基于LSP与GLCM融合的禾本科牧草种子特征提取算法
|
摘要
针对禾本科牧草种子相似性较高、识别困难的问题,采用局部相似模式(LSP)和灰度共生矩阵(GLCM)的方法,对禾本科牧草种子的分类识别进行研究。结果表明:1)局部相似模式与灰度共生矩阵融合的方法可以有效的提取禾本科牧草种子的纹理特征,能够识别颜色、形状、大小等特征都十分相似的牧草种子,且其识别率优于传统的LSP特征算子和GLCM特征算子。2)与传统LSP算法相比,结合灰度共生矩阵算法后,得到的特征受到相似种类种子图像的影响较小,具有更广泛的适应性。因此,基于LSP和GLCM的融合算法可以有效地提取相似禾本科种子图像的纹理统计特征,采用线性判别分析分类器(LDA)进行分类,识别率最高达到98.64%。
Aiming at the problem of high similarity and identification difficulties of Gramineous grass seeds,local similarity pattern(LSP) and gray level co-occurrence matrix(GLCM) are combined to study the classification and identification of Gramineous grass seeds.The results show that:1) The fusion of LSP and GLCM can effectively extract the texture characteristics of grass seeds to identify the seeds whose color,shape,size and other characteristics are very similar,and the accuracy is superior to that of the traditional LSP or GLCM operators.2) Compared with the traditional LSP algorithm,the extraction algorithm of LSP fused with GLCM can extract features,which are less affected by the similar species of seed images,representing a wider adaptability in Gramineous grass seed identification.Therefore,the extraction algorithm of the fusion of LSP and GLCM can effectively extract the statistical texture features of similar Gramineous grass seeds,which could reach the highest recognition accuracy of 98.64% when combined with LDA classifier.
Aiming at the problem of high similarity and identification difficulties of Gramineous grass seeds,local similarity pattern(LSP) and gray level co-occurrence matrix(GLCM) are combined to study the classification and identification of Gramineous grass seeds.The results show that:1) The fusion of LSP and GLCM can effectively extract the texture characteristics of grass seeds to identify the seeds whose color,shape,size and other characteristics are very similar,and the accuracy is superior to that of the traditional LSP or GLCM operators.2) Compared with the traditional LSP algorithm,the extraction algorithm of LSP fused with GLCM can extract features,which are less affected by the similar species of seed images,representing a wider adaptability in Gramineous grass seed identification.Therefore,the extraction algorithm of the fusion of LSP and GLCM can effectively extract the statistical texture features of similar Gramineous grass seeds,which could reach the highest recognition accuracy of 98.64% when combined with LDA classifier.
引文
[1] 刘立侠,李桂民,路云侠,王德利.中国牧草生物技术的研究现状和展望[J].草地学报,2009,17(3):389-397,401Liu L X,Li G M,Lu Y X,Wang D L.Current status and prospect in the biotechnology of forage breeding in China[J].Acta Agrestia Sinica,2009,17(3):389-397,401 (in Chinese)
[2] 游明鸿,刘金平,白史且,张新全,李达旭.行距对“川草2号”老芒麦生殖枝及种子产量性状的影响[J].草业学报,2011,20(6):299-304You M H,Liu J P,Bai S Q,Zhang X Q,Li D X.Influences of row spacing on fertile tillers and characters of seed yield of Elymus sibiricus cv Chuancao No.2[J].Acta Prataculturae Sinica,2011,20(6):299-304 (in Chinese)
[3] 朱振磊,张永亮,潘多锋,韩建国,申忠宝,韩云华,鲍青龙,王显国.行距与播种量对无芒雀麦种子产量及产量组分的影响[J].草地学报,2011,19(4):631-636Zhu Z L,Zhang Y L,Pan D F,Han J G,Shen Z B,Han Y H,Bao Q L,Wang X G.Effect of row spacing and seeding rate on seed yield and yield components of Bromus inermis Leyss[J].Acta Agrestia Sinica,2011,19(4):631-636 (in Chinese)
[4] Hillel D.Role of irrigation in agricultural systems[J].Agronomy,1990,30(32):5-30
[5] Beukes D J,Barnard S A.Effects of level and timing of irrigation on growth and water use of lucerne[J].South African Journal of Plant & Soil,1985,2(4):197-202
[6] 毛培胜.牧草与草坪草种子科学与技术[M].北京:中国农业大学出版社,2011:166-167Mao P S.Forage and Turfgrass Seed Science and Technology[M].Beijing:China Agricultural University Press,2001:166-167 (in Chinese)
[7] Thompson D J,Clark K W.Effects of clipping and nitrogen fertilization on tiller development and flowering in Kentucky bluegrass[J].Canadian Journal of Plant Science,1993,73(2):569-575
[8] 王佺珍,韩建国,周禾,刘富渊,仲勇.氮肥与植株密度互作对鸭茅种子产量的效应 [J].草业科学,2005,22(5):38-44Wang Q Z,Han J G,Zhou H,Liu F Y,Zhong Y.The effect of nitrogen and density of planting coupling on dactylis glomerato seed yield[J].Pratacultural Science,2005,22(5):38-44 (in Chinese)
[9] 王明亚,毛培胜.中国禾本科牧草种子生产技术研究进展 [J].种子,2012,31(9):55-60Wang M Y,Mao P S.Research advancement on seed production technology of forage grasses in China[J].Seed,2012,31(9):55-60 (in Chinese)
[10] 唐华俊,辛晓平,杨桂霞,张保辉,王旭,张宏斌,闫玉春.现代数字草业理论与技术研究进展及展望[J].中国草地学报,2009,31(4):1-8Tang H J,Xin X P,Yang G X,Zhang B H,Wang X,Zhang H B,Yan Y C.Advance and prospects in theories and techniques of modern digital grassland[J].Chinese Journal of Grassland,2009,31(4):1-8 (in Chinese)
[11] 金云翔,徐斌,杨秀春,李金亚,高添,于海达,马海龙.草原生物量及碳密度遥感估算:以内蒙古正蓝旗为例[J].中国农学通报,2013,29(5):11-16Jin Y X,Xu B,Yang X C,Li J Y,Gao T,Yu H D,Ma H L.Remote sensing estimation of grassland biomass and carbon density:Case of Zhenglan Banner[J].Chinese Agricultural Science Bulletin,2013,29(5):11-16 (in Chinese)
[12] Wen Q,Zhang Z,Zhao X,Yi L,Wang X,Hu S G,Liu B,Zeng T.Regularity and causes of grassland variations in China over the past 30 years using remote sensing data[J].International Journal of Image & Data Fusion,2015,6(4):330-347
[13] 李金亚.科尔沁沙地草原沙化时空变化特征遥感监测及驱动力分析[D].北京:中国农业科学院,2014Li J Y.Spatio-temporal variations and its driving factors of the grassland sandy desertification in the Horqin Sand Land based on remote sensing[D].Beijing:Chinese Academy of Agricultural Sciences,2014 (in Chinese)
[14] 赵学观,王秀,李翠玲,高原源,王松林,冯青春.基于主成分分析及LVQ神经网络的番茄种子品种识别[J].浙江农业学报,2017,29(8):1375-1383Zhao X G,Wang X,Li C L,Gao YY,Wang S L,Feng Q C.Tomato seed varieties recognition based on principal component analysis and LVQ neural network[J].Acta Agriculturae Zhejiangensis,29(8):1375-1383 (in Chinese)
[15] 刘双喜,张宏建,王金星,王震,张春庆,李岩.基于可见光波段的色彩概率聚类模型的玉米杂交种子识别[J].光谱学与光谱分析,2018,38(8):2516-2523Liu S X,Zhang H J,Wang J X Wang Z,Zhang C Q,Li Y.Hybrid seed recognition of maize based on probability clustering model using visible light color features[J].Spectroscopy and Spectral Analysis,2018,38(8):2516-2523 (in Chinese)
[16] 龙怡霖,蔡骋.基于随机森林的缺损杂草种子识别[J].计算机应用与软件,2016,33(8):185-189Long Y L,Cai P.Random forest-based damaged weed seeds recognition[J].Computer Applications and Software,2016,33(8):185-189 (in Chinese)
[17] Pourreza H R,Masoudifar M,Manafzade M M.LSP:Local similarity pattern,a new approach for rotation in variant noisy texture analysis[C].In:Proceedings of the IEEE International Conference on Image Processing,Brussels,Belgium:2011:837-840
[18] Haralick R M.Statistical and structural approaches to texture[J].Proceedings of the IEEE,1979,67(5):786-804
[19] Ulaby F T,Kouyate F,Brisco B,Williams T H.Textural information in SAR images [J].IEEE Transactions on Geoscience and Remote Sensing,1986,24(2):235-245
[2] 游明鸿,刘金平,白史且,张新全,李达旭.行距对“川草2号”老芒麦生殖枝及种子产量性状的影响[J].草业学报,2011,20(6):299-304You M H,Liu J P,Bai S Q,Zhang X Q,Li D X.Influences of row spacing on fertile tillers and characters of seed yield of Elymus sibiricus cv Chuancao No.2[J].Acta Prataculturae Sinica,2011,20(6):299-304 (in Chinese)
[3] 朱振磊,张永亮,潘多锋,韩建国,申忠宝,韩云华,鲍青龙,王显国.行距与播种量对无芒雀麦种子产量及产量组分的影响[J].草地学报,2011,19(4):631-636Zhu Z L,Zhang Y L,Pan D F,Han J G,Shen Z B,Han Y H,Bao Q L,Wang X G.Effect of row spacing and seeding rate on seed yield and yield components of Bromus inermis Leyss[J].Acta Agrestia Sinica,2011,19(4):631-636 (in Chinese)
[4] Hillel D.Role of irrigation in agricultural systems[J].Agronomy,1990,30(32):5-30
[5] Beukes D J,Barnard S A.Effects of level and timing of irrigation on growth and water use of lucerne[J].South African Journal of Plant & Soil,1985,2(4):197-202
[6] 毛培胜.牧草与草坪草种子科学与技术[M].北京:中国农业大学出版社,2011:166-167Mao P S.Forage and Turfgrass Seed Science and Technology[M].Beijing:China Agricultural University Press,2001:166-167 (in Chinese)
[7] Thompson D J,Clark K W.Effects of clipping and nitrogen fertilization on tiller development and flowering in Kentucky bluegrass[J].Canadian Journal of Plant Science,1993,73(2):569-575
[8] 王佺珍,韩建国,周禾,刘富渊,仲勇.氮肥与植株密度互作对鸭茅种子产量的效应 [J].草业科学,2005,22(5):38-44Wang Q Z,Han J G,Zhou H,Liu F Y,Zhong Y.The effect of nitrogen and density of planting coupling on dactylis glomerato seed yield[J].Pratacultural Science,2005,22(5):38-44 (in Chinese)
[9] 王明亚,毛培胜.中国禾本科牧草种子生产技术研究进展 [J].种子,2012,31(9):55-60Wang M Y,Mao P S.Research advancement on seed production technology of forage grasses in China[J].Seed,2012,31(9):55-60 (in Chinese)
[10] 唐华俊,辛晓平,杨桂霞,张保辉,王旭,张宏斌,闫玉春.现代数字草业理论与技术研究进展及展望[J].中国草地学报,2009,31(4):1-8Tang H J,Xin X P,Yang G X,Zhang B H,Wang X,Zhang H B,Yan Y C.Advance and prospects in theories and techniques of modern digital grassland[J].Chinese Journal of Grassland,2009,31(4):1-8 (in Chinese)
[11] 金云翔,徐斌,杨秀春,李金亚,高添,于海达,马海龙.草原生物量及碳密度遥感估算:以内蒙古正蓝旗为例[J].中国农学通报,2013,29(5):11-16Jin Y X,Xu B,Yang X C,Li J Y,Gao T,Yu H D,Ma H L.Remote sensing estimation of grassland biomass and carbon density:Case of Zhenglan Banner[J].Chinese Agricultural Science Bulletin,2013,29(5):11-16 (in Chinese)
[12] Wen Q,Zhang Z,Zhao X,Yi L,Wang X,Hu S G,Liu B,Zeng T.Regularity and causes of grassland variations in China over the past 30 years using remote sensing data[J].International Journal of Image & Data Fusion,2015,6(4):330-347
[13] 李金亚.科尔沁沙地草原沙化时空变化特征遥感监测及驱动力分析[D].北京:中国农业科学院,2014Li J Y.Spatio-temporal variations and its driving factors of the grassland sandy desertification in the Horqin Sand Land based on remote sensing[D].Beijing:Chinese Academy of Agricultural Sciences,2014 (in Chinese)
[14] 赵学观,王秀,李翠玲,高原源,王松林,冯青春.基于主成分分析及LVQ神经网络的番茄种子品种识别[J].浙江农业学报,2017,29(8):1375-1383Zhao X G,Wang X,Li C L,Gao YY,Wang S L,Feng Q C.Tomato seed varieties recognition based on principal component analysis and LVQ neural network[J].Acta Agriculturae Zhejiangensis,29(8):1375-1383 (in Chinese)
[15] 刘双喜,张宏建,王金星,王震,张春庆,李岩.基于可见光波段的色彩概率聚类模型的玉米杂交种子识别[J].光谱学与光谱分析,2018,38(8):2516-2523Liu S X,Zhang H J,Wang J X Wang Z,Zhang C Q,Li Y.Hybrid seed recognition of maize based on probability clustering model using visible light color features[J].Spectroscopy and Spectral Analysis,2018,38(8):2516-2523 (in Chinese)
[16] 龙怡霖,蔡骋.基于随机森林的缺损杂草种子识别[J].计算机应用与软件,2016,33(8):185-189Long Y L,Cai P.Random forest-based damaged weed seeds recognition[J].Computer Applications and Software,2016,33(8):185-189 (in Chinese)
[17] Pourreza H R,Masoudifar M,Manafzade M M.LSP:Local similarity pattern,a new approach for rotation in variant noisy texture analysis[C].In:Proceedings of the IEEE International Conference on Image Processing,Brussels,Belgium:2011:837-840
[18] Haralick R M.Statistical and structural approaches to texture[J].Proceedings of the IEEE,1979,67(5):786-804
[19] Ulaby F T,Kouyate F,Brisco B,Williams T H.Textural information in SAR images [J].IEEE Transactions on Geoscience and Remote Sensing,1986,24(2):235-245