Unsupervised classification algorithm for early weed detection in Row-Crops by combining spatial and spectral information
| Autor: | Marine Louargant, Gawain Jones, Christelle Gée, Thibault Maillot, Romain Faroux, Sylvain Villette, Jean-Noël Paoli |
|---|---|
| Přispěvatelé: | Jones , Gawain, Agroécologie [Dijon], Université de Bourgogne (UB)-Institut National de la Recherche Agronomique (INRA)-Université Bourgogne Franche-Comté [COMUE] (UBFC)-AgroSup Dijon - Institut National Supérieur des Sciences Agronomiques, de l'Alimentation et de l'Environnement, AIRINOV SAS, program 'ANR CoSAC' [ANR-14-CE18-0007], Horizon 2020 project IWMPRAISE [727321], Institut National de la Recherche Agronomique ( INRA ) -Université de Bourgogne ( UB ) -AgroSup Dijon - Institut National Supérieur des Sciences Agronomiques, de l'Alimentation et de l'Environnement-Université Bourgogne Franche-Comté ( UBFC ), Institut National de la Recherche Agronomique (INRA)-Université de Bourgogne (UB)-AgroSup Dijon - Institut National Supérieur des Sciences Agronomiques, de l'Alimentation et de l'Environnement-Université Bourgogne Franche-Comté [COMUE] (UBFC) |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
[SDV.SA]Life Sciences [q-bio]/Agricultural sciences
010504 meteorology & atmospheric sciences Computer science weed detection SVM [SDV]Life Sciences [q-bio] Multispectral image Row crop Image processing 01 natural sciences Crop image processing spatial information multispectral information automatic training data set generation Spatial analysis [ SDV.SA ] Life Sciences [q-bio]/Agricultural sciences 0105 earth and related environmental sciences 2. Zero hunger Pixel biology [ SDV ] Life Sciences [q-bio] business.industry Sowing 04 agricultural and veterinary sciences 15. Life on land biology.organism_classification Weed control Support vector machine Agriculture 040103 agronomy & agriculture automatic training dataset generation 0401 agriculture forestry and fisheries General Earth and Planetary Sciences Sugar beet business Weed Algorithm |
| Zdroj: | Remote Sensing 5 (10), 2-18. (2018) Remote Sensing; Volume 10; Issue 5; Pages: 761 Remote Sensing Remote Sensing, MDPI, 2018, 10 (5), pp.2-18. ⟨10.3390/rs10050761⟩ Remote Sensing, MDPI, 2018, 10 (5), 〈10.3390/rs10050761〉 Remote Sensing, MDPI, 2018, 10 (5), ⟨10.3390/rs10050761⟩ |
| ISSN: | 2072-4292 |
| DOI: | 10.3390/rs10050761⟩ |
| Popis: | International audience; In agriculture, reducing herbicide use is a challenge to reduce health and environmental risks while maintaining production yield and quality. Site-specific weed management is a promising way to reach this objective but requires efficient weed detection methods. In this paper, an automatic image processing has been developed to discriminate between crop and weed pixels combining spatial and spectral information extracted from four-band multispectral images. Image data was captured at 3 m above ground, with a camera (multiSPEC 4C, AIRINOV, Paris) mounted on a pole kept manually. For each image, the field of view was approximately 4 m × 3 m and the resolution was 6 mm/pix. The row crop arrangement was first used to discriminate between some crop and weed pixels depending on their location inside or outside of crop rows. Then, these pixels were used to automatically build the training dataset concerning the multispectral features of crop and weed pixel classes. For each image, a specific training dataset was used by a supervised classifier (Support Vector Machine) to classify pixels that cannot be correctly discriminated using only the initial spatial approach. Finally, inter-row pixels were classified as weed and in-row pixels were classified as crop or weed depending on their spectral characteristics. The method was assessed on 14 images captured on maize and sugar beet fields. The contribution of the spatial, spectral and combined information was studied with respect to the classification quality. Our results show the better ability of the spatial and spectral combination algorithm to detect weeds between and within crop rows. They demonstrate the improvement of the weed detection rate and the improvement of its robustness. On all images, the mean value of the weed detection rate was 89% for spatial and spectral combination method, 79% for spatial method, and 75% for spectral method. Moreover, our work shows that the plant in-line sowing can be used to design an automatic image processing and classification algorithm to detect weed without requiring any manual data selection and labelling. Since the method required crop row identification, the method is suitable for wide-row crops and high spatial resolution images (at least 6 mm/pix). |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|