RT Journal Article
T1 Automated Identification of Crop Tree Crowns from UAV Multispectral Imagery by Means of Morphological Image Analysis
A1 Sarabia, Ricardo
A1 Aquino Martín, Arturo
A1 Ponce Real, Juan Manuel
A1 López, Gilberto
A1 Andújar Márquez, José Manuel
AB Within the context of precision agriculture, goods insurance, public subsidies, fire damageassessment, etc., accurate knowledge about the plant population in crops represents valuableinformation. In this regard, the use of Unmanned Aerial Vehicles (UAVs) has proliferated as analternative to traditional plant counting methods, which are laborious, time demanding and proneto human error. Hence, a methodology for the automated detection, geolocation and counting ofcrop trees in intensive cultivation orchards from high resolution multispectral images, acquired byUAV-based aerial imaging, is proposed. After image acquisition, the captures are processed by meansof photogrammetry to yield a 3D point cloud-based representation of the study plot. To exploit theelevation information contained in it and eventually identify the plants, the cloud is deterministicallyinterpolated, and subsequently transformed into a greyscale image. This image is processed, byusing mathematical morphology techniques, in such a way that the absolute height of the treeswith respect to their local surroundings is exploited to segment the tree pixel-regions, by globalstatistical thresholding binarization. This approach makes the segmentation process robust againstsurfaces with elevation variations of any magnitude, or to possible distracting artefacts with heightslower than expected. Finally, the segmented image is analysed by means of an ad-hoc momentrepresentation-based algorithm to estimate the location of the trees. The methodology was testedin an intensive olive orchard of 17.5 ha, with a population of 3919 trees. Because of the plot’s plantdensity and tree spacing pattern, typical of intensive plantations, many occurrences of intra-row treeaggregations were observed, increasing the complexity of the scenario under study. Notwithstanding,it was achieved a precision of 99.92%, a sensibility of 99.67% and an F-score of 99.75%, thus correctlyidentifying and geolocating 3906 plants. The generated 3D point cloud reported root-mean squareerrors (RMSE) in the X, Y and Z directions of 0.73 m, 0.39 m and 1.20 m, respectively. These resultssupport the viability and robustness of this methodology as a phenotyping solution for the automatedplant counting and geolocation in olive orchards.
PB MDPI
SN 2072-4292
YR 2020
FD 2020-02
LK http://hdl.handle.net/10272/18271
UL http://hdl.handle.net/10272/18271
LA eng
NO Sarabia, R., Aquino Martín, A., Ponce  Real, J. M., López, G., & Andújar Márquez, J. M. (2020). Automated Identification of Crop Tree Crowns from UAV Multispectral Imagery by Means of Morphological Image Analysis. Remote Sensing, 12(5), 748. DOI: https://doi.org/10.3390/rs12050748
DS Repositorio Institucional de la Universidad de Huelva
RD 30 may 2026