Skip to main content

New remote sensing technologies for optimizing herbicide applications in weed-crop systems

Final Report Summary - TOAS (New remote sensing technologies for optimizing herbicide applications in weed-crop systems)

This project aimed to generate georeferenced weed infestation maps of selected annual crops (wheat, sunflower and maize) and permanent woody crops (olive-tree and poplar orchards) by using aerial images collected with an unmanned aerial vehicles (UAV) or drone. The specific objectives were concentrated on the configuration and use of the UAV and sensors for image acquisition, the evaluation of the specifications (sensor type, imagery characteristics, crop-weed phenological stage) required for each type of crop, and the development of automatic and robust image analysis procedures for crop assessment and weed mapping using the captured remote images in order to optimize the herbicide applications or other weed control operations.
The project reached all the planned objectives. In a first phase, a quadrocopter platform with vertical take-off and landing, model microdrones md4-1000, was chosen to collect the set of aerial images over the experimental crop-fields. These images were collected with two different sensors, a still point-and-shoot camera (model Olympus PEN E-PM1) and a six-band multispectral camera (model Tetracam mini-MCA-6), at several flight altitudes (between 30 and 100 m) with the objective of evaluating the optimal spectral and spatial resolutions needed for weed discrimination in early-season. In same of the fields, a multi-temporal study was also performed in order to study the optimal flight altitude as affected by type of sensor and crop, which allow establish an agreement among image spatial resolution, number of images per area unit, and flight length. Optimal period and flight altitude (subsequently, spatial resolution) for image acquisition in each crop system was determined according to results reported on weed discrimination.
The remote images were successfully collected in ten different parcels located in Cordoba, Seville, Jaen and Madrid. The main phases of the UAV workflow were defined: 1) mission planning, 2) UAV flight and image acquisition, and 3) image orthorectification and mosaicking. In this stage, it was quantified the effect of relationships of flight altitude with the image spatial resolution, area covered by each image and flight length. At a lower altitude, the number of images needed to cover the whole field may be a limiting factor due to the energy required for a greater flight length and computational requirements for the further mosaicking process. Spectral differences between weeds, crop and bare soil were also significant in the vegetation indices studied, mainly at a 30 m altitude. These results suggested that an agreement among spectral and spatial resolutions is needed to optimise the flight mission according to every agronomical objective as affected by the size of the smaller object to be discriminated (weed plants or weed patches). The results and recommendations reported in this piece of investigation is very helpful to make a decision about the optimal moment to collect the UAV images, the selection of the camera, and the configuration of the flights.
Next, image analysis procedures were developed in two separated phases as affected by the type of crop: annual or woody. On the one hand, customised and auto-adaptive algorithms were created for each of the annual crops studied, aiming to automatize the crop-weed-soil discrimination process in any crop-field condition. The successful of this investigation was associated to the development of advanced algorithms for the management and analysis of the UAV images. In this project, the object-based image analysis (OBIA) methodology was implemented in order to solve the limitations of the spectral similarity between pixels of weed and crop. On the one hand, customised and auto-adaptive algorithms were created for each of the annual crops studied, aiming to automatize the crop-weed-soil discrimination process in any crop-field condition. By using an advanced version of the Otsu algorithm, the method was implemented in all the crops with high accuracy (>85%). On the other hand, an innovative UAV-based procedure was developed in woody crops to compute the 3-dimensional features of the trees and to map tree and weed cover. The weed maps created in the previous objective were used to design a new map used for a site-specific weed control strategy. These new maps are customised in several formats (ASCII, shape-file, raster, table) and easily interpretable by a sprayer machinery with computer and navigation systems. Thus, the machinery could apply the weed treatment (e.g. herbicide) only in the weed-infested areas.
As conclusion, this project developed and tested the full protocol to create in-season post-emergence weed maps, as well as it identified and quantified each factor implied in this process has. The final results will allow generating efficient decision support system data that can be used to apply site-specific weed management strategies. In addition, these maps help to understand the linkages between crop development and field-related factors (e.g. weed emergences) and to optimize crop and weeding management operations in the context of precision agriculture with relevant economic and agro-environmental implications. The ultimate objective of this project is in line with the European policy for the Sustainable Use of Pesticides, which promotes reductions in herbicide applications and the utilisation of adequate doses according to the weed density. Obtaining weed distribution maps in early season by using imagery from UAVs is considered a milestone in Weed Research, thus generating evident economic benefits. The successful implementation of the TOAS project would lead to a 15-35% reduction in farm costs and a 20-30% decrease in the use of crop protection chemicals, doing evident the potential contributions to reduce agricultural cost and thus enhance European competitiveness in the agricultural regions.

Regarding the prospects of the research career development and re-integration of the fellow of this project, the investigator awarded in 2014 a “Ramon y Cajal” grand in the area of Agriculture that conducted to a 5-year contract in the Institute for Sustainable Agriculture (IAS-CSIC).

More information about this project and its evolution can be consulted in