Hostname: page-component-78c5997874-fbnjt Total loading time: 0 Render date: 2024-11-11T04:17:26.352Z Has data issue: false hasContentIssue false
  • English
  • Français

Discrimination of Common Ragweed (Ambrosia artemisiifolia) and Mugwort (Artemisia vulgaris) Based on Bag of Visual Words Model

Published online by Cambridge University Press:  06 March 2017

Anton Ustyuzhanin ,
Karl-Heinz Dammer ,
Antje Giebel ,
Cornelia Weltzien  and
Michael Schirrmann
Anton Ustyuzhanin*
Affiliation:
Scientist, Scientist, Research Technician, Senior Scientist, and Senior Scientist, Department of Engineering for Crop Production, Leibniz Institute for Agricultural Engineering and Bioeconomy, Max-Eyth-Allee 100, 14469 Potsdam, Germany
Karl-Heinz Dammer
Affiliation:
Scientist, Scientist, Research Technician, Senior Scientist, and Senior Scientist, Department of Engineering for Crop Production, Leibniz Institute for Agricultural Engineering and Bioeconomy, Max-Eyth-Allee 100, 14469 Potsdam, Germany
Antje Giebel
Affiliation:
Scientist, Scientist, Research Technician, Senior Scientist, and Senior Scientist, Department of Engineering for Crop Production, Leibniz Institute for Agricultural Engineering and Bioeconomy, Max-Eyth-Allee 100, 14469 Potsdam, Germany
Cornelia Weltzien
Affiliation:
Scientist, Scientist, Research Technician, Senior Scientist, and Senior Scientist, Department of Engineering for Crop Production, Leibniz Institute for Agricultural Engineering and Bioeconomy, Max-Eyth-Allee 100, 14469 Potsdam, Germany
Michael Schirrmann
Affiliation:
Scientist, Scientist, Research Technician, Senior Scientist, and Senior Scientist, Department of Engineering for Crop Production, Leibniz Institute for Agricultural Engineering and Bioeconomy, Max-Eyth-Allee 100, 14469 Potsdam, Germany
*
*Corresponding author’s E-mail: austyuzhanin@atb-potsdam.de
Get access Rights & Permissions [Opens in a new window]

Abstract

Common ragweed is a plant species causing allergic and asthmatic symptoms in humans. To control its propagation, an early identification system is needed. However, due to its similar appearance with mugwort, proper differentiation between these two weed species is important. Therefore, we propose a method to discriminate common ragweed and mugwort leaves based on digital images using bag of visual words (BoVW). BoVW is an object-based image classification that has gained acceptance in many areas of science. We compared speeded-up robust features (SURF) and grid sampling for keypoint selection. The image vocabulary was built using K-means clustering. The image classifier was trained using support vector machines. To check the robustness of the classifier, specific model runs were conducted with and without damaged leaves in the trainings dataset. The results showed that the BoVW model allows the discrimination between common ragweed and mugwort leaves with high accuracy. Based on SURF keypoints with 50% of 788 images in total as training data, we achieved a 100% correct recognition of the two plant species. The grid sampling resulted in slightly less recognition accuracy (98 to 99%). In addition, the classification based on SURF was up to 31 times faster.

Ambrosia artemisiifolia es una especie vegetal que causa síntomas de alergia y asma en humanos. Para controlar su propagación, se necesita un sistema de identificación temprana. Sin embargo, debido a su semejanza a Artemisia vulgaris, la identificación adecuada entre estas dos especies de malezas es importante. Por esto proponemos un método para discriminar las hojas de A. artemisiifolia y A. vulgaris con base en imágenes digitales usando el modelo de bolsa de palabras visuales (BoVW). BoVW es una clasificación de imágenes con base en objetos definidos que ha ganado aceptación en muchas áreas de la ciencia. Comparamos las características robustas aceleradas (SURF) y el muestreo en cuadrícula para la selección de puntos clave. El vocabulario de imágenes fue construido usando el agrupamiento de promedios K. El clasificador de imágenes fue entrenado usando máquinas vectoriales de apoyo. Para examinar la robustez del clasificador, se realizaron corridas de modelos específicos con y sin daño en las hojas en los set de datos para entrenamiento. Los resultados mostraron que el modelo BoVW permite la discriminación entre A. artemisiifolia y A. vulgaris con alta exactitud. Con base en los puntos clave SURF, con 50% del total de 788 imágenes como datos de entrenamiento, logramos un 100% de reconocimiento correcto de estas dos especies de plantas. El muestreo en cuadrícula resultó en una exactitud de reconocimiento ligeramente menor (98 a 99%). Adicionalmente, la clasificación basada en SURF fue hasta 31 veces más rápida.

Keywords

Common ragweed, Ambrosia artemisiifolia L.;mugwort, Artemisia vulgaris L.Weed speciesdigital imagesmachine learningspeeded-up robust featuresplant recognition
Type
Weed Management-Techniques
Information
Weed Technology , Volume 31 , Issue 2 , March 2017 , pp. 310 - 319
Copyright
© Weed Science Society of America, 2017 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

Associate Editor for this paper: Andrew Kniss, University of Wyoming

References

Literature Cited

Aakif, A, Khan, MF (2015) Automatic classification of plants based on their leaves. Biosyst Eng 139:66–75 CrossRefGoogle Scholar
Auda, Y, Blasco, F, Gastellu-Etchegorry, JP, Marty, G, Dechamp, C (2002) Preliminary studies of the detection of ambrosia populations by spatial remote sensing. Revue Française d’Allergologie et d’Immunologie Clinique 42:533–538 CrossRefGoogle Scholar
Bahmanyar, R, Cui, SY, Datcu, M (2015) A comparative study of bag-of-words and bag-of-topics models of EO image patches. IEEE Geosci Remote Sens Lett 12:1357–1361 CrossRefGoogle Scholar
Bay, H, Ess, A, Tuytelaars, T, Van Gool, L (2008) Speeded-up robust features (SURF). Comput Vis Image Underst 110:346–359 CrossRefGoogle Scholar
Brandes, D, Nitsche, J (2006) Verbreitung, Ökologie, Soziologie von Ambrosia artemisiifolia L. in Mitteleuropa. Tuexenia 27:167–194 Google Scholar
Bromuri, S, Zufferey, D, Hennebert, J, Schumacher, M (2014) Multi-label classification of chronically ill patients with bag of words and supervised dimensionality reduction algorithms. J Biomed Inform 51:165–175 CrossRefGoogle ScholarPubMed
Csurka, G, Dance, CR, Fan, L, Willamowski, J, Bray, C (2004) Visual categorization with bags of keypoints. Pages 59–74 in Proceedings of the Workshop on Statistical Learning in Computer Vision. Prague, Czech Republic: ECCV Google Scholar
Dammer, K, Intress, J, Beuche, H, Selbeck, J, Dworak, V (2012) Discrimination of Ambrosia artemisiifolia L. and Artemisia vulgaris L. by hyperspectral image analysis during the growing season. Weed Res 53:146–156 CrossRefGoogle Scholar
Dietterich, TG (1998) Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput 10:1895–1923 CrossRefGoogle ScholarPubMed
Fakhri, A, Nasir, A, Nordin, A, Rahman, M, Nashriyah, Mat, Rasid Mamat, A (2014) Automatic identification of Ficus deltoidea Jack (Moraceae) varieties based on leaf. Mod Appl Sci 8:121–131 Google Scholar
Faraki, M, Palhang, M, Sanderson, K (2015) Log-Euclidean bag of words for human action recognition. IET Comput Vision 9:331–339 CrossRefGoogle Scholar
Fumanal, B, Girod, C, Fried, G, Bretagnolle, F, Chauvel, B (2008) Can the large ecological amplitude of Ambrosia artemisiifolia explain its invasive success in France? Weed Res 48:349–359 CrossRefGoogle Scholar
Guo, W, Fukatsu, T, Ninomiya, S (2015) Automated characterization of flowering dynamics in rice using field-acquired time-series RGB images. Plant Methods 11:7 CrossRefGoogle ScholarPubMed
Hammaoui-Laguel, L, Vautard, R, Liu, L, Solmon, F, Viovy, N, Khvorostyanov, D, Essl, F, Chuine, I, Colette, A, Semenov, MA, Schaffhauser, A, Storkey, J, Thibaudon, M, Epstein, MM (2015) Effects of climate change and seed dispersal on airborne ragweed pollen loads in Europe. Nat Clim Change 5:766–771 CrossRefGoogle Scholar
Hemming, J, Rath, T (2001) Computer-vision-based weed identification under field conditions using controlled lighting. J Agric Eng Res 78:233–243 CrossRefGoogle Scholar
Hernandez-Serna, A, Jimenez-Segura, LF (2014) Automatic identification of species with neural networks. PeerJ 2:e563 CrossRefGoogle ScholarPubMed
Jarić, S, Mitrović, M, Vrbničanin, S, Karadžić, B, Đurđević, L, Kostić, O, Mačukanović-Jocić, M, Gajić, G, Pavlović, P (2011) A contribution to studies of the ruderal vegetation of southern Srem, Serbia. Arch Biol Sci (Belgrade) 63:1181–1197 CrossRefGoogle Scholar
Ji, SW, Li, YX, Zhou, ZH, Kumar, S, Ye, JP (2009) A bag-of-words approach for Drosophila gene expression pattern annotation. BMC Bioinformatics 10:119 CrossRefGoogle ScholarPubMed
Jin, T, Hou, X, Li, P, Zhou, F (2015) A novel method of automatic plant species identification using sparse representation of leaf tooth features. PLoS One 10:e0139482 CrossRefGoogle ScholarPubMed
Jung, A, Kardevan, P, Reisinger, P (2006) Detection of common ragweed (Ambrosia artemisiifolia L.) reflectance spectrum by means of field measurements. Pages 153–160 in Proceedings of the 4th International Plant Protection Symposium (11th Trans-Tisza Plant Protection Forum). Debrecen, Hungary: Debrecen University Google Scholar
Kaden, NN, Terentjeva, NN (1979) Etimologicheskiy slovar nauchnyih nazvaniy sosudistyih rasteniy dikorastuschih i razvodimyih v SSSR [Etymological Dictionary of Scientific Names of Wild and Farmed in the USSR Vascular Plants] Issue 1. Moscow: Moscow State University Press. 268 pGoogle Scholar
Larsen, AB, Vestergaard, JS, Larsen, R (2014) HEp-2 cell classification using shape index histograms with donut-shaped spatial pooling. IEEE Trans Med Imag 33:1573–1580 CrossRefGoogle ScholarPubMed
Leiblein, M (2008) Biomasse-Entwicklung und Konkurrenzbiologie des invasiven Neophyten Ambrosia artemisiifolia. Pages 97–102 in Feit U & Korn H eds. Treffpunkt Biologische Vielfalt VIII. Bonn: Bundesamt für Naturschutz Google Scholar
Leiblein-Wild, M, Tackenberg, O (2014) Phenotypic variation of 38 European Ambrosia artemisiifolia populations measured in a common garden experiment. Biol Invasions 16:2003–2015 CrossRefGoogle Scholar
Li, ZS, Imai, J, Kaneko, M (2011) Block-based bag of words for robust face recognition under variant conditions of facial expression, illumination, and partial occlusion. IEICE Trans Fundam Electron Commun Comput Sci E94–A:533–541 CrossRefGoogle Scholar
Lozano-Vega, G, Benezeth, Y, Marzani, F, Boochs, F (2014) Modular method of detection, localization, and counting of multiple-taxon pollen apertures using bag-of-words. J Electron Imaging 23:053025 CrossRefGoogle Scholar
Lüdke, HJ (2009) Ragweed allergy in Brandenburg. Allergologie 32:402 CrossRefGoogle Scholar
Makra, L, Matyasovszky, I, Hufnagel, L, Tusnady, G (2015) The history of ragweed in the world. Appl Ecol Env Res 13:489–512 Google Scholar
McCarthy, C, Rees, S, Baillie, C (2012) Preliminary evaluation of shape and colour image sensing for automated weed identification in sugarcane. Pages 3–7 in Proceedings of the 34th Annual Conference. Palm Cove, Australia: Australian Society of Sugar Cane Technologists Google Scholar
Mukti, FA, Eswaran, C, Hashim, N (2015) Detection and classification of diabetic retinopathy anomalies using bag-of-words model. J Med Imaging Health Inf 5:1009–1019 CrossRefGoogle Scholar
Ngom, R, Gosselin, P (2014) Development of a remote sensing-based method to map likelihood of common ragweed (Ambrosia artemisiifolia) presence in urban areas. IEEE J Sel Top Appl Earth Obs Remote Sens 7:126–139 CrossRefGoogle Scholar
Oyallon, E, Rabin, J (2015) An analysis of the SURF method. Image Process On Line 5:176–218 CrossRefGoogle Scholar
Ren, Y (2016) A comparative study of irregular pyramid matching in bag-of-bags of words model for image retrieval. Signal Image Video Process 10:471–478 CrossRefGoogle Scholar
Silc, U (2009) Vegetation of the Zale cemetery (Ljubljana). Hacquetia 8:41–47 CrossRefGoogle Scholar
Sirbu, C (2008) Chorological and phytocoenological aspects regarding the invasion of some alien plants, on the Romanian territory. Acta Hortic Bot Bucurest 35:60–68 Google Scholar
Storkey, A, Stratonovitch, P, Chapman, DS, Vidotto, F, Semenov, MA (2014) Process-based approach to predicting the effect of climate change on the distribution of an invasive allergenic plant in Europe. PLoS One 9:e88156 CrossRefGoogle ScholarPubMed
Sun, H, Sun, X, Wang, HQ, Li, Y, Li, XJ (2012) Automatic target detection in high-resolution remote sensing images using spatial sparse coding bag-of-words model. IEEE Geosci Remote Sens Lett 9:109–113 CrossRefGoogle Scholar
Swain, K, Nørremark, M, Jørgensen, R, Midtiby, H, Green, O (2011) Weed identification using an automated active shape matching (AASM) technique. Biosyst Eng 110:450–457 CrossRefGoogle Scholar
Ustyuzhanin, A, Intreß, J, Schirrmann, M, Chochlov, N, Dammer, K (2015) Identifizierung von Beifußblättriger Ambrosie (Ambrosia artemisiifolia) mittels Bildverarbeitung in einem Winterroggenfeld. Gesunde Pflanz 67:165–173 CrossRefGoogle Scholar
Verschwele, A (2014) Die Beifuss-Ambrosie auf Ackerflachen – ein Problem? Pages 21–26 in Starfinger U, Sölter U & Verschwele A eds. Julius-Kühn-Archiv 445: Ambrosia in Deutschland – lässt sich die Invasion aufhalten? Quedlinburg, Germany: Julius-Kühn-Institut, Bundesforschungsinstitut für Kulturpflanzen Google Scholar
Vetrivel, A, Gerke, M, Kerle, N, Vosselman, G (2016) Identification of structurally damaged areas in airborne oblique images using a visual-bag-of-words approach. Remote Sens 8:231 CrossRefGoogle Scholar
Vuković, I, Mesić, M, Bajić, M, Krtalić, A, Gold, H, Kisić, I, Bašić, F, Zgorelec, Ž, Sajko, K (2007) Spatial distribution of ambrosia weediness in soybean at different rates of nitrogen fertilization, based on digital imagery analysis. Agriculturae Conspectus Scientificus 72:103–111 Google Scholar
Wang, J, Liu, P, She, MFH, Nahavandi, S, Kouzani, A (2013) Bag-of-words representation for biomedical time series classification. Biomed Signal Process Control 8:634–644 CrossRefGoogle Scholar
Weis, M, Gerhards, R (2007) Feature extraction for the identification of weed species in digital images for the purpose of site-specific weed control. Pages 537–545 in Stafford J, ed. Precision Agriculture ’07, 6th European Conference on Precision Agriculture. Wageningen, Netherlands: Wageningen Academic Publishers Google Scholar
Willamowski, J, Arregui, D, Csurka, G, Dance, CR, Fan, L (2004) Categorizing nine visual classes using local appearance descriptors. Pages 1–11 in Proceedings of ICPR Workshop on Learning for Adaptable Visual Systems. Cambridge, United Kingdom: IEEE Computer Society Google Scholar
Yang, J, Jiang, Y, Hauptmann, A, Ngo, C (2007) Evaluating bag-of-visual-words representations in scene classification. Pages 197–206 in Proceedings of the 9th International Workshop on Multimedia Information Retrieval. Augsburg, Germany: ACM SIGMM CrossRefGoogle Scholar
Yanikoglu, B, Aptoula, E, Tirkaz, C (2014) Automatic plant identification from photographs. Mach Vis Appl 25:1369–1383 CrossRefGoogle Scholar