基于高光谱成像的柑橘黄龙病无损检测
|
摘要
采用高光谱成像技术,结合最小二乘支持向量机(LS-SVM)和偏最小二乘判别分析(PLS-DA)2种方法,探索柑橘黄龙病快速无损检测的可行性。在380~1 080 nm光谱范围内,采集正常、轻度黄龙病、中度黄龙病、重度黄龙病和缺素5种柑橘叶片的高光谱图像。采用方差分析方法,分析了正常、轻度黄龙病、中度黄龙病、重度黄龙病和缺素5种叶片的叶绿素、淀粉和可溶性糖含量间的差异,表明3指标可作为判别黄龙病的指示性指标。采用偏最小二乘法,建立了叶绿素、可溶性糖及淀粉3指标含量的定量分析数学模型,模型预测均方根误差分别为7.46、5.51、5.88,提供了柑橘黄龙病高光谱成像快速检测依据。提取高光谱图像感兴趣区域的平均光谱,通过分析正常、轻度黄龙病、中度黄龙病、重度黄龙病和缺素5种叶片的代表性光谱,在750 nm处吸光度存在差异。采用2阶导数处理样品光谱,消除了450~650 nm和800~1 000 nm波段的基线漂移,放大了有效光谱信息。采用主成分分析(PCA)和连续投影算法(SPA)筛选柑橘黄龙病LS-SVM定性判别模型的输入变量,建立了LS-SVM定性判别模型,同时与PLS-DA进行对比。采用未参与建模的预测集样品评价模型性能,结果表明PLS-DA模型判别柑橘黄龙病的准确率更高,模型误判率为5.6%。实验结果表明,高光谱成像技术结合偏最小二乘判别分析方法可实现柑橘黄龙病快速无损检测与黄龙病病情等级判别。
In order to explore the feasibility of the quick non-destructive detection of citrus Huanglong disease,the hyperspectral image technique combined with least square support vector machine( LSSVM) and partial least squares discriminate analysis( PLS-DA) were used. The hyperspectral images of the normal,the Huanglong disease of slight,moderate and serious,the lack element citrus leaves were collected in wavelength range of 380 ~ 1 080 nm. By using variance analysis method,the differences in content of chlorophyll,soluble sugar and starch of leaves of the normal,the Huanglong disease of slight,moderate,serious and the lack element were analyzed,and the chlorophyll,soluble sugar and starch were the indicator which could be used to discriminate Huanglong disease. The partial least squares( PLS) method was adopted to establish the mathematical model of quantitative analysis of chlorophyll,soluble sugar and starch, and root mean square error of forecast model were 7. 46,5. 51,5. 88 respectively,which provided the basis for rapid detection of citrus Huanglong disease hyperspectral images. The average spectrum of hyperspectral images was extracted in interested area. The differences in absorbance at 750 nm was found by analyzing five kinds of leaves of representative spectrum of thenormal,the Huanglong disease of slight,moderate and serious,the lack element. The 2-order derivative was used to process the sample spectrum,the baseline drift in 450 ~ 650 nm and 800 ~ 1 000 nm band was eliminated and the effective spectral information was enlarged. Using principal component analysis( PCA) and successive projections algorithm( SPA) to screen the input variables of the model of least squares support vector machine( LS-SVM) qualitative discrimination of citrus Huanglong disease,the LS-SVM model was built for qualitative discrimination and compared with the partial least squares qualitative discriminate model( PLS-DA) at the same time. The prediction sample set which was used to evaluate the performance of model was not used to establish the model. The results showed that the accuracy of PLS-DA model of citrus Huanglong disease was higher,three leaves of lack element were misclassified as serious Huanglong disease,and the misclassification rate was 5. 6%. The experimental results showed that the hyperspectral image technology combined with PLS-DA can achieve rapid and nondestructive detection of citrus Huanglong disease and the degree of Huanglong disease.
In order to explore the feasibility of the quick non-destructive detection of citrus Huanglong disease,the hyperspectral image technique combined with least square support vector machine( LSSVM) and partial least squares discriminate analysis( PLS-DA) were used. The hyperspectral images of the normal,the Huanglong disease of slight,moderate and serious,the lack element citrus leaves were collected in wavelength range of 380 ~ 1 080 nm. By using variance analysis method,the differences in content of chlorophyll,soluble sugar and starch of leaves of the normal,the Huanglong disease of slight,moderate,serious and the lack element were analyzed,and the chlorophyll,soluble sugar and starch were the indicator which could be used to discriminate Huanglong disease. The partial least squares( PLS) method was adopted to establish the mathematical model of quantitative analysis of chlorophyll,soluble sugar and starch, and root mean square error of forecast model were 7. 46,5. 51,5. 88 respectively,which provided the basis for rapid detection of citrus Huanglong disease hyperspectral images. The average spectrum of hyperspectral images was extracted in interested area. The differences in absorbance at 750 nm was found by analyzing five kinds of leaves of representative spectrum of thenormal,the Huanglong disease of slight,moderate and serious,the lack element. The 2-order derivative was used to process the sample spectrum,the baseline drift in 450 ~ 650 nm and 800 ~ 1 000 nm band was eliminated and the effective spectral information was enlarged. Using principal component analysis( PCA) and successive projections algorithm( SPA) to screen the input variables of the model of least squares support vector machine( LS-SVM) qualitative discrimination of citrus Huanglong disease,the LS-SVM model was built for qualitative discrimination and compared with the partial least squares qualitative discriminate model( PLS-DA) at the same time. The prediction sample set which was used to evaluate the performance of model was not used to establish the model. The results showed that the accuracy of PLS-DA model of citrus Huanglong disease was higher,three leaves of lack element were misclassified as serious Huanglong disease,and the misclassification rate was 5. 6%. The experimental results showed that the hyperspectral image technology combined with PLS-DA can achieve rapid and nondestructive detection of citrus Huanglong disease and the degree of Huanglong disease.
引文
1陈凯男,李充璧.柑橘黄龙病研究概况[J].浙江农业科学,2015,56(7):1048-1050.CHEN Kainan,LI Chongbi.Citrus Huanglongbing research situation[J].Zhejiang Agricultural Science,2015,56(7):1048-1050.(in Chinese)
2 梅慧兰,邓小玲,洪添胜,等.柑橘黄龙病高光谱早期鉴别及病情分级[J].农业工程学报,2014,30(9):140-148.MEI Huilan,DENG Xiaoling,HONG Tiansheng,et al.Early detection and grading of citrus Huanglongbing using hyperspectral imaging technique[J].Transactions of the CSAE,2014,30(9):140-148.(in Chinese)
3 单振菊,冯震,周根,等.南方5省区柑橘黄龙病病原16Sr DNA片段的克隆与序列分析[J].华南农业大学学报,2008,29(2):25-30.SHAN Zhenju,FENG Zhen,ZHOU Gen,et al.Cloning and sequence analysis of 16Sr DNA of citrus Huanglongbing agents collected from five provinces in South China[J].Journal of South China Agricultural University,2008,29(2):25-30.(in Chinese)
4 YU Jihua,ZHANG Minrong,TAO Jian,et al.The control research progress of liberobacter asiaticum in Taizhou city[J].Agricultural Science&Technology,2015,16(6):1237-1242.
5 王爱民,邓晓玲.柑橘黄龙病诊断技术研究进展[J].广东农业科学,2008,6(1):101-104.WANG Aimin,DENG Xiaoling.The research on the diagnosis technology of citrus Huanglongbing[J].Guangdong Agricultural Sciences,2008,6(1):101-104.(in Chinese)
6 马淏,吉海彦,WON Suk Lee.基于Vis-NIR光谱的柑橘叶片黄龙病检测及其光谱特性研究[J].光谱学与光谱分析,2014,34(10):2713-2718.MA Hao,JI Haiyan,WON Suk Lee.Detection of citrus greening based on Vis-NIR spectroscopy and spectral feature analysis[J].Spectroscopy and Spectral Analysis,2014,34(10):2713-2718.(in Chinese)
7 付惠敏,陈传武,白先进.有百年历史的日益严重的毁灭性柑橘病害—黄龙病[J].广西农学报,2011,26(4):70-78.FU Huimin,CHEN Chuanwu,BAI Xianjin.The growing devastating citrus diseases have one hundred year old historyHuanglongbing[J].Journal of Guangxi Agriculture,2011,26(4):70-78.(in Chinese)
8 DENG Xiaoling,TANG Weiwen.Application of polymerase chain reaction to the detection of citrus Huanglongbing pathogen[J].Journal of Zhejiang Agricultural University,1998,24(5):557-562.
9 穆晓琨,高飞云,李充璧.黄龙病柑橘内生菌16Sr DNA的提取及分析[J].农业与技术,2015,35(5):5-8.MU Xiaogun,GAO Feiyun,LI Chongbi.Citrus Huanglongbing endophytes 16 sr-DNA extraction and analysis[J].Agriculture and Technology,2015,35(5):5-8.(in Chinese)
10 王志辉,丁丽霞.基于叶片高光谱特性分析的树种识别[J].光谱学与光谱分析,2010,30(7):1825-1829.WANG Zhihui,DING Lixia.Tree species discrimination based on leaf-level hyperspectral characteristic analysis[J].Spectroscopy and Spectral,2010,30(7):1825-1829.(in Chinese)
11 梁爽,赵庚星,朱西存.苹果树叶片叶绿素含量高光谱估测模型研究[J].光谱学与光谱分析,2012,32(5):1367-1370.LIANG Shuang,ZHAO Gengxing,ZHU Xicun.Hyperspectral estimation models of chlorophyll content in apple lesves[J].Spectroscopy and Spectral,2012,32(5):1367-1370.(in Chinese)
12 岳学军,全东平,洪添胜,等.柑橘叶片叶绿素含量高光谱无损检测模型[J].农业工程学报,2015,31(1):294-303.YUE Xuejun,QUAN Dongping,HONG Tiansheng,et al.Non-destructive hyperspectral measurement model of chlorophyll content for citrus leaves[J].Transactions of The CSAE,2015,31(1):294-303.(in Chinese)
13 邓小玲,郑建宝,梅慧兰,等.基于高光谱成像技术的柑橘黄龙病病情诊断及分类[J].西北农林科技大学学报:自然科学版,2013,41(7):99-106.DENG Xiaoling,ZHENG Jianbao,MEI Huilan,et al.Identification and classification of citrus Huanglongbing disease based on hyperspectral imaging[J].Journal of Northwest A&F University:Natural Science Edition,2013,41(7):99-106.(in Chinese)
14 SINDHUJA Sankaran,REZA Ehsani.Visible-near infrared spectroscopy based citrus greening detection:evaluation of spectral feature extraction techniques[J].Crop Protection,2011,30(11):1508-1513.
15 ZHANG Baohua,LIA Jiangbo,FAN Shuxiang,et al.Hyperspectral imaging combined with multivariate analysis and band math for detection of common defects on peaches(prunus persica)[J].Computers and Electronics in Agriculture,2015,114(5):14-24.
16 CEN Haiyan,LU Renfu,ZHU Qibing,et al.Nondestructive detection of chilling injury in cucumber fruit using hyperspectral imaging with feature selection and supervised classification[J].Postharvest Biology and Technology,2016,111(2):352-361.
17 POURREZA A,LEE W S,ETXEBERRIA E,et al.An evaluation of a vision-based sensor performance in Huanglongbing disease identification[J].Biosystems Engineering,2015,130:13-22.
18 李修华,李民赞,WON SUK Lee,等.柑桔黄龙病的可见-近红外光谱特征[J].光谱学与光谱分析,2014,34(6):1553-1559.LI Xiuhua,LI Minzhan,WON SUK Lee,et al.Visible-NIR spectral feature of citrus greening disease[J].Spectroscopy and Spectral Analysis,2014,34(6):1553-1559.(in Chinese)
19 邓晓玲,唐伟文.应用PCR技术检测柑桔黄龙病病原的研究[J].华南农业大学学报,1996,17(3):119-120.DENG Xiaoling,TANG Weiwen.The studies on detection of citrus Huanglongbing pathogen by polymerase chain reaction[J].Journal of South China Agricultural University,1996,17(3):119-120.(in Chinese)
20 袁亦文,蒋自珍,王德善.柑橘黄龙病病情分级标准探讨[J].浙江农业科学,2010(1):121-123.YUAN Yiwen,JIANG Zhizhen,WANG Deshan.Study on classification standard of Citrus Huanglongbing disease[J].Zhenjiang Agricultural Science,2010(1):121-123.(in Chinese)
21 李江波,饶秀勤,应义斌,等.基于高光谱成像技术检测脐橙溃疡[J].农业工程学报,2010,26(8):222-231.LI Jiangbo,RAO Xiuqin,YING Yibin,et al.Detection of navel oranges canker based on hyperspectral imaging technology[J].Transactions of The CSAE,2010,26(8):222-231.(in Chinese)
22 刘燕德,邓清.基于高光谱成像技术的脐橙叶片的叶绿素的含量及其分布测量[J].发光学报,2015,36(8):30-35.LIU Yande,DENG Qing.Measurement of chlorophyll distribution in navel orange leaves based on hyperspectral imaging technique[J].Chinese Journal of Luminescence,2015,36(8):30-35.(in Chinese)
23 黄双萍,洪添胜,岳学军,等.基于高光谱的柑橘叶片磷含量估算模型实验[J].农业机械学报,2013,44(4):202-207,195.HUANG Shuangping,HONG Tiansheng,YUE Xuejun,et al.Hyperspectral estimation model of total phosphorus content for citrus leaves[J].Transactions of the Chinese Society for Agricultural Machinery,2013,44(4):202-207,195.(in Chinese)
24 刘雪梅,章海亮.基于DPLS和LS-SVM的梨品种近红外光谱识别[J].农业机械学报,2012,43(9):160-164.LIU Xuemei,ZHANG Hailiang.Identification of varieties of pear using near infrared spectra based on DPLS and LS-SVM model[J].Transactions of the Chinese Society for Agricultural Machinery,2012,43(9):160-164.(in Chinese)
25 迟茜,王转卫,杨婷婷,等.基于近红外高光谱成像的猕猴桃早期隐性损伤识别[J].农业机械学报,2015,46(3):235-241.CHI Qian,WANG Zhuanwei,YANG Tingting,et al.Recognition of early hidden bruises on kiwifruits based on near-infrared hyperspectral imaging technology[J].Transactions of the Chinese Society for Agricultural Machinery,2015,46(3):235-241.(in Chinese)
26 高洪智,卢启鹏,丁海泉,等.基于连续投影算法的土壤总氮近红外特征波长的选取[J].光谱学与光谱分析,2009,29(11):2951-2955.GAO Hongzhi,LU Qipeng,DING Haiquan,et al.Choice of characteristic near-infrared wavelengths for soil total nitrogen based on successive projection algorithm[J].Spectroscopy and Spectral,2009,29(11):2951-2955.(in Chinese)
2 梅慧兰,邓小玲,洪添胜,等.柑橘黄龙病高光谱早期鉴别及病情分级[J].农业工程学报,2014,30(9):140-148.MEI Huilan,DENG Xiaoling,HONG Tiansheng,et al.Early detection and grading of citrus Huanglongbing using hyperspectral imaging technique[J].Transactions of the CSAE,2014,30(9):140-148.(in Chinese)
3 单振菊,冯震,周根,等.南方5省区柑橘黄龙病病原16Sr DNA片段的克隆与序列分析[J].华南农业大学学报,2008,29(2):25-30.SHAN Zhenju,FENG Zhen,ZHOU Gen,et al.Cloning and sequence analysis of 16Sr DNA of citrus Huanglongbing agents collected from five provinces in South China[J].Journal of South China Agricultural University,2008,29(2):25-30.(in Chinese)
4 YU Jihua,ZHANG Minrong,TAO Jian,et al.The control research progress of liberobacter asiaticum in Taizhou city[J].Agricultural Science&Technology,2015,16(6):1237-1242.
5 王爱民,邓晓玲.柑橘黄龙病诊断技术研究进展[J].广东农业科学,2008,6(1):101-104.WANG Aimin,DENG Xiaoling.The research on the diagnosis technology of citrus Huanglongbing[J].Guangdong Agricultural Sciences,2008,6(1):101-104.(in Chinese)
6 马淏,吉海彦,WON Suk Lee.基于Vis-NIR光谱的柑橘叶片黄龙病检测及其光谱特性研究[J].光谱学与光谱分析,2014,34(10):2713-2718.MA Hao,JI Haiyan,WON Suk Lee.Detection of citrus greening based on Vis-NIR spectroscopy and spectral feature analysis[J].Spectroscopy and Spectral Analysis,2014,34(10):2713-2718.(in Chinese)
7 付惠敏,陈传武,白先进.有百年历史的日益严重的毁灭性柑橘病害—黄龙病[J].广西农学报,2011,26(4):70-78.FU Huimin,CHEN Chuanwu,BAI Xianjin.The growing devastating citrus diseases have one hundred year old historyHuanglongbing[J].Journal of Guangxi Agriculture,2011,26(4):70-78.(in Chinese)
8 DENG Xiaoling,TANG Weiwen.Application of polymerase chain reaction to the detection of citrus Huanglongbing pathogen[J].Journal of Zhejiang Agricultural University,1998,24(5):557-562.
9 穆晓琨,高飞云,李充璧.黄龙病柑橘内生菌16Sr DNA的提取及分析[J].农业与技术,2015,35(5):5-8.MU Xiaogun,GAO Feiyun,LI Chongbi.Citrus Huanglongbing endophytes 16 sr-DNA extraction and analysis[J].Agriculture and Technology,2015,35(5):5-8.(in Chinese)
10 王志辉,丁丽霞.基于叶片高光谱特性分析的树种识别[J].光谱学与光谱分析,2010,30(7):1825-1829.WANG Zhihui,DING Lixia.Tree species discrimination based on leaf-level hyperspectral characteristic analysis[J].Spectroscopy and Spectral,2010,30(7):1825-1829.(in Chinese)
11 梁爽,赵庚星,朱西存.苹果树叶片叶绿素含量高光谱估测模型研究[J].光谱学与光谱分析,2012,32(5):1367-1370.LIANG Shuang,ZHAO Gengxing,ZHU Xicun.Hyperspectral estimation models of chlorophyll content in apple lesves[J].Spectroscopy and Spectral,2012,32(5):1367-1370.(in Chinese)
12 岳学军,全东平,洪添胜,等.柑橘叶片叶绿素含量高光谱无损检测模型[J].农业工程学报,2015,31(1):294-303.YUE Xuejun,QUAN Dongping,HONG Tiansheng,et al.Non-destructive hyperspectral measurement model of chlorophyll content for citrus leaves[J].Transactions of The CSAE,2015,31(1):294-303.(in Chinese)
13 邓小玲,郑建宝,梅慧兰,等.基于高光谱成像技术的柑橘黄龙病病情诊断及分类[J].西北农林科技大学学报:自然科学版,2013,41(7):99-106.DENG Xiaoling,ZHENG Jianbao,MEI Huilan,et al.Identification and classification of citrus Huanglongbing disease based on hyperspectral imaging[J].Journal of Northwest A&F University:Natural Science Edition,2013,41(7):99-106.(in Chinese)
14 SINDHUJA Sankaran,REZA Ehsani.Visible-near infrared spectroscopy based citrus greening detection:evaluation of spectral feature extraction techniques[J].Crop Protection,2011,30(11):1508-1513.
15 ZHANG Baohua,LIA Jiangbo,FAN Shuxiang,et al.Hyperspectral imaging combined with multivariate analysis and band math for detection of common defects on peaches(prunus persica)[J].Computers and Electronics in Agriculture,2015,114(5):14-24.
16 CEN Haiyan,LU Renfu,ZHU Qibing,et al.Nondestructive detection of chilling injury in cucumber fruit using hyperspectral imaging with feature selection and supervised classification[J].Postharvest Biology and Technology,2016,111(2):352-361.
17 POURREZA A,LEE W S,ETXEBERRIA E,et al.An evaluation of a vision-based sensor performance in Huanglongbing disease identification[J].Biosystems Engineering,2015,130:13-22.
18 李修华,李民赞,WON SUK Lee,等.柑桔黄龙病的可见-近红外光谱特征[J].光谱学与光谱分析,2014,34(6):1553-1559.LI Xiuhua,LI Minzhan,WON SUK Lee,et al.Visible-NIR spectral feature of citrus greening disease[J].Spectroscopy and Spectral Analysis,2014,34(6):1553-1559.(in Chinese)
19 邓晓玲,唐伟文.应用PCR技术检测柑桔黄龙病病原的研究[J].华南农业大学学报,1996,17(3):119-120.DENG Xiaoling,TANG Weiwen.The studies on detection of citrus Huanglongbing pathogen by polymerase chain reaction[J].Journal of South China Agricultural University,1996,17(3):119-120.(in Chinese)
20 袁亦文,蒋自珍,王德善.柑橘黄龙病病情分级标准探讨[J].浙江农业科学,2010(1):121-123.YUAN Yiwen,JIANG Zhizhen,WANG Deshan.Study on classification standard of Citrus Huanglongbing disease[J].Zhenjiang Agricultural Science,2010(1):121-123.(in Chinese)
21 李江波,饶秀勤,应义斌,等.基于高光谱成像技术检测脐橙溃疡[J].农业工程学报,2010,26(8):222-231.LI Jiangbo,RAO Xiuqin,YING Yibin,et al.Detection of navel oranges canker based on hyperspectral imaging technology[J].Transactions of The CSAE,2010,26(8):222-231.(in Chinese)
22 刘燕德,邓清.基于高光谱成像技术的脐橙叶片的叶绿素的含量及其分布测量[J].发光学报,2015,36(8):30-35.LIU Yande,DENG Qing.Measurement of chlorophyll distribution in navel orange leaves based on hyperspectral imaging technique[J].Chinese Journal of Luminescence,2015,36(8):30-35.(in Chinese)
23 黄双萍,洪添胜,岳学军,等.基于高光谱的柑橘叶片磷含量估算模型实验[J].农业机械学报,2013,44(4):202-207,195.HUANG Shuangping,HONG Tiansheng,YUE Xuejun,et al.Hyperspectral estimation model of total phosphorus content for citrus leaves[J].Transactions of the Chinese Society for Agricultural Machinery,2013,44(4):202-207,195.(in Chinese)
24 刘雪梅,章海亮.基于DPLS和LS-SVM的梨品种近红外光谱识别[J].农业机械学报,2012,43(9):160-164.LIU Xuemei,ZHANG Hailiang.Identification of varieties of pear using near infrared spectra based on DPLS and LS-SVM model[J].Transactions of the Chinese Society for Agricultural Machinery,2012,43(9):160-164.(in Chinese)
25 迟茜,王转卫,杨婷婷,等.基于近红外高光谱成像的猕猴桃早期隐性损伤识别[J].农业机械学报,2015,46(3):235-241.CHI Qian,WANG Zhuanwei,YANG Tingting,et al.Recognition of early hidden bruises on kiwifruits based on near-infrared hyperspectral imaging technology[J].Transactions of the Chinese Society for Agricultural Machinery,2015,46(3):235-241.(in Chinese)
26 高洪智,卢启鹏,丁海泉,等.基于连续投影算法的土壤总氮近红外特征波长的选取[J].光谱学与光谱分析,2009,29(11):2951-2955.GAO Hongzhi,LU Qipeng,DING Haiquan,et al.Choice of characteristic near-infrared wavelengths for soil total nitrogen based on successive projection algorithm[J].Spectroscopy and Spectral,2009,29(11):2951-2955.(in Chinese)