ANALYSIS OF WEED COVER USING COMPUTER VISION

Authors

  • Mário Cupertino da Silva Júnior
  • Francisco de Assis de Carvalho Pinto
  • Daniel Marçal de Queiroz
  • Gonzalo Ruiz Ruiz
  • Jaime Gómez Gil
  • Luis Manuel Navas Gracia
  • Adson Andrade Barros
  • Luis Gustavo Figueiredo França

DOI:

https://doi.org/10.63330/sasciencesv6n2-112

Keywords:

Agricultural sensing, Computer vision, Digital image processing, Precision agriculture, Weeds

Abstract

The present study addresses the determination of the percentage of weed coverage thru digital images applied to precision agriculture. This is an applied research study aimed at developing algorithms capable of performing the automatic and georeferenced acquisition of images and determining the percentage of weed coverage, providing information to support the localized application of herbicides. The methodology consisted of developing algorithms in the LabVIEW environment for the simultaneous acquisition of images and GPS coordinates, digital image processing, vegetation segmentation, identification of crop rows, and calculation of weed coverage. The system was evaluated in an experimental area cultivated with beans, using images obtained by RGB and near-infrared (NIR) cameras, whose results were analyzed thru confusion matrices, overall accuracy, and Kappa index. The results demonstrated that the algorithms allowed for the automation of image acquisition and processing, enabling the determination of the percentage of weed coverage in the interrows. The images obtained by the RGB camera showed better classification performance compared to the NIR images, while the no-till system favored the identification of vegetation in both cameras. It is concluded that the developed algorithms proved to be efficient in estimating weed cover, constituting a support tool for precision agriculture practices.

Downloads

Download data is not yet available.

References

BAIO, F.H.R. Aplicação localizada de defensivos baseada na variabilidade espacial das plantas daninhas. 2001. 113 f. Dissertação (Mestrado em Agronomia) – Escola Superior de Agricultura Luiz de Queiroz, Piracicaba, SP.

CONGALTON, R.G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sensing of Environment, n. 37, p. 35-46, 1991.

DUDA, R.O.; HART, P.E. Use of the Hough transformation to detect lines and curves in pictures. Communicationsofthe ACM, v. 15, n. 1, p. 11-15, 1972.

ESPINOSA, A.T.; PONCE, A.J.R.; REYES, M.M. Desarrollo de software para el manejo de información catastral. ReportesTécnicosenIngeniería del Software, v. 9, n. 1, p. 1-13, 2007.

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS – FAO. Disponível em: <http://www.fao.org/news/story/en/item/29402/icode/>. Acessoem: 11 jan. 2010.

GÉE, C.; BOSSU, J.; JONES, G.; TRUCHETET, F. Crop/weed discrimination in perspective agronomic images. ComputersandElectronics in Agriculture, v. 60, p. 49-59, 2008.

GOMES, O.F.M. Processamento e análise de imagens aplicados à caracterização automática de materiais.2001. 151 f. Dissertação (Mestrado em Ciências da Engenharia Metalúrgica) – Pontifícia Universidade Católica do Rio de Janeiro, Rio de Janeiro, RJ.

HUDSON, W.D.; RAMM, C.W. Correct formulation of the kappa coefficient of agreement. Photogrammetric Engineering & Remote Sensing, v. 53, n. 4, p. 421-422, 1987.

LANDIS, J.R.; KOCH, G.G. The measurement of observer agreement for categorical data. Biometrics, v. 33, n. 1, p. 159-174, 1977.

MOLIN, J.P. Agricultura de precisão: fundamentos e estado atual da arte. In: CÂMARA, G.M. Soja: tecnologia da produção II. Piracicaba: ESALQ/Departamento de Produção Vegetal, 2000. p. 423-437.

RONCHI, C.P.; TERRA, A.A.; SILVA, A.A.; FERREIRA, L.R. Acúmulo de nutrientes pelo cafeeiro sob interferência de plantas daninhas. Planta Daninha,v. 21, n. 2, p. 219-227, 2003.

SHIRATSUCHI, L.S. Mapeamento da variabilidade espacial das plantas daninhas com a utilização de ferramentas da agricultura de precisão. 2001. 96 f. Dissertação (Mestrado em Agronomia) – Escola Superior de Agricultura Luiz de Queiroz, Piracicaba, SP.

TELLAECHE, A.; BURGOSARTIZZU, X.P.; PAJARES, G.; RIBEIRO, A.; FERNÁNDEZ-QUINTANILLA, C. A new vision-based approach to differential spraying in precision agriculture. Computers and Electronics in Agriculture, v. 60, p. 144-155, 2008.

WOEBBECKE, D.M.; MEYER, G.E.; VON BARGEN, K.; MORTESEN, D.A. Color indices for weed identification under various soil, residue, and lighting conditions. Transactions of the ASAE, v. 38, n. 1, p. 259-269, 1995.

YANG, X.; BEYENAL, H.; GARY, H.; LEWANDOWSKI, Z. Evaluation of biofilm image thresholding methods. WaterResearch, v. 35, p. 1149-1158, 2001.

Published

2026-07-31

How to Cite

da Silva Júnior, M. C. ., Pinto, F. de A. de C. ., de Queiroz, D. M. ., Ruiz, G. R. ., Gil, J. G. ., Gracia, L. M. N. ., Barros, A. A. ., & França, L. G. F. . (2026). ANALYSIS OF WEED COVER USING COMPUTER VISION. South American Sciences, 6(2), e26333. https://doi.org/10.63330/sasciencesv6n2-112