A research managed by the IMAPING Group (Institute for Sustainable Agriculture, CSIC-Spanish National Research Council-Córdoba, Spain) in which Dra. Kelly has also participated (University of California-Berkeley-USA), has made possible the weed mapping in early season using multispectral and ultra-high spatial resolution imagery from an Unmanned Aerial Vehicle (UAV). The technology and the processes of image analysis have been published in the following journals: PLOS ONE*, Computers and Electronics in Agriculture*, and Sensors*. Other works on these topics have been published and can be followed in https://toasproject.wordpress.com/http://goo.gl/JeKd3S.
Weeds compete for light, space, water and nutrients with crops and it reduces the crop production in 35%. To avoid these losses, weed infestations must be controlled and the most used tool is herbicide application. Despite weeds are distributed in patches (Figure 1A) and is not necessary to apply herbicides in parts of the field where there are no weeds, weeds in current extensive agricultural systems are controlled with herbicides treatments in the whole field. It causes unnecessary use of herbicides and consequently a relevant economic and environmental impact. To avoid this situation, it is necessary to identify the crop areas with weed infestations and to determine their exact position in the field for addressing the herbicides applications only to infested areas and also deciding what herbicides must be applied depending on weed species or group of species (e.g., broadleaved, grass or resistant weeds). Therefore, it is necessary to detect, discriminate and map those crop areas infested by weeds. Our technological innovation resides in the application of flexible and robust remote sensing techniques based in OBIA (Object Based Image Analysis) and using an UAV (Figure 1B), which  is composed of a base station and can be equipped by several sensors with different spectral ranges, for example in visible range (Red-Blue-Green) or in visible and near infrared (Red-Green-Blue-NIR).
The UAV was programmed to fly at low altitude (30-120 m of altitude) so it allows to get very high spatial resolution images (pixels with several mm or few cm). Image acquisition must be accompanied by a field sampling (Figure 1C) that has been designed by our group and was necessary to validate the results of the image analysis. Our analysis procedures have made possible mapping the presence of weed patches showing that an average of 70% of the field was not infested and was not necessary to apply herbicides (Figure 1D).
Our studies have been performed in wheat, maize and sunflower crops. Currently, new research is been conducted in order to detect some problematic weeds (for example: herbicide resistant Papaver weeds) and crop volume in several wood crops. In fact, due to the extraordinary versatility of UAS, other researches in wood crops such as olive tree or poplar are been developed to control the architecture and volume of the trees, being an important base for a lot of agricultural works.
As conclusion, our results show that it is feasible to reduce the herbicide application in world-wide distributed crops, which has evident agro-economic and environmental benefits for farmers. “We are searching to apply an automatized and low-cost technology which is available even in cloudy days that can be adapted to the circumstances and numerous agronomic, environmental, forestry or any other objectives requiring the generation of maps”.  
This research has been funded projects: RHEA (ref.: 7FP-NMP-2009-LARGE-3, Grant Agreement: 245986, www.rhea-project.eu, TOAS (ref.: 7FP-PEOPLE-2011-CIG, Grant Agreement: 293991, http://toasproject.wordpress.com), TELEPLAM (MINECO-FEDER) and RECUPERA2020 (MINECO-FEDER, http://recupera.telegrafico.es/es).

* J. M. Peña-Barragán, J. Torres-Sánchez, A. Serrano-Pérez, A. I. de Castro-Megías and F. López-Granados. 2015. Quantifying efficacy and limits of Unmanned Aerial Vehicle (UAV) technology for weed seedling detection as affected by sensor resolution. Sensors, 15: 5609-5626 (open access). doi:10.3390/s150305609. Download PDF
*J. Torres-Sánchez, J.M. Peña-Barragán, A.I. de Castro-Megías and F. López-Granados. 2014. Multi-temporal mapping of the vegetation fraction in early-season wheat fields using images from UAV. Computers and Electronics in Agriculture, 103: 104-113.
*J. Torres-Sánchez, F. López-Granados, A.I. de Castro-Megías and J.M. Peña-Barragán. 2013. Configuration and Specifications of an Unmanned Aerial Vehicle (UAV) for Early Site Specific Weed Management. PLOS ONE, 8(3): e58210 (open access). doi:10.1371/journal.pone.0058210. Download PDF
*J.M. Peña-Barragán, J. Torres-Sánchez, A.I. de Castro-Megías, M. Kelly and F. López-Granados. 2013. Weed mapping in early-season maize fields using object-based analysis of unmanned aerial vehicle (UAV) images. PLOS ONE, 8(10): e77151 (open access). doi:10.1371/journal.pone.0077151. Download PDF

Utilizamos cookies en este sitio para mejorar su experiencia de usuario. Más información

Aviso de cookies