Dicotyledon Weed Quantification Algorithm for Selective Herbicide Application in Maize Crops: Statistical Evaluation of the Potential Herbicide Savings
This work contributes a statistical model and simulation
framework yielding the best estimate possible for the potential
herbicide reduction when using the MoDiCoVi algorithm all the
while requiring a efficacy comparable to conventional spraying. In
June 2013 a maize field located in Denmark were seeded. The field
was divided into parcels which was assigned to one of two main
groups: 1) Control, consisting of subgroups of no spray and full dose
spraty; 2) MoDiCoVi algorithm subdivided into five different leaf
cover thresholds for spray activation. In addition approximately 25%
of the parcels were seeded with additional weeds perpendicular to
the maize rows. In total 299 parcels were randomly assigned with
the 28 different treatment combinations. In the statistical analysis,
bootstrapping was used for balancing the number of replicates. The
achieved potential herbicide savings was found to be 70% to 95%
depending on the initial weed coverage. However additional field
trials covering more seasons and locations are needed to verify
the generalisation of these results. There is a potential for further
herbicide savings as the time interval between the first and second
spraying session was not long enough for the weeds to turn yellow,
instead they only stagnated in growth.
[1] Wallinga, Groeneveld, and Lotz, “Measures that describe weed spatial
patterns at different levels of resolution and their applications for patch
spraying of weeds,” Weed Res., vol. 38, no. 5, pp. 351–359, 1 Oct. 1998.
[2] S. L. Young and F. J. Pierce, Automation: The Future of Weed Control
in Cropping Systems:. Springer Netherlands, 2014.
[3] T. W. Berge, S. Goldberg, K. Kaspersen, and J. Netland, “Towards
machine vision based site-specific weed management in cereals,”
Comput. Electron. Agric., vol. 81, no. 0, pp. 79–86, Feb. 2012.
[4] R. Gerhards and S. Christensen, “Real-time weed detection, decision
making and patch spraying in maize, sugarbeet, winter wheat and winter
barley,” Weed Res., vol. 43, no. 6, pp. 385–392, 1 Dec. 2003.
[5] S. Christensen, H. T. Søgaard, P. Kudsk, M. Nørremark, I. Lund, E. S.
Nadimi, and R. Jørgensen, “Site-specific weed control technologies,”
Weed Res., vol. 49, no. 3, pp. 233–241, 1 Jun. 2009.
[6] X. P. Burgos-Artizzu, A. Ribeiro, M. Guijarro, and G. Pajares,
“Real-time image processing for crop/weed discrimination in maize
fields,” Comput. Electron. Agric., vol. 75, no. 2, pp. 337–346, Feb. 2011.
[7] R. D. Lamm, D. C. Slaughter, and D. K. Giles, “Precision weed control
system for cotton,” Trans. ASAE, vol. 45, no. 1, pp. 231–238, 2002.
[8] D. C. Slaughter, “The biological engineer: Sensing the difference
between crops and weeds,” in Automation: The Future of Weed Control
in Cropping Systems. Springer Netherlands, 2014, pp. 71–95.
[9] R. Gerhards and H. Oebel, “Practical experiences with a system for
site-specific weed control in arable crops using real-time image analysis
and GPS-controlled patch spraying,” Weed Res., vol. 46, no. 3, pp.
185–193, 1 Jun. 2006.
[10] D. C. Slaughter, D. K. Giles, and D. Downey, “Autonomous robotic
weed control systems: A review,” Comput. Electron. Agric., vol. 61,
no. 1, pp. 63–78, Apr. 2008.
[11] R. N. Jørgensen, N. Krueger, H. S. Midtiby, and M. S. Laursen, “Spray
boom for selectively spraying a herbicidal composition onto dicots,”
Patent 20 140 001 276, 2 Jan., 2014.
[12] M. J. Kropff and C. J. T. Spitters, “A simple model of crop loss by weed
competition from early observations on relative leaf area of the weeds,”
Weed Res., vol. 31, no. 2, pp. 97–105, 1 Apr. 1991. [13] C. J. T. Spitters and R. Aerts, “Simulation of competition for light and
water in crop-weed associations,” Asp. Appl. Biol., vol. 4, pp. 467–483,
1983.
[14] M. S. Laursen, H. S. Midtiby, R. N. Jørgensen, and N. Kr¨uger,
“Validation of MoDiCoVi - monocot and dicot coverage ratio vision
based method for real time estimation of canopy coverage ration between
cereal and dicotyledon weeds,” in International Conference on Precision
Agriculture, Indianapolis, US, July 2012, 2012.
[15] A. Ali, J. C. Streibig, S. Christensen, and C. Andreasen, “Estimation
of weeds leaf cover using image analysis and its relationship with
fresh biomass yield of maize under field conditions,” in Proceedings
2011 International Conference on Information and Communication
Technologies in Agriculture, Food and Environment. ceur-ws.org, 2011,
pp. 41–49.
[16] M. S. Laursen, R. N. Jørgensen, H. S. Midtiby, K. Jensen, M. P.
Christiansen, T. M. Giselsson, A. K. Mortensen, and P. K. Jensen,
“Dicotyledon weed quantification algorithm for selective herbicide
application in maize crops,” Sensors, vol. 16, no. 11, p. 1848, 2016.
[17] K. Kristensen, “The use of spatial and randomisation-based methods for
analysis of trials with treatments randomised into rows and columns,”
J. Stat. Plan. Inference, vol. 140, no. 6, pp. 1542–1549, Jun. 2010.
[18] B. M. Bolker, M. E. Brooks, C. J. Clark, S. W. Geange, J. R. Poulsen,
M. H. H. Stevens, and J.-S. S. White, “Generalized linear mixed models:
a practical guide for ecology and evolution,” Trends Ecol. Evol., vol. 24,
no. 3, pp. 127–135, Mar. 2009.
[19] A. Onofri, E. A. Carbonell, H.-P. Piepho, A. M. Mortimer, and R. D.
Cousens, “Current statistical issues in weed research,” Weed Res.,
vol. 50, no. 1, pp. 5–24, 1 Feb. 2010.
[20] B. Efron, “Bootstrap methods: Another look at the jackknife,” Ann. Stat.,
vol. 7, no. 1, pp. 1–26, Jan. 1979.
[21] R Core Team, “R: A language and environment for statistical
computing,” Vienna, Austria, 2012.
[22] D. M. Woebbecke, G. E. Meyer, K. V. Bargen, and D. A. Mortensen,
“Color indices for weed identification under various soil, residue, and
lighting conditions,” Trans. ASAE, vol. 38, no. l, pp. 259–269, 1995.
[23] B. Efron and R. J. Tibshirani, An Introduction to the Bootstrap, ser.
Chapman & Hall/CRC Monographs on Statistics & Applied Probability.
Taylor & Francis, 1994.
[24] R. J. Carroll and D. Ruppert, Transformation and weighting in
regression. CRC Press, 1988, vol. 30.
[25] S. S. Seefeldt, J. E. Jensen, and E. P. Fuerst, “Log-Logistic analysis of
herbicide Dose-Response relationships,” Weed Technol., vol. 9, no. 2,
pp. 218–227, 1 Apr. 1995.
[26] J. Ascard, “Dose–response models for flame weeding in relation to plant
size and density,” Weed Res., vol. 34, no. 5, pp. 377–385, 1 Oct. 1994.
[27] M. Weisa and R. Gerhards, “Feature extraction for the identification
of weed species in digital images for the purpose of site-specific
weed control,” in Precision agriculture’07. Papers presented at the 6th
European Conference on Precision Agriculture, Skiathos, Greece, 3-6
June, 2007., 2007, pp. 537–544.
[28] M. S. Laursen, R. N. Jørgensen, M. Dyrmann, and R. N. Poulsen,
“RoboWeedSupport - sub millimeter weed image acquisition in cereal
crops with speeds up till 50 km/h,” in European Conference on Precision
Agriculture, Edinburgh, Scotland, July 2017, 2017.
[29] WeedIt, “WEEDit - plant detection technology,”
http://www.weedit.com.au/, 2015, accessed: 2015-3-12.
[30] T. Komives and P. Reisinger, “Precision weed control in sunflower
and maize - experiences from hungary,” in Proceedings, 25th
German Conference on Weed Biology and Weed Control, Volume
1, Braunschweig, Germany, March 13-15, 2012., vol. 1. Julius
K¨uhn Institut, Bundesforschungsinstitut f¨ur Kulturpflanzen, 2012, pp.
207–214.
[31] P. Lutman and P. Miller, “Spatially variable herbicide application
technology; opportunities for herbicide minimisation and protection of
beneficial weeds,” Tech. Rep., 2007.
[32] C. Tardif-Paradis, M.-J. Simard, G. D. Leroux, B. Panneton, R. E.
Nurse, and A. Vanasse, “Effect of planter and tractor wheels on row
and inter-row weed populations,” Crop Prot., vol. 71, no. 0, pp. 66–71,
May 2015.
[1] Wallinga, Groeneveld, and Lotz, “Measures that describe weed spatial
patterns at different levels of resolution and their applications for patch
spraying of weeds,” Weed Res., vol. 38, no. 5, pp. 351–359, 1 Oct. 1998.
[2] S. L. Young and F. J. Pierce, Automation: The Future of Weed Control
in Cropping Systems:. Springer Netherlands, 2014.
[3] T. W. Berge, S. Goldberg, K. Kaspersen, and J. Netland, “Towards
machine vision based site-specific weed management in cereals,”
Comput. Electron. Agric., vol. 81, no. 0, pp. 79–86, Feb. 2012.
[4] R. Gerhards and S. Christensen, “Real-time weed detection, decision
making and patch spraying in maize, sugarbeet, winter wheat and winter
barley,” Weed Res., vol. 43, no. 6, pp. 385–392, 1 Dec. 2003.
[5] S. Christensen, H. T. Søgaard, P. Kudsk, M. Nørremark, I. Lund, E. S.
Nadimi, and R. Jørgensen, “Site-specific weed control technologies,”
Weed Res., vol. 49, no. 3, pp. 233–241, 1 Jun. 2009.
[6] X. P. Burgos-Artizzu, A. Ribeiro, M. Guijarro, and G. Pajares,
“Real-time image processing for crop/weed discrimination in maize
fields,” Comput. Electron. Agric., vol. 75, no. 2, pp. 337–346, Feb. 2011.
[7] R. D. Lamm, D. C. Slaughter, and D. K. Giles, “Precision weed control
system for cotton,” Trans. ASAE, vol. 45, no. 1, pp. 231–238, 2002.
[8] D. C. Slaughter, “The biological engineer: Sensing the difference
between crops and weeds,” in Automation: The Future of Weed Control
in Cropping Systems. Springer Netherlands, 2014, pp. 71–95.
[9] R. Gerhards and H. Oebel, “Practical experiences with a system for
site-specific weed control in arable crops using real-time image analysis
and GPS-controlled patch spraying,” Weed Res., vol. 46, no. 3, pp.
185–193, 1 Jun. 2006.
[10] D. C. Slaughter, D. K. Giles, and D. Downey, “Autonomous robotic
weed control systems: A review,” Comput. Electron. Agric., vol. 61,
no. 1, pp. 63–78, Apr. 2008.
[11] R. N. Jørgensen, N. Krueger, H. S. Midtiby, and M. S. Laursen, “Spray
boom for selectively spraying a herbicidal composition onto dicots,”
Patent 20 140 001 276, 2 Jan., 2014.
[12] M. J. Kropff and C. J. T. Spitters, “A simple model of crop loss by weed
competition from early observations on relative leaf area of the weeds,”
Weed Res., vol. 31, no. 2, pp. 97–105, 1 Apr. 1991. [13] C. J. T. Spitters and R. Aerts, “Simulation of competition for light and
water in crop-weed associations,” Asp. Appl. Biol., vol. 4, pp. 467–483,
1983.
[14] M. S. Laursen, H. S. Midtiby, R. N. Jørgensen, and N. Kr¨uger,
“Validation of MoDiCoVi - monocot and dicot coverage ratio vision
based method for real time estimation of canopy coverage ration between
cereal and dicotyledon weeds,” in International Conference on Precision
Agriculture, Indianapolis, US, July 2012, 2012.
[15] A. Ali, J. C. Streibig, S. Christensen, and C. Andreasen, “Estimation
of weeds leaf cover using image analysis and its relationship with
fresh biomass yield of maize under field conditions,” in Proceedings
2011 International Conference on Information and Communication
Technologies in Agriculture, Food and Environment. ceur-ws.org, 2011,
pp. 41–49.
[16] M. S. Laursen, R. N. Jørgensen, H. S. Midtiby, K. Jensen, M. P.
Christiansen, T. M. Giselsson, A. K. Mortensen, and P. K. Jensen,
“Dicotyledon weed quantification algorithm for selective herbicide
application in maize crops,” Sensors, vol. 16, no. 11, p. 1848, 2016.
[17] K. Kristensen, “The use of spatial and randomisation-based methods for
analysis of trials with treatments randomised into rows and columns,”
J. Stat. Plan. Inference, vol. 140, no. 6, pp. 1542–1549, Jun. 2010.
[18] B. M. Bolker, M. E. Brooks, C. J. Clark, S. W. Geange, J. R. Poulsen,
M. H. H. Stevens, and J.-S. S. White, “Generalized linear mixed models:
a practical guide for ecology and evolution,” Trends Ecol. Evol., vol. 24,
no. 3, pp. 127–135, Mar. 2009.
[19] A. Onofri, E. A. Carbonell, H.-P. Piepho, A. M. Mortimer, and R. D.
Cousens, “Current statistical issues in weed research,” Weed Res.,
vol. 50, no. 1, pp. 5–24, 1 Feb. 2010.
[20] B. Efron, “Bootstrap methods: Another look at the jackknife,” Ann. Stat.,
vol. 7, no. 1, pp. 1–26, Jan. 1979.
[21] R Core Team, “R: A language and environment for statistical
computing,” Vienna, Austria, 2012.
[22] D. M. Woebbecke, G. E. Meyer, K. V. Bargen, and D. A. Mortensen,
“Color indices for weed identification under various soil, residue, and
lighting conditions,” Trans. ASAE, vol. 38, no. l, pp. 259–269, 1995.
[23] B. Efron and R. J. Tibshirani, An Introduction to the Bootstrap, ser.
Chapman & Hall/CRC Monographs on Statistics & Applied Probability.
Taylor & Francis, 1994.
[24] R. J. Carroll and D. Ruppert, Transformation and weighting in
regression. CRC Press, 1988, vol. 30.
[25] S. S. Seefeldt, J. E. Jensen, and E. P. Fuerst, “Log-Logistic analysis of
herbicide Dose-Response relationships,” Weed Technol., vol. 9, no. 2,
pp. 218–227, 1 Apr. 1995.
[26] J. Ascard, “Dose–response models for flame weeding in relation to plant
size and density,” Weed Res., vol. 34, no. 5, pp. 377–385, 1 Oct. 1994.
[27] M. Weisa and R. Gerhards, “Feature extraction for the identification
of weed species in digital images for the purpose of site-specific
weed control,” in Precision agriculture’07. Papers presented at the 6th
European Conference on Precision Agriculture, Skiathos, Greece, 3-6
June, 2007., 2007, pp. 537–544.
[28] M. S. Laursen, R. N. Jørgensen, M. Dyrmann, and R. N. Poulsen,
“RoboWeedSupport - sub millimeter weed image acquisition in cereal
crops with speeds up till 50 km/h,” in European Conference on Precision
Agriculture, Edinburgh, Scotland, July 2017, 2017.
[29] WeedIt, “WEEDit - plant detection technology,”
http://www.weedit.com.au/, 2015, accessed: 2015-3-12.
[30] T. Komives and P. Reisinger, “Precision weed control in sunflower
and maize - experiences from hungary,” in Proceedings, 25th
German Conference on Weed Biology and Weed Control, Volume
1, Braunschweig, Germany, March 13-15, 2012., vol. 1. Julius
K¨uhn Institut, Bundesforschungsinstitut f¨ur Kulturpflanzen, 2012, pp.
207–214.
[31] P. Lutman and P. Miller, “Spatially variable herbicide application
technology; opportunities for herbicide minimisation and protection of
beneficial weeds,” Tech. Rep., 2007.
[32] C. Tardif-Paradis, M.-J. Simard, G. D. Leroux, B. Panneton, R. E.
Nurse, and A. Vanasse, “Effect of planter and tractor wheels on row
and inter-row weed populations,” Crop Prot., vol. 71, no. 0, pp. 66–71,
May 2015.
@article{"International Journal of Biological, Life and Agricultural Sciences:75109", author = "Morten Stigaard Laursen and Rasmus Nyholm Jørgensen and Henrik Skov Midtiby and Anders Krogh Mortensen and Sanmohan Baby", title = "Dicotyledon Weed Quantification Algorithm for Selective Herbicide Application in Maize Crops: Statistical Evaluation of the Potential Herbicide Savings", abstract = "This work contributes a statistical model and simulation
framework yielding the best estimate possible for the potential
herbicide reduction when using the MoDiCoVi algorithm all the
while requiring a efficacy comparable to conventional spraying. In
June 2013 a maize field located in Denmark were seeded. The field
was divided into parcels which was assigned to one of two main
groups: 1) Control, consisting of subgroups of no spray and full dose
spraty; 2) MoDiCoVi algorithm subdivided into five different leaf
cover thresholds for spray activation. In addition approximately 25%
of the parcels were seeded with additional weeds perpendicular to
the maize rows. In total 299 parcels were randomly assigned with
the 28 different treatment combinations. In the statistical analysis,
bootstrapping was used for balancing the number of replicates. The
achieved potential herbicide savings was found to be 70% to 95%
depending on the initial weed coverage. However additional field
trials covering more seasons and locations are needed to verify
the generalisation of these results. There is a potential for further
herbicide savings as the time interval between the first and second
spraying session was not long enough for the weeds to turn yellow,
instead they only stagnated in growth.", keywords = "Weed crop discrimination, macrosprayer,
herbicide reduction, site-specific, sprayer-boom.", volume = "11", number = "4", pages = "272-10", }
