国境卫生检疫杂志,2017,40(5):334-337.Zhang H,Zhang Y B,Wang Z Z,et al.Development of a web-based search and identification system on Ceratopogonidae.[J].Chinese Frontier Health Quarantine,2017,40(5):334-337.
[17] Watson A,Neill M O,Kitching I.Automated identification of live moths (Macrolepidoptera) using digital automated identification system (DAISY) [J].Systematics & Biodiversity,2004,1(3):287-300.
[18] 李震,洪添胜,文韬,等.基于机器视觉技术识别实蝇成虫 [J].果树学报,2014,31(4):679-683.Li Z,Hong T S,Wen T,et al.Mature fruit fly identification using machine vision [J].Journal of Fruit Science,2014,31(4):679-683.
[19] 叶菁,吕俊峰,廖辉.2011～2016年广东口岸进境水果截获疫情分析与对策 [J].植物检疫,2017,31(4):81-84.Ye J,Lu J F,Liao H.Situation analysis of the pest intercepted and countermeasures for the imported fruits in Guangdong during 2011-2016.[J].Plant Quarantine,2017,31(4):81-84.
[20] 高媛惠,盖云鹏,李斌,等.2013～2014年我国口岸截获实蝇疫情分析 [J].植物检疫,2016,30(5):68-71.Gao Y H,Gai Y P,Li B,et al.Analysis of fruit flies intercepted in entry quarantine during 2013-2014 [J].Plant Quarantine,2016,30(5):68-71.
[21] 马锞,张瑞萍,陈耀华,等.瓜实蝇的生物学特性及综合防治研究概况 [J].广东农业科学,2010,37(8):131-135.Ma K,Zhang R P,Chen Y H,et al.Reviews in biology characteristic and integrated control of melon fly [J].Guangdong Agricultural Sciences,2010,37(8):131-135.
[22] 秦誉嘉.橘小实蝇在全球的种群结构、定殖风险及潜在分布研究 [D].北京:中国农业大学,2017.Qin Y J.Global population structure,establishment risk and potential geographical distribution of Bactrocera dorsails(Diptera:Tephritidae) [D].Beijing:China Agricultural University,2017.
[23] 马兴莉,李志红,胡学难,等.橘小实蝇、瓜实蝇和南亚果实蝇对广东省造成的经济损失评估 [J].植物检疫,2013,27(3):50-56.Ma X L,Li Z H,Hu X N,et al.The assessment of the economic losses caused by Bactrocera dorsalis,B.cucurbitae and B.tau to Guangdong province [J].Plant Quarantine,2013,27(3):50-56.
[24] 李震,邓忠易,洪添胜,等.基于神经网络的实蝇成虫图像识别算法 [J].农业机械学报,2017,48(S1):129-135.Li Z,Deng Z Y,Hong T S,et al.Image recognition algorithm for fruit flies based on BP neural network [J].T Chin Soc Agric Mach,2017,48(S1):129-135.
[25] 张国权,李战明,李向伟,等.HSV空间中彩色图像分割研究 [J].计算机工程与应用,2010,46(26):179-181.Zhang G Q,Li Z M,Li X W,et al.Research on color image segmentation in HSV space [J].Computer Engineering and Applications,2010,46(26):179-181.
[26] 张辰,杨文柱,刘召海,等.基于HSV综合显著性的彩色图像分割方法 [J].计算机工程与设计,2013,34(11):3944-3947.Zhang C,Yang W Z,Liu Z H,et al.Color image segmentation based on compositive HSV saliency [J].Computer Engineering and Design,2013,34(11):3944-3947.
[27] 王植,贺赛先.一种基于Canny理论的自适应边缘检测方法 [J].中国图像图形学报,2004,9(8):954-962.Wang Z,He S X.An adaptive edge-detection method based on Canny algorithm [J].Journal of Image and Graphics,2004,9(8):954-962.
[28] 吴刚,谭彧,郑永军,等.基于改进Hough变换的收获机器人行走目标直线检测 [J].农业机械学报,2010,41(2):176-179.Wu G,Tan Y,Zheng Y J,et al.Walking goal line detection based on improved Hough transform on harvesting robot [J].T Chin Soc Agric Mach,2010,41(2):176-179.
[29] 程琮,范华.Levene方差齐性检验 [J].中国卫生统计,2005,22(6):408-420.Cheng Z,Fan H.Levene variance homogeneity test [J].Chinese Journal of Health Statistics,2005,22(6):408-420.
[30] 胡竹菁.平均数差异显著性检验统计检验力和效果大小的估计原理与方法 [J].心理学探新,2010,30(1):68-73.Hu Z Q.The estimation principle and method of statistical test force and effect size by mean difference significance test [J].Psychological Exploration,2010,30(1):68-73.
[31] Morton B.Brown,Alan B.Forsythe.Robust tests for the equality of variances [J].Journal of the American Statistical Association,1974,69(346):364-367.
[2] 蔡小娜,黄大庄,沈佐锐,等.用于昆虫分类鉴定的几何形态计量学方法研究:相对扭曲分析 [J].生物数学学报,2016,31(2):254-262.Cai X N,Huang D Z,Shen Z R,et al.Study on geometric morphometric method for insect classification and identification [J].Journal of Biomathematics,2016,31(2):254-262.
[3] 张蕾,陈小琳,侯新文,等.实蝇科果实蝇属昆虫数字图像自动识别系统的构建和测试 [J].昆虫学报,2011,54(2):184-196.Zhang L,Chen X L,Hou X W,et al.Construction and testing of automated fruit fly identification systembactrocera macquart (Diptera:Tephritidae) [J].Acta Entomologica Sinica,2011,54(2):184-196.
[4] 彭莹琼,廖牧鑫,张永红,等.基于BP神经网络模型的果实蝇自动分类系统 [J].江西农业学报,2016,38(6):1205-1210.Peng Y Q,Liao M X,Zhang Y H,et al.A study on the automatic classification system for fruit flies based on BP neural network model [J].Acta Agriculturae Jiangxi,2016,38(6):1205-1210.
[5] 文韬,洪添胜,李震,等.基于机器视觉的橘小实蝇运动轨迹跟踪与数量检测 [J].农业工程学报,2011,27(10):137-141.Wen T,Hong T S,Li Z,et al.Statistics and tracking of Bactrocera dorsalis based on machine vision [J].Transactions of the CSAE,2011,27(10):137-141.
[6] 汪露,黄丽莉,杨慧勇,等.果实蝇属昆虫图像识别系统的开发与测试 [J].植物检疫,2013,27(5):29-35.Wang L,Huang L L,Yang H Y,et al.Developing and testing of image identification system for Bactrocera spp.[J].Plant Quarantine,2013,27(5):29-35 .
[7] 周志艳,罗锡文,张扬,等.农作物虫害的机器检测与监测技术研究进展 [J].昆虫学报,2010,53(1):98-109.Zhou Z Y,Luo X W,Zhang Y,et al.Machine-based technologies for detecting and monitoring insect pests of crops:a review [J].Acta Entomologica Sinica,2010,53(1):98-109.
[8] 杨红珍,张建伟,李湘涛,等.基于图像的昆虫远程自动识别系统的研究 [J].农业工程学报,2008,24(1):188-192.Yang H Z,Zhang J W,Li X T,et al.Remote automatic identification system based on insect image [J].Transactions of the CSAE,2008,24(1):188-192.
[9] 赵汗青,沈佐锐,于新文.数学形态学在昆虫分类学上的应用研究:Ⅰ.在目级阶元上的应用研究 [J].昆虫学报,2003,46(1):45-50.Zhao H Q,Shen Z R,Yu X W.Use of math-morphological features in insect taxonomy:Ⅰ.At the order level [J].Acta Entomologica Sinica,2003,46(1):45-50.
[10] 赵汗青,沈佐锐,于新文.数学形态学在昆虫分类学上的应用研究:Ⅱ.在总科阶元上的应用研究 [J].昆虫学报,2003,46(2):201-208.Zhao H Q,Shen Z R,Yu X W.Use of math-morphological features in insect taxonomy:Ⅱ.At superfamily level [J].Acta Entomologica Sinica,2003,46(2):201-208.
[11] 沈佐锐,赵汗青,于新文.数学形态学在昆虫分类学上的应用研究:Ⅲ.在科阶元上的应用研究 [J].昆虫学报,2003,46(3):339-344.Shen Z R,Zhao H Q,Yu X W.Use of math-morphological features in insect taxonomy:Ⅲ.At the family level [J].Acta Entomologica Sinica,2003,46(3):339-344.
[12] 李阳,沈佐锐,董学超,等.利用翅的数学形态特征对20种夜蛾进行分类鉴定的研究 [J].中国农业大学学报,2016,21(9):80-89.Li Y,Shen Z R,Dong X C,et al.Math-morphological characters of wings for classification and identification of twenty species [J].Journal of China Agricultural University,2016,21(9):80-89.
[13] 蔡小娜,韩旭,沈佐锐,等.基于蛾翅翅脉特征的夜蛾昆虫数字化分类研究(鳞翅目:夜蛾科) [J].环境昆虫学报,2016,38(2):348-353.Cai X N,Han X,Shen Z R,et al.Digital classification of noctuid moths (Lepidoptera:Noctuidae) base on wings vein characte [J].Journal of Environmental Entomology,2016,38(2):348-353.
[14] 娄定风,章桂明,焦懿,等.基于形状与纹理算法的通用昆虫图像模式识别研究 [J].植物检疫,2012,26(4):10-15.Lou D F,Zhang G M,Jiao Y,et al.Research on insect image recognition with a common algorithm based on shape and texture [J].Plant Quarantine,2012,26(4):10-15.
[15] 李正,倪远平,刘迪,等.实蝇图像识别中的形态特征提取研究 [J].计算机仿真,2011,28(7):254-257.Li Z,Ni Y P,Liu D,et al.Math-morphological feature extraction in classification of tephritidae [J].Computer Simulation,2011,28(7):254-257.
[16] 张慧,张玉波,王祝祝,等.蠓科昆虫网络检索鉴定系统的设计与实现 [J].中国国境卫生检疫杂志,2017,40(5):334-337.Zhang H,Zhang Y B,Wang Z Z,et al.Development of a web-based search and identification system on Ceratopogonidae.[J].Chinese Frontier Health Quarantine,2017,40(5):334-337.
[17] Watson A,Neill M O,Kitching I.Automated identification of live moths (Macrolepidoptera) using digital automated identification system (DAISY) [J].Systematics & Biodiversity,2004,1(3):287-300.
[18] 李震,洪添胜,文韬,等.基于机器视觉技术识别实蝇成虫 [J].果树学报,2014,31(4):679-683.Li Z,Hong T S,Wen T,et al.Mature fruit fly identification using machine vision [J].Journal of Fruit Science,2014,31(4):679-683.
[19] 叶菁,吕俊峰,廖辉.2011～2016年广东口岸进境水果截获疫情分析与对策 [J].植物检疫,2017,31(4):81-84.Ye J,Lu J F,Liao H.Situation analysis of the pest intercepted and countermeasures for the imported fruits in Guangdong during 2011-2016.[J].Plant Quarantine,2017,31(4):81-84.
[20] 高媛惠,盖云鹏,李斌,等.2013～2014年我国口岸截获实蝇疫情分析 [J].植物检疫,2016,30(5):68-71.Gao Y H,Gai Y P,Li B,et al.Analysis of fruit flies intercepted in entry quarantine during 2013-2014 [J].Plant Quarantine,2016,30(5):68-71.
[21] 马锞,张瑞萍,陈耀华,等.瓜实蝇的生物学特性及综合防治研究概况 [J].广东农业科学,2010,37(8):131-135.Ma K,Zhang R P,Chen Y H,et al.Reviews in biology characteristic and integrated control of melon fly [J].Guangdong Agricultural Sciences,2010,37(8):131-135.
[22] 秦誉嘉.橘小实蝇在全球的种群结构、定殖风险及潜在分布研究 [D].北京:中国农业大学,2017.Qin Y J.Global population structure,establishment risk and potential geographical distribution of Bactrocera dorsails(Diptera:Tephritidae) [D].Beijing:China Agricultural University,2017.
[23] 马兴莉,李志红,胡学难,等.橘小实蝇、瓜实蝇和南亚果实蝇对广东省造成的经济损失评估 [J].植物检疫,2013,27(3):50-56.Ma X L,Li Z H,Hu X N,et al.The assessment of the economic losses caused by Bactrocera dorsalis,B.cucurbitae and B.tau to Guangdong province [J].Plant Quarantine,2013,27(3):50-56.
[24] 李震,邓忠易,洪添胜,等.基于神经网络的实蝇成虫图像识别算法 [J].农业机械学报,2017,48(S1):129-135.Li Z,Deng Z Y,Hong T S,et al.Image recognition algorithm for fruit flies based on BP neural network [J].T Chin Soc Agric Mach,2017,48(S1):129-135.
[25] 张国权,李战明,李向伟,等.HSV空间中彩色图像分割研究 [J].计算机工程与应用,2010,46(26):179-181.Zhang G Q,Li Z M,Li X W,et al.Research on color image segmentation in HSV space [J].Computer Engineering and Applications,2010,46(26):179-181.
[26] 张辰,杨文柱,刘召海,等.基于HSV综合显著性的彩色图像分割方法 [J].计算机工程与设计,2013,34(11):3944-3947.Zhang C,Yang W Z,Liu Z H,et al.Color image segmentation based on compositive HSV saliency [J].Computer Engineering and Design,2013,34(11):3944-3947.
[27] 王植,贺赛先.一种基于Canny理论的自适应边缘检测方法 [J].中国图像图形学报,2004,9(8):954-962.Wang Z,He S X.An adaptive edge-detection method based on Canny algorithm [J].Journal of Image and Graphics,2004,9(8):954-962.
[28] 吴刚,谭彧,郑永军,等.基于改进Hough变换的收获机器人行走目标直线检测 [J].农业机械学报,2010,41(2):176-179.Wu G,Tan Y,Zheng Y J,et al.Walking goal line detection based on improved Hough transform on harvesting robot [J].T Chin Soc Agric Mach,2010,41(2):176-179.
[29] 程琮,范华.Levene方差齐性检验 [J].中国卫生统计,2005,22(6):408-420.Cheng Z,Fan H.Levene variance homogeneity test [J].Chinese Journal of Health Statistics,2005,22(6):408-420.
[30] 胡竹菁.平均数差异显著性检验统计检验力和效果大小的估计原理与方法 [J].心理学探新,2010,30(1):68-73.Hu Z Q.The estimation principle and method of statistical test force and effect size by mean difference significance test [J].Psychological Exploration,2010,30(1):68-73.
[31] Morton B.Brown,Alan B.Forsythe.Robust tests for the equality of variances [J].Journal of the American Statistical Association,1974,69(346):364-367.