摘要
本文以农林业主要蛾类害虫为研究对象，利用数学形态学和几何形态测量学方法对5总科8科39种蛾类昆虫进行了数字化分类鉴定研究，并分别在总科、科及种分类阶元上探讨了这些方法在昆虫分类中应用的可行性和可靠性及筛选出的用于分类鉴定的特征参数的重要性。
首先利用数字化技术获取蛾翅和蛾翅脉图像，并对蛾翅图像进行阈值分割和平滑去噪等预处理操作。然后利用BugShape1.0软件从蛾前后翅图像中提取矩形度、延长度、叶状性、偏心率、球状性、似圆度和不变矩Hu1、Hu2等共计26项与大小尺度和方向均无关的数学形态特征，以用于蛾类昆虫在各阶元上的分类鉴定。对分类正确率较低的种类，再利用几何形态测量学方法对蛾翅脉图像进行进一步的分析。
研究结果表明，利用蛾翅数学形态特征成功地实现了对蛾类昆虫在总科、科及种分类阶元上的分类鉴定，其中仅有天蛾科昆虫的识别正确率较低，回归和交叉判别的正确率分别为：87.7%和84.0%，在总科和科阶元以及其它蛾类昆虫种类的识别正确率均较高。
在总科阶元上的回归和交叉识别正确率分别为100.0%和97.4%，筛选出了6项特征参数用于分类鉴定，其重要性大小为：（FW矩形度、FWHu5、HW偏心率）＞HW似圆度＞（HW矩形度、HWHu5）；科阶元上的识别正确率分别为100.0%和92.1%，筛选出的6项特征参数重要性大小为：（FWHu6、FW矩形度、FWHu5）＞HWHu2＞（HW似圆度、FW似圆度）。
在种阶元上的回归和交叉识别正确率均大于93.3%，筛选出的用于夜蛾科昆虫分类鉴定的11项特征参数的重要性为：（FW偏心率、FWHu5、FWHu7）＞FWHu2＞FW似圆度＞FW球状性＞FWHu3＞（FW叶状性、FWHu1、FWHu6）＞FWHu4；筛选出的用于刺蛾科昆虫分类鉴定的7项特征参数的重要性为：（HW叶状性、HWHu7）＞（HWHu6、FWHu5、FW叶状性）＞（FW矩形度、HW延长度）；用于灯蛾科分类鉴定的13项特征参数的重要性为：（FWHu3、FWHu7、HWHu7、HW偏心率）＞（HWHu4、HWHu3、FW似圆度、FW球状性、FWHu4）＞（FW延长度、FWHu5、FW叶状性、FWHu1）；用于枯叶蛾科分类鉴定的13项特征参数的重要性为：（HWHu3、FW偏心率、FWHu6、FW球状性）＞（FW似圆度、HW矩形度、FWHu4、FWHu3、FWHu1）＞（FWHu2、HW延长度、FW矩形度、HWHu1）；用于舟蛾科分类鉴定的11项特征参数的重要性为：（FW偏心率、HWHu1、FWHu5、FW矩形度）＞（FW球状性、FW似圆度、FW叶状性、HW矩形度）＞（HWHu7、FWHu1、FWHu2）；用于毒蛾科分类鉴定的8项特征参数的重要性为：（FW似圆度、FWHu4、HW矩形度）＞（FW延长度、FWHu1、HWHu6）＞（HW叶状性、FW矩形度）；用于尺蛾科分类鉴定的8项特征参数的重要性为：（FWHu2、FW叶状性、HW似圆度）＞（HW延长度、HWHu6、FWHu1）＞（FWHu3、FW偏心率）。
蛾翅数学形态特征在不同分类阶元或不同种类上应用的有效性和重要性有所不同，但是，在各分类阶元上，利用最终筛选出的蛾翅数学形态特征能够很好的用于其分类鉴定，这表明各数学形态特征在不同的分类阶元上有着不同的作用。
对于分类识别正确率较低的天蛾科昆虫，利用几何形态测量学方法对其蛾翅脉图像进行了进一步的研究，利用TpsDig软件按翅脉交点顺序共选取了前翅17个翅脉交点作为标记点，并利用软件TpsSuper和TpsRelw对其进行了普氏叠加和相对扭曲分析，最后天蛾科昆虫的回归和交叉识别正确率达到了100.0%和99.7%。这表明蛾翅数学形态特征和翅脉的几何形态特征均具有很好的分类和鉴别作用，为未来逐步实现蛾类昆虫的自动识别奠定了基础。
最后利用人工神经网络方法对夜蛾科昆虫做了进一步的分类鉴定研究，研究结果表明采用BP神经网络可以实现昆虫的分类鉴定，且试验表明与数学方法相结合，如采用主成分分析对昆虫的数学形态特征数据进行预处理，可以提高BP神经网络对昆虫分类鉴定的准确度。这为日后逐步形成自动鉴定系统做了初步的探索研究。
The digital classification and identification of thirty nine moths (five superfamilies,eight familys) were studied by using the methods of mathematical morphology andGeometric morphometry, taking the moths which are harmful to agriculture and forestryas material. Each method was evaluated on the feasibility and reliability, and theclassificatory importance of each morphological character was analyzed in classificationof the moths at superfamily level, family level and species level.
At first, digital technology was used to get images of wing and vein of moths andpretreatments such as threshold segmentation and smooth denoising were applied toimages of moth wing. Then twenty six math-morphological characters(MMCs) such aseccentricity, sphericity, lobation, roundness, rectangularity, elongation and movementinvariants including Hu1and Hu2were selected for being invariant to the image size anddirection by software BugShape1.0. These MMCs which were used for classification anddiscrimination of the moths at each level were extracted from the images of rightforewing and right hindwing of thirty nine moths. For the species which had relativelylower classification accuracy resulting from the method, Geometric morphometry wasused to make further analysis on their images.
The analytic results showed that by using MMCs of moth wings, classification anddiscrimination of the moths at each level were realized successfully. High classificationaccuracies were got for all moths at superfamily level, family level and species levelexcept for Sphingidae insects with accuracies of regression and intersecting discriminantanalysis87.7%and84.0%respectively
The accuracies of regression and intersecting discriminant analysis were100.0%and97.4%respectively, and six MMCs were selected as the classification variables atsuperfamily level. The contribution of these variables were ranked as follows:(FW-rectangularity, FW-Hu5, HW-eccentricity)＞HW-roundness＞(HW-rectangularity,HW-Hu5). Whereas, at family level, the accuracies of regression and intersectingdiscriminant analysis were100.0%and92.1%respectively, and the contribution of sixselected variables were ranked as follows:(FW-Hu6, FW-rectangularity, FW-Hu5)＞HW-Hu2＞(HW-roundness, FW-roundness).
The accuracies of regression and intersecting discriminant analysis were both higherthan93.3%at species level. The contribution of eleven selected variables used toNoctuidae insects were ranked as follows:(FW-eccentricity, FWHu5, FWHu7）＞FWHu2＞FW-roundness＞FW-sphericity＞FWHu3＞(FW-lobation, FWHu1, FWHu6) ＞FWHu4; for Eucleidae insects:(HW-lobation, HWHu7)＞(HWHu6, FWHu5,FW-lobation)＞(FW-rectangularity, HW-elongation); for Arctiidae insects:(FWHu3,FWHu7, HWHu7, HW-eccentricity)＞(HWHu4, HWHu3, FW-roundness, FW-sphericity,FWHu4)＞(FW-elongation, FWHu5, FW-lobation, FWHu1); for Lasiocampidae insects:(HWHu3, FW-eccentricity, FWHu6, FW-sphericity)＞(FW-roundness,HW-rectangularity, FWHu4, FWHu3, FWHu1)＞(FWHu2, HW-elongation,FW-rectangularity, HWHu1); for Notodontidae insects:(FW-eccentricity, HWHu1,FWHu5, FW-rectangularity)＞(FW-sphericity, FW-roundness, FW-lobation,HW-rectangularity)＞(HWHu7, FWHu1, FWHu2); for Lymantriidae insects:(FW-roundness, FWHu4, HW-rectangularity)＞(FW-elongation, FWHu1, HWHu6)＞(HW-lobation, FW-rectangularity); and for Geometridae insects:(FWHu2, FW-lobation,HW-roundness)＞(HW-elongation, HWHu6, FWHu1)＞(FWHu3, FW-eccentricity).
Differences were existed for the effectiveness and importance of MMCs of mothwings used to different taxonomic category or different species. However, at each level,the selected MMCs could be used for classification and discrimination well, which meantthat each MMC played varied role at different level.
The method of geometric morphometry was used to analyze the vein images ofSphingidae insects which had relatively lower classification accuracy. Seventeenintersection points of the veins were selected as landmarks by using TpsDig software insequence of the intersection points. Procrustes superimposition and relative warpanalyses were done by applying TpsSuper and TpsRelw softwares. Finally, the accuraciesof regression and intersecting discriminant analysis reached to100.0%and99.7%respectively. The research indicated that both the MMCs of the moth wings and theGeometric morphological characters of the veins had good function for classification anddiscrimination and therefore, laid foundation for gradually realizing the automaticrecognition of moths in the future.
Finally, artificial neural network method was applied to do further study. The resultsindicated that BP neural network was able for classification and identification ofNoctuidae insects and identification accuracy could be enhanced by combining withmathematical method such as principal component analysis which was used for datapreprocessing of the mathematical morphological characteristics in this study. The studyplayed an elementary and exploratory role in forming automatic identification system.
引文
[1]朱弘复.蛾类图册[M].北京:科学出版社,1980.
[2]彩万志,庞雄飞,花保祯,等.普通昆虫学[M].北京:中国农业大学出版社,2001.
[3]蔡邦华.昆虫分类学(中册)[M].北京:中国农业大学出版社,1973.
[4]袁锋.昆虫分类学[M].北京:中国农业出版社,1996.
[5]管致和.昆虫学通论(上册)[M].北京:中国农业出版社,1993.
[6]O'Neill MA,Gaston KJ..Automated species identification:why not?[J]. Phil Trans R Soc Lond,2004,359:655-667.
[7]陈小琳,侯新文,刘成林,等.昆虫图像自动鉴别技术[J].昆虫知识,2008,45(2):317-322.
[8]吴丹,刘修国,尚建嘎.数学形态学在图像处理与分析中的应用及展望[J].工程图学学报,2003,(2):120-125.
[9]岳宏伟,李扬,蔡肯,等.数学形态学在图像处理中的应用与展望[J].影像技术,2006,(2):19-21.
[10]杨述斌,彭复员.数学形态学在图像处理中的应用与发展[J].武汉化工学院学报,2004,26(1):70-73.
[11]Schowalter T D. Insect ecology: an ecosystem approach (2nd ed)[M]. San Diago: Academic Press,2006.
[12]Liu J D. The expert system for identification of tortricinae (Lepidoptera) using image analysis of venation[J].Entomol.Sin.,1996,3(1):1-8.
[13]Liu J D. How to construct the expert system for species identification using venation of tortricinae(Lepidoptera)[J]. Entomol.Sin.,1996,3(2):133-137.
[14]Albrecht A, Kaila L. Variation of wing venation in elachistidae (Lepidoptera: Gelechioidea): methodology andimplications to systematics[J]. Syst. Entomol.,1997,22:185-198.
[15]Adsavakulchai S, Baimai V, Prachyabrued W, et al. Morphometric study using wing image analysis foridentification of the bactrocera dorsalis complex (Diptera:Tephritidae)[J]. World Wide Web J. Biol.,1998,3.(URL: http://www.epress.com/w3jbio/vol3/Adsavakulchai/index.html).
[16]Weeks P J D, O'Neill M A, Gaston K J, et al. Automating insect identification: exploring the limitations of aprototype system[J]. Appl. Entomol.,1999,123:1-8.
[17]O'Neill MA. DAISY: a practical tool for semi-automated species identification[J].2008.URL: http://www.fao.org/ag/agp/agps/c-cab/castudies/pdf/3–001.pdf)
[18]Rohlf FJ. TpsUtil. Ecology and evolution, SUNY. New York: Stony Brook,1998. http://life.bio.sunysb.edu/morph.
[19]Rohlf FJ. TpsDig. Ecology and evolution, SUNY. New York: Stony Brook,1998. http://life.bio.sunysb.edu/morph.
[20]Rohlf FJ. TpsSuper. Ecology and evolution, SUNY. New York: Stony Brook,1998. http://life.bio.sunysb.edu/morph.
[21]Rohlf FJ. TpsRelw. Ecology and evolution, SUNY. New York: Stony Brook,1998. http://life.bio.sunysb.edu/morph.
[22]Arbuckle T, Schr der S, Steinhage V, et al. Biodiversity informatics in action: identification and monitoring ofbee species using ABIS sustainability in the information society[C]. Conf. Environ. Inf.,2001,1:425-430.
[23]Houle D, Mezey J, Galpern P, et al. Automated measurement of Drosophila wings[J]. BMC Evolutionarybiology,2003,3pp13.[http://www.biomedcentral.com/1471-2148/3/25]
[24]Tofilski A. DrawWing, a program for numerical description of insect wings[J]. Insect Sci.,2004,4(17):1-5.
[25]Gilchrist A S, Crisafulli D C A. Using variation in wing shape to distinguish between wild and mass-rearedindividuals of queensland fruit fly, Bactrocera tryoni[J]. Neth.Entomol. Soc.,2006,119:175-178.
[26]MacLeod N,2007. Automated Taxon Identification in Systematics: Theory, Approaches and Applications.CRC Press, London.
[27]Prieto C, Munguira M, Romo H,2009. Morphometric analysis of genitalia and wing pattern elements in thegenus Cupido (Lepidoptera, Lycaenidae): are Cupido minimus and C. carswelli different species? Deul.Entomol. Z.,56:137–147.
[28]Andrade C., Vieira R., Ananina G., Klaczko L.2009. Evolution of the male genitalia: morphological variationof the aedeagi in a natural population of Drosophila mediopunctata. Genetica135:13–23.
[29]Nikolaou N, Sampaziotis P, Aplikioti M, et al. Vestis: a versatile semi-automatic taxon identification systemfrom digital images [J]. Tools for Identifying Biodiversity: Progress and Problems,2010,231-236.
[30]沈佐锐,于新文.昆虫数学形态学研究及其应用展望[J].昆虫学报,1998,41(增刊):140-148.
[31]于新文.昆虫图像数字技术的研究开发[D].北京:中国农业大学,1999.
[32]张建伟.基于计算机视觉技术的蝴蝶自动识别研究[D].北京:中国农业大学,2006.
[33]赵汗青,沈佐锐,于新文.数学形态特征应用于昆虫自动鉴别的研究.中国农业大学学报,2002,7(3):38-42
[34]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.
[35]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.
[36]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.
[37]潘鹏亮,杨红珍,沈佐锐,等.翅脉的数学形态特征在蝴蝶分类鉴定中的应用研究[J].昆虫分类学报,2008,30(2):151-160.
[38]潘鹏亮,沈佐锐,杨红珍,等.三种绢蝶翅脉数字化特征的提取及初步分析[J].动物分类学报,2008,33(3):566-571.
[39]潘鹏亮,沈佐锐,高灵旺,等.昆虫翅脉特征自动获取技术的初步研究[J].昆虫分类学报,2008,30(1):72-80.
[40]张红涛,毛罕平,邱道尹.储量害虫图像识别中的特征提取[J]．农业工程学报,2009,25(2):126-130.
[41]邱道尹,张红涛,陈铁军,等.模糊识别技术在储粮害虫检测中的应用[J].农业系统科学与综合研究,2002,18(2):122-125．
[42]张蕾,陈小琳,侯新文.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.
[43]梁广勤,杨国海,梁帆,等.利用染色体和同工酶技术鉴定寡鬃实蝇幼虫[J].植物检疫,1994,8(1):4-9.
[44]魏文娟,任炳忠.我国直翅目昆虫细胞分类学研究现状[J].昆虫知识,2004,41(2):123-126.
[45]徐丽霞,王琴琴.染色体核型分析技术的发展[J].医学综述,2009,15(2):188-190.
[46]屈娟,罗辉.利用酶电泳方法快速鉴定斑潜蝇[J].植物检疫,1997,11(增刊):47-49.
[47]卢荣胜.尺蛾科,灯蛾科部分种类酯酶同工酶的研究[D].陕西师范大学,2000.
[48]Dadour I.R.,Yeates D.K.,Postle A.C..两种区分昆虫兰实蝇和地中海实蝇的快速鉴定方法[J].植物检疫,1993,7(6):447-448.
[49]杜波,迟德福.PCR-SSCP技术在昆虫学研究中的应用[J].昆虫知识,2007,44(3):333-336.
[50]包立军,蔡平,陈正凯.生物技术在昆虫分类中的应用[J].淮阴工学院学报,2006,15(3):53-56.
[51]朱家颖,李正跃,肖春,等.生物技术在昆虫分类中的研究现状与展望[J].江西农业大学学报,2003,25(5):765-771.
[52]任海英,李彤.昆虫分子生物学分类技术及在植保上的应用前景[J].绿色科技,2012,(9):64-66.
[53]刘勇,宋毓,李晓宇.基于线粒体COI基因的DNA条形码技术在昆虫分子鉴定中的应用[J].植物检疫,2010,24(2):46-50.
[54]韦健红,吴文如,喻良文,等.DNA条形码技术在生物分类中的应用[J].广东药学院学报,2010,26(4):430-433.
[55]杨聪慧,韩辉林,迟美妍,等.DNA条形码技术在北京百花山地区夜蛾科物种鉴定中的应用[J].昆虫学报,2012,55(9):1082-1092.
[56]刘慎思.DNA条形码技术在主要实蝇类害虫识别中的应用[D].西南大学,2012.
[57]廖力,徐淼锋,张卫东.应用DNA条形码技术鉴别谷实夜蛾与棉铃虫[J].植物检疫,2012,26(6):12-16.
[58]李文芬.地中海实蝇及其近缘种和桔小实蝇复合种基因芯片的研制及应用[D].湖南师范大学,2008.
[59]唐侃.利用微卫星标记鉴定寡毛实蝇属幼虫及利用线粒体序列分析中国桔小实蝇种群结构的研究[D].广西师范大学,2006.
[60]刘燕.储粮昆虫图像模式识别研究[D].成都理工大学,2002.
[61]于新文,沈佐锐,高灵旺,等.昆虫图像几何形状特征的提取技术研究[J].中国农业大学学报,2003,8(3):47-50.
[62]周红,王宏坡.基于Visual C++.NET的昆虫图像自动识别系统的研究[J].天津农学院学报,2005,12(2):39-41.
[63]唐强.基于图像模式识别技术的昆虫识别研究[D].昆明理工大学,2006.
[64]程小梅.基于图像的昆虫识别研究与设计[D].西北大学,2008.
[65]黄世国.基于图像的昆虫识别关键技术研究[D].西北大学,2008.
[66]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.
[67]李小林,林大辉,李威,等.基于图像的昆虫远程自动识别[J]。福建农林大学学报,2008,37(5):456-459.
[68]王江宁,宋志顺,纪力强,等.相对特征在昆虫目级阶元图像自动鉴定中的应用[J].动物分类学报,2010,35(3):585-590.
[69]竺乐庆,张真,张培毅.基于颜色直方图及双树复小波变换(DTCWT)的昆虫图像识别[J].昆虫学报,2010,53(1):91-97.
[70]姚青,吕军,杨保军,等.基于图像的昆虫自动识别与计数研究进展[J].中国农业科学,2011,44(14):2886-2899.
[71]刘芳,沈佐锐,张建伟,等.基于颜色特征的昆虫自动鉴定方法[J].昆虫知识,2008,45(1):150-153.
[72]姜红军,王可佳,海鹰,等.基于数学形态学的医学图像自动阈值分割[J].内蒙古师范大学学报,2008,37(3):397-399.
[73]郭臻,陈远知.图像阈值分割算法研究[J].中国传媒大学学报,2008,15(2):77-82.
[74]沈圆,李国勇.图像的阈值分割法设计[J].机械工程与自动化,2012,(5):45-46.
[75]陶文兵,金海.一种新的基于图谱理论的图像阈值分割方法[J].计算机学报,2007,30(1):110-119.
[76]高月红.灰度图像分割算法的研究[J].科级信息,2009,(27):409-410.
[77]朱友勇.基于图论的图像分割方法的研究与应用-基于图论的图像阈值分割方法研究[D].江南大学,2009.
[78]杨海峰,侯朝桢.基于二维灰度直方图的蚁群图像分割[J].激光与红外,2005,35(8):614-617.
[79]韩帅.灰度图像分割算法的研究[D].山东大学,2007.
[80]石殿国.基于图论的灰度图像分割[D].武汉理工大学,2009.
[81]闫海霞.基于数学形态学的图像边缘检测和增强算法的研究[D].吉林大学,2009.
[82]何宇.基于数学形态学的图像矢量化研究[J].软件导刊,2012,11(9):175-177.
[83]王建斌,樊新华,顾建锋.数学形态学在昆虫鉴定中的应用[J].河北农业科学,2012,16(4):80-82.
[84]双娜.数字图像中平滑去噪技术研究及实现[J].图形图像,2010,(6):55-57.
[85]任获荣.数学形态学及其应用[D].西安电子科技大学,2004.
[86]范立南.图像处理与模式识别[M].北京:科学出版社,2007.
[87]杨琨,曾立波,王殿成.数学形态学腐蚀膨胀运算的快速算法[J].计算机工程与应用,2005,34:54-56.
[88]马向华.数字图像中平滑去噪技术探讨[J].装备制造技术,2012,(7):36-38.
[89]袁雄峰.鼠类头骨标本数学形态特征分析与自动识别技术研究[D].中国农业大学,2011.
[90]谢敏,沈佐锐,施宗伟,等.3种菟丝子种子数学形态特征的初步研究[J].植物保护,2008,34(6):53-56.
[91]闫宝荣,花保祯.几何形态测量学及其在昆虫分类学和系统发育中的应用[J].昆虫分类学报,2010,32(4):313-320.
[92]Bookstein F L. Morphometric tools for landmark data: geometry and biology[M].Cambridge UniversityPress,1991.
[93]Bookstein F L. Morphometrics in evolutionary biology: the geometry of size and shape change, withexamples from fishes[M]. Academy of Natural Sciences,1985.
[94]曾志杰,何文,陈善勤.甘肃临夏盆地晚中新世鬣狗类群头骨的几何形态测量学及生态形态学分析[J].古脊椎动物学报,2010,48(3):235-246.
[95]萧旭峰,吴文哲.生物形状的科学－浅谈几何形态测量学之发展与应用[J].科学月刊,1998,29:624-633.
[96]Rohlf F J. Shape statistics: Procrustes superimpositions and tangent spaces[J]. Journal of Classification,1999,16:197-223.
[97]刘武,Ronald C.,刑松.湖北建始更新世早期人类牙齿几何形态测量分析[J].中国科学:地球科学,2010,40(6):724-736.
[98] Kendall D G. Exact distributions for shapes of random triangles in convex sets[J]. Advances in AppliedProbability,1985,17:308-329.
[99] Kendall D G, Le H L. Exact shape-densities for random triangles in convex polygons[J]. Advances in AppliedProbability,1986,18:59-72.
[100]罗传毅.方差分析的简便方法及应用[J].吉林化工学院学报,1985,(4):58-64.
[101]李吉勋.方差分析的实用方法[J].畜牧兽医杂志,1991,(2):31-34.
[102]石红霞.方差分析计算方法综述[J].洛阳农业高等专科学校学报,2002,22(2):111-112.
[103]陆荷英,吴玉林.试述方差分析的原理和方法[J].数理医药学杂志,2002,15(6):535.
[104]胡庆军,吴翊.主方差分析方法[J].国防科技大学学报,2000,22(2):117-120.
[105]章文波,陈红艳.实用数据统计分析及SPSS12.0应用[M].北京:人民邮电出版社,2006.
[106]于秀林,任雪松.多元统计分析[M].北京:中国统计出版社,1999.
[107]范金城,梅长林.数据分析[M].北京:科学出版社,2002.
[108]高陈燕.主成分分析法数学推导方法的研究[J].衡水学院学报,2010,12(4):7-8.
[109]黄传坤,王远清.主成分分析方法的应用研究[J].现代商贸工业,2011,(17):26-28.
[110]陈一心.中国动物志:昆虫纲.第十六卷,鳞翅目:夜蛾科[M].北京:科学出版社,1999.
[111]朱弘复,陈一心.中国经济昆虫志.第三册,鳞翅目:夜蛾科(一)[M].北京:科学出版社,1963.
[112] Güler Y, Aytekin AM, Ca atay N,2006. Systematical studies on Anthidiini (Hymenoptera: Megachilidae): Ageometric morphometric approach[J]. Acta Entomologica Sinica,49:474-483.
[113] Weeks PJD, Gauld ID, Gaston KJ, et al. Automating the identification of insects: a new solution to an oldproblem. Bull.Entomol.Res.,1997,87:203211.
[114]Villemant C, Simbolotti G, Kenis M,2007. Discrimination of Eubazus (Hymenoptera, Braconidae) siblingspecies using geometric morphometrics analysis of wing venation[J]. Systematic Entomology,32:625-634.
[115]Villemant C, Simbolotti G, Kenis M,2007. Discrimination of Eubazus (Hymenoptera, Braconidae) siblingspecies using geometric morphometrics analysis of wing venation[J]. Systematic Entomology,32:625-634.
[116]Klingenberg CP,2003. Developmental instability as a research tool: using patterns of fluctuating asymmetryto infer the developmentall origins of morphological integration. In: Polak M ed. Developmental Instability,Causes and Consequences. Oxford University Press,427-442.
[117]Yasemin GULER,A.Murat AYTEKIN,Nese CAGATAY.Systematical studies on Anthidiini (Hymenoptera：Megachilidae)A geometric morphometric approach[J].Acta Entomologlca Sinica,2006,49(3):474-483.
[118]Bookstein FL,1996. Applying landmark methods to biological outline data. In: Mardia K V, Gill C A, DrydenI L, eds. Image Fusion and Shape Variability Techniques. Leeds: Leeds University Press.79-87
[119]Bookstein FL,1997. Landmark methods for forms without landmarks: Morphometrics of group differences inoutline shape. Med Image Anal,1:225-243
[120]Sheets HD, Keonho K, Mitchell CE,2004. A combined landmark and outlinebased approach to ontogeneticshape change in the ordovician trilobite triarthrus becki. In: Elewa, A.(Ed.), Applications of Morphometricsin Paleontology and Biology.New York: Springer,67-81.
[121] Sampson PD, Bookstein FL, Sheehan H, Bolson EL,1996. Eigenshape analysis of left ventricular outlinesfrom contrast ventriculograms [A]. In: Marcus L.F,Corti M, Loy A, Naylor G J P., Slice DE.(Eds.),Advances in Morphometrics. Plenum,,New York,1996,131-152.
[122]Zelditch ML, Swiderski DL, Sheets HD, et al.,2004. Geometric morphometrics for biologists: A primer. SanDiego: Elsevier Academic Press.1-443.
[123]Slice DE,2005. Modern morphometrics in physical anthropology. New York: Kluwer Academic.1-383.
[124]刘武,Ronald CLARKE,邢松.湖北建始更新世早期人类牙齿几何形态测量分析[J].中国科学,2010,40(6):724-736.
[125] Adams DC, Rohlf FJ, Slice DE,2004. Geometric morphometrics: ten years of progress following the‘revolution’[J]. Ital. J. Zool.,71:5-16.
[126]王志有,于洪梅,李井会,等. BP人工神经网络紫外分光光度法同时测定三种氨基酸[J].生物数学学报,2005,20(2):240-244.
[127]鲍一丹,吴燕平,何勇. BP神经网络最优组合预测方法及其应用[J].农机化研究,2004,(3):162-164.
[128]石峰,娄臻亮,张永清.基于遗传算法和神经网络的冷挤压工艺参数模糊优化设计[J].机械工程学报,2002,38(8):45-49.
[129]刘燕德,应义斌.BP神经网络在农产品生产与检测中的应用[J].生物数学学报,2003,18(4):439-444.
[130]何勇,李晓丽,邵咏妮.基于主成分分析和神经网络的近红外光谱苹果品种鉴别方法研究[J].光谱学与光谱分析,2006,26(5):55-55.
[131]张方樱,巢佰崇,陈雪丰,2005.车牌自动识别系统.地理空间信息,3(3):4950
[132]连建兵,王浩全,2011.浅谈人脸识别技术及应用.机械管理开发,(2):99100
[133]飞思科技产品研发中心.MATLAB6.5辅助神经网络分析与设计[M].北京:电子工业出版社,2003.
[134]李军梅,胡以华,陶小红.基于主成分分析与BP神经网络的识别方法研究[J].红外与激光工程,2005,34(6):719-723.
[135]李增芳,何勇,宋海燕.基于主成分分析和集成神经网络的发动机故障诊断模型研究[J].农业工程学报,2006,22(4):131-134.
[136]吴景社,康绍忠,王景雷,等.基于主成分分析和模糊聚类方法的全国节水灌溉分区研究[J].农业工程学报,2004,20(4):64-68.
[137]张文彬,计时鸣.人工神经网络在刀具磨损检测中的应用[J].机床与液压,2004,(1):126-127.
[138]王歌,谢松云,党正.基于双神经网络分类器的脱机手写体汉字识别[J].西北工业大学学报,2010,28(4):574-578.
[139]于兰峰,王金诺.基于遗传算法和神经网络的塔机结构动态优化设计[J].中国机械工程,2008,19(1):61-63.
[140]张敏,钟志友,杨乐,等.基于BP神经网络的果蔬热导率预测模型[J].农业机械学报,2010,41(10):117-121.
[141]杨红珍,沈佐锐,李湘涛.昆虫自动鉴定技术研究与展望[J].四川动物,2011,30(5):834-838.