Weed mapping using UAV-imagery

Title Weed mapping using UAV-imageryTitle (native language)

Category Recording or mapping technology

Short summary forpractitioners (Practiceabstract) in English)

This paper approaches the problem of weed mapping for precision agriculture, using imageryprovided by Unmanned Aerial Vehicles (UAVs) from sunflower and maize crops. Precision agriculturereferred to weed control is mainly based on the design of early post-emergence site-specific controltreatments according to weed coverage, where one of the most important challenges is the spectralsimilarity of crop and weed pixels in early growth stages. Our work tackles this problem in the contextof object-based image analysis (OBIA) by means of supervised machine learning methods combinedwith pattern and feature selection techniques, devising a strategy for alleviating the user intervention inthe system while not compromising the accuracy. This work firstly proposes a method for choosing aset of training patterns via clustering techniques so as to consider a representative set of the wholefield data spectrum for the classification method. Furthermore, a feature selection method is used toobtain the best discriminating features from a set of several statistics and measures of differentnature.

Short summary forpractitionersWebsiteAudiovisual materialLinks to other websitesAdditional commentsKeywords Farming practiceAdditional keywords Remote sensing; Unmanned aerial vehicles (UAV); Weed detection; Object based image analysisGeographical location(NUTS) EU

Other geographicallocationCropping systems Arable cropsField operations Crop and soil scoutingSFT users Farmer | ContractorEducation level of users AllFarm size (ha) 0-2 | 2-10 | 10-50 | 50-100 | 100-200 | 200-500 | >500

Scientific articleTitle Selecting patterns and features for between- and within- crop-row weed mapping using UAV-imagery

Full citation Pérez-Ortiz, M.; Peña, J.M.; Gutiérrez, P.A.; Torres-Sánchez, J.; Hervás-Martínez, C.; López-Granados, F.(2016). Expert Systems with Applications, DOI:10.1016/j.eswa.2015.10.043

Effects of this SFTProductivity (crop yield per ha) No effectQuality of product No effectRevenue profit farm income Some increaseSoil biodiversity No effectBiodiversity (other than soil) No effectInput costs Some decreaseVariable costs Some decreasePost-harvest crop wastage No effectEnergy use Some decreaseCH4 (methane) emission No effectCO2 (carbon dioxide) emission No effectN2O (nitrous oxide) emission No effectNH3 (ammonia) emission No effectNO3 (nitrate) leaching No effectFertilizer use No effectPesticide use No effectIrrigation water use No effectLabor time Some decreaseStress or fatigue for farmer No effectAmount of heavy physical labour No effectNumber and/or severity of personal injury accidents No effectNumber and/or severity of accidents resulting in spills property damage incorrectapplication of fertiliser/pesticides etc. No effect

Pesticide residue on product No effectWeed pressure Some decreasePest pressure (insects etc.) No effectDisease pressure (bacterial fungal viral etc.) No effect

Information related to how easy it is to start using the SFTThis SFT replaces a tool or technology that is currently used. The SFT is better than thecurrent tool no opinion

The SFT can be used without making major changes to the existing system no opinionThe SFT does not require significant learning before the farmer can use it disagreeThe SFT can be used in other useful ways than intended by the inventor agreeThe SFT has effects that can be directly observed by the farmer agreeUsing the SFT requires a large time investment by farmer no opinionThe SFT produces information that can be interpreted directly agree

View this technology on the Smart-AKIS platform.

This factsheet was generated on 2018-Apr-03 11:57:16.