高级检索

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

檀香咖啡豹蠹蛾虫害的树干区域分类研究

陈珠琳 王雪峰

陈珠琳, 王雪峰. 檀香咖啡豹蠹蛾虫害的树干区域分类研究[J]. 北京林业大学学报, 2018, 40(1): 74-82. doi: 10.13332/j.1000-1522.20170306
引用本文: 陈珠琳, 王雪峰. 檀香咖啡豹蠹蛾虫害的树干区域分类研究[J]. 北京林业大学学报, 2018, 40(1): 74-82. doi: 10.13332/j.1000-1522.20170306
Chen Zhu-lin, Wang Xue-feng. Classification of sandalwood trunk area damaged by Zeuzera coffeae nietner in complex background[J]. Journal of Beijing Forestry University, 2018, 40(1): 74-82. doi: 10.13332/j.1000-1522.20170306
Citation: Chen Zhu-lin, Wang Xue-feng. Classification of sandalwood trunk area damaged by Zeuzera coffeae nietner in complex background[J]. Journal of Beijing Forestry University, 2018, 40(1): 74-82. doi: 10.13332/j.1000-1522.20170306

檀香咖啡豹蠹蛾虫害的树干区域分类研究

doi: 10.13332/j.1000-1522.20170306
基金项目: 

国家自然科学基金项目 31670642

详细信息
    作者简介:

    陈珠琳。主要研究方向:森林资源监测与计算机视觉。Email:825511059@qq.com 地址:100091 北京市海淀区东小府1号中国林业科学研究院资源信息研究所

    责任作者:

    王雪峰,博士,研究员。主要研究方向:森林资源监测与计算机视觉。Email:xuefeng@ifrit.ac.cn 地址:同上

  • 中图分类号: S763.38;TP391.41

Classification of sandalwood trunk area damaged by Zeuzera coffeae nietner in complex background

  • 摘要: 目的檀香是一种典型的珍贵树种,但不易成活且常遭受病虫害的干扰。本研究针对受咖啡豹蠹蛾危害的檀香,为提高受害檀香树干区域的分类效果,根据不同区域在颜色和纹理方面的特点,提出了一种复杂背景下受咖啡豹蠹蛾危害的檀香树干区域分类方法。方法(1) 本文首先结合高斯高通滤波、Otsu法、2G-R-B因子以及形态学运算,将树干从复杂背景中提取出来,再通过L*a*b*系统中的L*通道和a*通道将树干分割为健康区域、虫害区域和排泄物区域。(2)提出“多纹理”并使用“相对颜色”特征变换方法以提取图像特征。(3)提出BP神经网络和RBF-SVM相结合的方法对区域进行识别。结果(1) 本文提出的图像分割算法有效地将受虫害危害的檀香从背景中提取出来,并成功分割为健康区域、虫害区域和排泄物区域。(2)与传统PCA处理相比,使用“多纹理”和“相对颜色特征”得到的分类精度有所提高。其中,“多纹理”特征通过差异扩大的方法构建出的图像特征,明显增加了类间方差;“相对颜色特征”则减小了光照对样本的影响,从而减小了类内方差。(3)通过将RBF-SVM与BP神经网络方法相结合,对比发现,使用一种方法进行三分类的精度较低,分别为74.44%和81.11%;两次二分类后提高了精度,分别为87.77%和85.56%;而两种方法相结合的方式所得到的分类精度最高,总体可达91.11%。结论本文通过数字图像对受咖啡豹蠹蛾危害的檀香树干区域进行分类,为该虫害的早期识别以及受损率计算提供了方法,可以有效地减小檀香生长所受危害,提高心材质量,也为数字图像技术在林业中的应用提供了参考。

     

  • 图  1  3种区域图像样本

    Figure  1.  Three kinds of regional image samples

    图  2  檀香图像分割结果

    Figure  2.  Segmentation result of sandalwood

    图  3  3种区域的纹理特征值分布图

    Figure  3.  Texture feature value distribution of three kinds of regions

    表  1  单纹理与多纹理特征分类精度比较

    Table  1.   Comparison in classification accuracy using single texture features and multi-texture features

    单纹理特征
    Single texture feature
    总体分类精度
    Overall classification accuracy/%
    多纹理特征
    Multi-texture feature
    总体分类精度
    Overall classification accuracy/%
    能量均值
    Energy mean value
    58能量均值-熵值均值
    Energy mean value-entropy mean value
    78
    能量方差
    Energy variance
    58能量方差-熵值均值
    Energy variance-entropy mean value
    80
    熵值均值
    Entropy mean value
    80能量方差-熵值方差
    Energy variance-entropy variance
    72
    熵值方差
    Entropy variance
    74能量方差-对比度均值
    Energy variance-contrast mean value
    72
    对比度均值
    Contrast mean value
    72能量方差-对比度方差
    Energy variance-contrast variance
    78
    对比度方差
    Contrast variance
    76相关性均值-熵值均值
    Correlation mean value-entropy mean value
    72
    相关性均值
    Correlation mean value
    62相关性方差-熵值均值
    Correlation variance-entropy mean value
    80
    相关性方差
    Correlation variance
    54对比度方差-熵值均值
    Contrast variance-entropy mean value
    78
    下载: 导出CSV

    表  2  R通道原始颜色特征与相对颜色特征分类精度比较

    Table  2.   Classification accuracy compare using original R channel color features and relative color features

    %
    颜色特征
    Color feature
    原始数据Original data相对颜色数据Relative color data
    排泄区域
    Excrement region
    健康区域
    Healthy region
    排泄区域
    Excrement region
    健康区域
    Healthy region
    均值Mean value84949898
    X轴最大值Maximal value of X axis568244100
    X轴最小值Minimum value of X axis100384652
    X轴差值Difference of X axis62407084
    Y轴峰值Peak value of Y axis72268698
    信息熵Information entropy605890100
    下载: 导出CSV

    表  3  使用RBF-SVM和BP神经网络对PCA处理纹理以及多纹理分类结果比较

    Table  3.   Classification accuracy compare of texture by PCA processing and multi-texture features using RBF-SVM and BP neural network

    %
    分类方法
    Classification method
    PCA处理后纹理分类精度
    Classification accuracy by texture features
    after PCA processing
    多纹理分类精度
    Classification accuracy by multi-texture feature
    排泄物区域
    Excrement
    region
    虫害区域
    Pest region
    健康区域
    Healthy
    region
    总体
    Total
    排泄物区域
    Excrement
    region
    虫害区域
    Pest region
    健康区域
    Healthy
    region
    总体
    Total
    RBF-SVM76.0063.3370.0070.0093.3380.0093.3388.89
    BP神经网络BP neural network80.0070.0080.0076.6796.6786.6793.3392.22
    下载: 导出CSV

    表  4  使用RBF-SVM和BP神经网络对PCA处理颜色以及相对颜色特征的分类结果比较

    Table  4.   Classification accuracy compare of color feature by PCA processing and relative color features using RBF-SVM and BP neural network

    %
    分类方法
    Classification method
    颜色特征分类精度
    Classification accuracy of color feature
    相对颜色特征分类精度
    Classification accuracy of relative color feature
    排泄物区域
    Excrement
    region
    虫害区域
    Pest region
    健康区域
    Healthy
    region
    总体
    Total
    排泄物区域
    Excrement
    region
    虫害区域
    Pest region
    健康区域
    Healthy
    region
    总体
    Total
    RBF-SVM70.0063.3370.0067.7893.3380.0090.0087.77
    BP神经网络BP neural network66.6770.0066.6767.7886.6786.6783.3385.56
    下载: 导出CSV

    表  5  不同方法、不同分类次数之间的分类精度比较

    Table  5.   Classification accuracy compare of different methods and classification number

    %
    分类方法
    Classification method
    排泄物区域
    Excrement region
    虫害区域
    Pest region
    健康区域
    Health region
    总体Total
    RBF-SVM三分类 RBF-SVMthree classes80.0063.3380.0074.44
    BP神经网络三分类 BP neural networkthree classes83.3373.3380.0081.11
    本文提出方法Method proposed in this paper93.3386.6793.3391.11
    下载: 导出CSV
  • [1] 张宁南, 徐大平, 王卫文, 等.檀香栽培现状与发展前景[J].林业实用技术, 2006(10): 14-16. http://d.old.wanfangdata.com.cn/Periodical/lysyjs200610005

    Zhang N N, Xu D P, Wang W W, et al. Present situation and development prospect of sandalwood cultivation[J]. Practical Foreatry Technology, 2006(10):14-16. http://d.old.wanfangdata.com.cn/Periodical/lysyjs200610005
    [2] 唐黎标.檀香树探秘[J].云南林业, 2010(4):49. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=ynly201004038

    Tang L B. Explore the sandalwood[J]. Forestry of Yunnan, 2010(4):49. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=ynly201004038
    [3] 吴智仁.檀香重要害虫咖啡木蠹蛾防治方法[J].福建农业, 2012(5):23. doi: 10.3969/j.issn.1671-6922.2012.05.006

    Wu Z R.Methods for the control of Zeuzera coffeae nietner[J]. Agriculture of Fujian, 2012(5):23. doi: 10.3969/j.issn.1671-6922.2012.05.006
    [4] 汪京京, 张武, 刘连忠, 等.农作物病虫害图像识别技术的研究综述[J].计算机工程与科学, 2014, 36(7):1363-1370. doi: 10.3969/j.issn.1007-130X.2014.07.026

    Wang J J, Zhang W, Liu L Z, et al. Summary of crop diseases and pests image recognition technology[J]. Computer Engineering & Science, 2014, 36(7):1363-1370. doi: 10.3969/j.issn.1007-130X.2014.07.026
    [5] 温芝元, 曹乐平.基于为害状色相多重分形的椪柑病虫害图像识别[J].农业机械学报, 2014, 45(3):262-267. http://d.old.wanfangdata.com.cn/Periodical/nyjxxb201403043

    Wen Z Y, Cao L P. Damage pattern recognition of Citrus reticulate Blanco based on multi-fractal analysis of image Hue[J]. Transactions of The Chinese Society of Agricultural Machinery, 2014, 45(3):262-267. http://d.old.wanfangdata.com.cn/Periodical/nyjxxb201403043
    [6] Xia C, Chon T S, Ren Z. Automatic identification and counting of small size pests in greenhouse conditions with low computational cost[J]. Ecological Informatics, 2015, 29(9):139-146. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=554dd7546a45a782f8e50cdd999f79f4
    [7] Sasaki Y, Okamoto T, Imou K, et al. Automatic diagnosis of plant disease-recognition between health and diseased leaf[J]. Journal of the Japanese Society of Agricultural Machinery, 1999, 61(2):119-126.
    [8] El-Helly M, El-Beltagy S, Rafea A. Image analysis based interface for diagnostic expert systems[C]//Proceedings of the Winter International Symposium on Information and Communication Technologies. Dublin: Trinity College Dublin, 2004: 1-6.
    [9] Huang K Y. Application of artificial neural network for detecting Phalaenopsis seedling diseased using color and texture features[J]. Computer and Electronics in Agriculture. 2007, 57(1):3-11. http://www.sciencedirect.com/science/article/pii/S0168169907000385
    [10] 唐建军, 王映龙, 彭莹琼, 等. BP神经网络在水稻病虫害诊断中的应用研究[J].安徽农业科学, 2010, 38(1): 199-200. doi: 10.3969/j.issn.0517-6611.2010.01.085

    Tang J J, Wang Y L, Peng Y Q, et al. Application study on BP neural network in the diagnosis of rice diseases and pests[J]. Journal of Anhui Agricultural Sciences, 2010, 38(1):199-200. doi: 10.3969/j.issn.0517-6611.2010.01.085
    [11] Zhang R, Ma J W. State of the art on remotely sensed data classification based on support vector machine[J]. Advances in Earth Science, 2009, 24(5):555-562. http://en.cnki.com.cn/Article_en/CJFDTOTAL-DXJZ200905018.htm
    [12] Song X F, Duan Z, Jiang X G. Comparison of artificial neural networks and support vector machine classifiers for land cover classification in northern China using a SPOT-5 HRG image[J]. International Journal of Remote Sensing, 2012, 33(10):3301-3320. doi: 10.1080/01431161.2011.568531
    [13] Huang C, Davis L S, Townshend J R G. An assessment of support vector machines for land cover classification[J]. International Journal of Remote Sensing, 2002, 23(4): 725-749. doi: 10.1080/01431160110040323
    [14] 张建华, 祁力钧, 冀荣华, 等.基于Gabor小波和颜色矩的棉花盲椿象分类方法[J].农业工程学报, 2012, 28(1): 133-138. doi: 10.3969/j.issn.1002-6819.2012.01.025

    Zhang J H, Qi L J, Ji R H, et al. Classification of cotton blind stinkbug based on Gabor wavelet and color moments[J]. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(1): 133-138. doi: 10.3969/j.issn.1002-6819.2012.01.025
    [15] 田有文, 王立地, 姜淑华, 等.基于图像处理和支持向量机的玉米病害识别[J].仪器仪表学报, 2006, 27(6): 2123-2124. http://d.old.wanfangdata.com.cn/Periodical/yqyb2006z3141

    Tian Y W, Wang L D, Jiang S H. Recognition of maize disease based on image processing and SVM[J]. Chinese Journal of Scientific Instrument, 2006, 27(6): 2123-2124. http://d.old.wanfangdata.com.cn/Periodical/yqyb2006z3141
    [16] 王献锋, 王震, 王旭启, 等.基于SVM和形态学的作物病斑图像分割方法[J].东北农业科学, 2015, 40(1): 51-60. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jlnykx201501012

    Wang X F, Wang Z, Wang X Q, et al. Crop spot image segmentation based on SVM and morphology[J]. Journal of Jilin Agricultural Sciences, 2015, 40(1):51-53. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jlnykx201501012
    [17] Burgos-Artizzu X P, Ribeiro A, Guijarro M, et al. Real-time image processing for crop/weed discrimination in maize fields[J]. Computers and Electronics in Agriculture, 2010, 75(2): 337-346. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=aa9f8d1c0e813a0014a112676b733123
    [18] Aitkenhead M J, Dalgetty I A, Mullins C E. Weed and crop discrimination using image analysis and artificial intelligence methods[J]. Computers and Electronics in Agriculture, 2003, 39(3): 157-171. doi: 10.1016/S0168-1699(03)00076-0
    [19] 赵进辉, 罗锡文, 周志艳.基于颜色与形状特征的甘蔗病害图像分割方法[J].农业机械学报, 2008, 39(9): 100-103. http://d.old.wanfangdata.com.cn/Periodical/nyjxxb200809023

    Zhao J H, Luo X W, Zhou Z Y. Image segmentation method for sugarcane diseases based on color and shape features[J]. Transactions of the Chinese Society of Agricultural Machinery, 2008, 39(3):100-103. http://d.old.wanfangdata.com.cn/Periodical/nyjxxb200809023
    [20] 陈佳娟, 纪寿文, 李娟, 等.采用计算机视觉进行棉花虫害程度的自动测定[J].农业工程学报, 2001, 17(2): 157-160. doi: 10.3321/j.issn:1002-6819.2001.02.040

    Chen J J, Ji S W, Li J, et al. Automatic measurement of danger degree of cotton insect pests using computer vision[J]. Transactions of the Chinese Society of Agricultural Engineering, 2001, 17(2):157-160. doi: 10.3321/j.issn:1002-6819.2001.02.040
    [21] 王雪峰, 毕于慧, 罗景芳.银中杨苗木光需求的图像分析法[J].林业科学, 2010, 46(4): 161-165. http://d.old.wanfangdata.com.cn/Periodical/lykx201004025

    Wang X F, Bi Y H, Luo J F. Image analysis method of light need of the Populus alba×P. berolinensis[J]. Scientia Silvae Sinicae, 2010, 46(4):161-165. http://d.old.wanfangdata.com.cn/Periodical/lykx201004025
    [22] Ulaby F T, Kouyate F, Brisco B. Texture information on SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 1986, 24(2):235-245. http://d.old.wanfangdata.com.cn/Periodical/zgkydxxb200903023
    [23] 江虹, 曾立波, 胡继明.优化的BP神经网络分类器的设计与实现[J].计算机工程与应用, 2001, 37(5): 122-125. doi: 10.3321/j.issn:1002-8331.2001.05.042

    Jiang H, Zeng L B, Hu J M. The design and performance of optimal BP neural network classifier[J]. Computer Engineering and Application, 2001, 37(5): 122-125. doi: 10.3321/j.issn:1002-8331.2001.05.042
    [24] 张伐伐, 李卫忠, 卢柳叶, 等. SVM多窗口纹理土地利用信息提取技术[J].遥感学报, 2012, 16(1): 67-78. http://d.old.wanfangdata.com.cn/Periodical/ygxb201201005

    Zhang F F, Li W Z, Lu L Y, et al. Technologies of extracting land utilization information based on SVM method with multi-window texture[J]. Journal of Remote Sensing, 2012, 16(1): 67-78. http://d.old.wanfangdata.com.cn/Periodical/ygxb201201005
  • 加载中
图(3) / 表(5)
计量
  • 文章访问数:  1974
  • HTML全文浏览量:  168
  • PDF下载量:  70
  • 被引次数: 0
出版历程
  • 收稿日期:  2017-09-04
  • 修回日期:  2017-12-05
  • 刊出日期:  2018-01-01

目录

    /

    返回文章
    返回