基于多光谱成像技术的小麦品种快速无损鉴定
|
摘要
为了研究多光谱成像技术对小麦品种快速无损鉴定的可行性,采用VideometerLab多光谱图像采集设备对5个小麦品种共500个样品在405～970 nm波段内的进行多光谱图像信息进行采集,获取其光谱、颜色和形态特征。利用主成分分析对5个小麦品种进行定性鉴别,同时,基于光谱特征和光谱图像特征分别比较了神经网络、支持向量机和随机森林3种模型的鉴定效果。结果显示:利用19个光谱特征值建立的模型中,BPNN识别模型效果最佳,其建模集和预测集的识别率分别为100%和91.25%。融合19个光谱特征和6个图像特征所建立的模型中,BPNN识别模型效果最佳,其建模集和预测集的识别率分别达到了100%和98.4%。结果表明,基于BPNN的多光谱特征融合能够有效的提高小麦品种鉴定效率,为小麦品种的快速无损检测提供了一个新途径。
To study the feasibility of multi-spectral imaging technology for rapid and non-destructive identification of wheat varieties, multi-spectral images that covered the range of 405-970 nm from 500 samples of 5 wheat varieties were collected to find out the specified spectral, color and morphological characteristics by using VideometerLab multiidentified by principal component analysis. Likewise, the recognition accuracy of 3 different models(neural network, support vector machine and random forest) was compared based on spectral features and spectral image features. The results showed that BPNN method had the best performance, which was 100% and 91.25%for the modeling set and prediction set respectively, when 19 spectral eigenvalues were employed. Moreover,the satisfactory recognition accuracy, which was 100% and 98.4% for the modeling set and prediction set respectively, was also achieved when 19 spectral features and 6 image features were integrated. It suggested that multi-spectral feature fusion based on BPNN can effectively improve the recognition accuracy of wheat varieties and provide a new way for rapid nondestructive detection of wheat varieties.
To study the feasibility of multi-spectral imaging technology for rapid and non-destructive identification of wheat varieties, multi-spectral images that covered the range of 405-970 nm from 500 samples of 5 wheat varieties were collected to find out the specified spectral, color and morphological characteristics by using VideometerLab multiidentified by principal component analysis. Likewise, the recognition accuracy of 3 different models(neural network, support vector machine and random forest) was compared based on spectral features and spectral image features. The results showed that BPNN method had the best performance, which was 100% and 91.25%for the modeling set and prediction set respectively, when 19 spectral eigenvalues were employed. Moreover,the satisfactory recognition accuracy, which was 100% and 98.4% for the modeling set and prediction set respectively, was also achieved when 19 spectral features and 6 image features were integrated. It suggested that multi-spectral feature fusion based on BPNN can effectively improve the recognition accuracy of wheat varieties and provide a new way for rapid nondestructive detection of wheat varieties.
引文
[1]钟永玲,曹慧.中国小麦产业发展现状、调控政策及建议[J].农学学报,2011,1(12):49-54.
[2]华国栋,刘耀鸿,李冠喜.小麦种子质量控制关键技术的初步研究[J].作物杂志,2012,(1):139-141.
[3]朱旭东,上官凌飞,孙欣,等.DNA标记在植物品种鉴定上的应用现状[J].中国农学通报,2014,30(30):234-240.
[4]郑永胜,张晗,王东建,等.基于荧光检测技术的小麦品种SSR鉴定体系的建立[J].中国农业科学,2014,47(19):3725-3735.
[5]吴则东,江伟,马龙彪.分子标记技术在农作物品种鉴定上的研究进展及未来展望[J].中国农学通报,2015,31(33):172-176.
[6]张雪原,赵攀峰,王凤格,等.玉米品种SSR分子标记与田间小区种植一致性鉴定结果的比较[J].玉米科学,2009,17(1):40-45,50.
[7]朱大洲,王坤,周光华,等.单粒大豆的近红外光谱特征及品种鉴别研究[J].光谱学与光谱分析,2010,30(12):3217-3221.
[8]刘天玲,苏琪雅,孙群,等.基于NIR分析和模式识别技术的玉米种子识别系统[J].光谱学与光谱分析,2012,32(6):1550-1553.
[9]黄敏,朱晓,朱启兵,等.基于高光谱图像的玉米种子特征提取与识别[J].光子学报,2012,41(7):868-873.
[10]邹伟,方慧,刘飞,等.基于高光谱图像技术的油菜籽品种鉴别方法研究[J].浙江大学学报:农业与生命科学版,2011,37(2):175-180.
[11]严衍禄.近红外光谱分析基础与应用[M].北京:中国轻工业出版社, 2005.
[12] Pua H, Kamruzzamanb M, Sun D W. Selection of feature wavelengths for developingmultispectral imaging systems for quality, safety and authenticity of muscle foods-areview[J]. Trends in Food Science and Technology, 2015,45:86-104.
[13] Xing J A, Symons S, Shahin M, et al. Detection of sprout damage in Canada Western RedSpring wheat with multiple wavebands using visible/near-infrared hyperspectral imaging[J].Biosystems Engineering, 2010,106(2):188-194.
[14] Wallays C, Missotten B, De Baerdemaeker J, et al. Hyperspectral waveband selection foron-line measurement of grain cleanness[J].Biosystems Engineering, 2009, 104(1):1-7.
[15]刘伟,刘长虹,郑磊.基于支持向量机的多光谱成像稻谷品种鉴定[J].农业工程学报,2014,30(10):145-151.
[16] Ma F, Wang J, Liu C, et al. Discrimination of kernel quality characteristics for sunflowerseeds based on multispectral imaging approach[J]. Food Analytical Methods,2015,8(7):1629-1636.
[17] Liu C, Liu W, Lu X, et al. Application of Multispectral Imaging to Determine QualityAttributes and Ripeness Stage in Strawberry Fruit[J]. Plos One, 2014, 9(2), e87818.
[18] Xiong C, Liu C, Pan W, et al. Non-destructive determination of total polyphenolscontent and classification of storage periods of Iron Buddha tea using multispectralimaging system[J]. Food Chemistry, 2015,176:130-136.
[19]翟俊海,赵文秀,王熙照.图像特征提取研究[J].河北大学学报:自然科学版, 2009,29(1):106-112.
[20] Ma F, Qin H, Shi K, et al. Feasibility of combining spectra with texture data ofmultispectral imaging to predict heme and non-heme iron contents in pork sausages[J].Food Chemistry, 2016, 190:142-149.
[21]张初,刘飞,孔汶汶,等.利用近红外高光谱图像技术快速鉴定西瓜种子品种[J].农业工程学报,2013,29(20):270-277.
[22]吴建国,石春海,张小明,等.用近红外反射光谱法分析稻米3种必需氨基酸含量的研究[J].作物学报,2003,29(05):688-692.
[23]杨小玲,由昭红,成芳.高光谱成像技术检测玉米种子成熟度[J].光谱学与光谱分析,2016,36(12):4028-4033.
[24]章毅,郭泉,王建勇.大数据分析的神经网络方法[J].工程科学与技术,2017,49(1):9-18.
[25]许思,赵光武,邓飞,等.基于高光谱的水稻种子活力无损分级检测[J].种子,2016,35(4):34-40.
[26]余鼎,程维民,王琦,等.基于近红外光谱技术的水稻种子成分分析模型的建立[J].光散射学报,2015,27(04):384-389.
[27]姚登举,杨静,詹晓娟.基于随机森林的特征选择算法[J].吉林大学学报:工学版,2014,44(01):137-141.
[28] Svante W. PLS-regression:a basic tool of chemometrics[J].Chemometrics&Intelligent Laboratory Systems,2001,58(2):109-130.
[29]邓小琴,朱启兵,黄敏.融合光谱、纹理及形态特征的水稻种子品种高光谱图像单粒鉴别[J].激光与光电子学进展,2015,52(02):128-134.
[30]黄艳艳,朱丽伟,李军会,等.应用近红外光谱技术快速鉴别玉米杂交种纯度的研究[J].光谱学与光谱分析,2011,31(03):661-664.
[2]华国栋,刘耀鸿,李冠喜.小麦种子质量控制关键技术的初步研究[J].作物杂志,2012,(1):139-141.
[3]朱旭东,上官凌飞,孙欣,等.DNA标记在植物品种鉴定上的应用现状[J].中国农学通报,2014,30(30):234-240.
[4]郑永胜,张晗,王东建,等.基于荧光检测技术的小麦品种SSR鉴定体系的建立[J].中国农业科学,2014,47(19):3725-3735.
[5]吴则东,江伟,马龙彪.分子标记技术在农作物品种鉴定上的研究进展及未来展望[J].中国农学通报,2015,31(33):172-176.
[6]张雪原,赵攀峰,王凤格,等.玉米品种SSR分子标记与田间小区种植一致性鉴定结果的比较[J].玉米科学,2009,17(1):40-45,50.
[7]朱大洲,王坤,周光华,等.单粒大豆的近红外光谱特征及品种鉴别研究[J].光谱学与光谱分析,2010,30(12):3217-3221.
[8]刘天玲,苏琪雅,孙群,等.基于NIR分析和模式识别技术的玉米种子识别系统[J].光谱学与光谱分析,2012,32(6):1550-1553.
[9]黄敏,朱晓,朱启兵,等.基于高光谱图像的玉米种子特征提取与识别[J].光子学报,2012,41(7):868-873.
[10]邹伟,方慧,刘飞,等.基于高光谱图像技术的油菜籽品种鉴别方法研究[J].浙江大学学报:农业与生命科学版,2011,37(2):175-180.
[11]严衍禄.近红外光谱分析基础与应用[M].北京:中国轻工业出版社, 2005.
[12] Pua H, Kamruzzamanb M, Sun D W. Selection of feature wavelengths for developingmultispectral imaging systems for quality, safety and authenticity of muscle foods-areview[J]. Trends in Food Science and Technology, 2015,45:86-104.
[13] Xing J A, Symons S, Shahin M, et al. Detection of sprout damage in Canada Western RedSpring wheat with multiple wavebands using visible/near-infrared hyperspectral imaging[J].Biosystems Engineering, 2010,106(2):188-194.
[14] Wallays C, Missotten B, De Baerdemaeker J, et al. Hyperspectral waveband selection foron-line measurement of grain cleanness[J].Biosystems Engineering, 2009, 104(1):1-7.
[15]刘伟,刘长虹,郑磊.基于支持向量机的多光谱成像稻谷品种鉴定[J].农业工程学报,2014,30(10):145-151.
[16] Ma F, Wang J, Liu C, et al. Discrimination of kernel quality characteristics for sunflowerseeds based on multispectral imaging approach[J]. Food Analytical Methods,2015,8(7):1629-1636.
[17] Liu C, Liu W, Lu X, et al. Application of Multispectral Imaging to Determine QualityAttributes and Ripeness Stage in Strawberry Fruit[J]. Plos One, 2014, 9(2), e87818.
[18] Xiong C, Liu C, Pan W, et al. Non-destructive determination of total polyphenolscontent and classification of storage periods of Iron Buddha tea using multispectralimaging system[J]. Food Chemistry, 2015,176:130-136.
[19]翟俊海,赵文秀,王熙照.图像特征提取研究[J].河北大学学报:自然科学版, 2009,29(1):106-112.
[20] Ma F, Qin H, Shi K, et al. Feasibility of combining spectra with texture data ofmultispectral imaging to predict heme and non-heme iron contents in pork sausages[J].Food Chemistry, 2016, 190:142-149.
[21]张初,刘飞,孔汶汶,等.利用近红外高光谱图像技术快速鉴定西瓜种子品种[J].农业工程学报,2013,29(20):270-277.
[22]吴建国,石春海,张小明,等.用近红外反射光谱法分析稻米3种必需氨基酸含量的研究[J].作物学报,2003,29(05):688-692.
[23]杨小玲,由昭红,成芳.高光谱成像技术检测玉米种子成熟度[J].光谱学与光谱分析,2016,36(12):4028-4033.
[24]章毅,郭泉,王建勇.大数据分析的神经网络方法[J].工程科学与技术,2017,49(1):9-18.
[25]许思,赵光武,邓飞,等.基于高光谱的水稻种子活力无损分级检测[J].种子,2016,35(4):34-40.
[26]余鼎,程维民,王琦,等.基于近红外光谱技术的水稻种子成分分析模型的建立[J].光散射学报,2015,27(04):384-389.
[27]姚登举,杨静,詹晓娟.基于随机森林的特征选择算法[J].吉林大学学报:工学版,2014,44(01):137-141.
[28] Svante W. PLS-regression:a basic tool of chemometrics[J].Chemometrics&Intelligent Laboratory Systems,2001,58(2):109-130.
[29]邓小琴,朱启兵,黄敏.融合光谱、纹理及形态特征的水稻种子品种高光谱图像单粒鉴别[J].激光与光电子学进展,2015,52(02):128-134.
[30]黄艳艳,朱丽伟,李军会,等.应用近红外光谱技术快速鉴别玉米杂交种纯度的研究[J].光谱学与光谱分析,2011,31(03):661-664.