Centre de documentation
scientifique

A CNN approach to simultaneously count plants and detect plantation-rows from UAV imagery / Lucas Prado Osco in ISPRS Journal of photogrammetry and remote sensing, vol 174 (April 2021)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 174 (April 2021) . - pp 1 - 17 Titre : A CNN approach to simultaneously count plants and detect plantation-rows from UAV imagery Type de document : Article/Communication Auteurs : Lucas Prado Osco, Auteur ; Mauro Dos Santos de Arruda, Auteur ; Diogo Nunes Gonçalves, Auteur ; et al., Auteur Année de publication : 2021 Article en page(s) : pp 1 - 17 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] apprentissage profond [Termes IGN] carte agricole [Termes IGN] Citrus sinensis [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] comptage [Termes IGN] cultures [Termes IGN] détection d'objet [Termes IGN] extraction de la végétation [Termes IGN] gestion durable [Termes IGN] image captée par drone [Termes IGN] maïs (céréale) [Termes IGN] rendement agricole Résumé : (auteur) Accurately mapping croplands is an important prerequisite for precision farming since it assists in field management, yield-prediction, and environmental management. Crops are sensitive to planting patterns and some have a limited capacity to compensate for gaps within a row. Optical imaging with sensors mounted on Unmanned Aerial Vehicles (UAV) is a cost-effective option for capturing images covering croplands nowadays. However, visual inspection of such images can be a challenging and biased task, specifically for detecting plants and rows on a one-step basis. Thus, developing an architecture capable of simultaneously extracting plant individually and plantation-rows from UAV-images is yet an important demand to support the management of agricultural systems. In this paper, we propose a novel deep learning method based on a Convolutional Neural Network (CNN) that simultaneously detects and geolocates plantation-rows while counting its plants considering highly-dense plantation configurations. The experimental setup was evaluated in (a) a cornfield (Zea mays L.) with different growth stages (i.e. recently planted and mature plants) and in a (b) Citrus orchard (Citrus Sinensis Pera). Both datasets characterize different plant density scenarios, in different locations, with different types of crops, and from different sensors and dates. This scheme was used to prove the robustness of the proposed approach, allowing a broader discussion of the method. A two-branch architecture was implemented in our CNN method, where the information obtained within the plantation-row is updated into the plant detection branch and retro-feed to the row branch; which are then refined by a Multi-Stage Refinement method. In the corn plantation datasets (with both growth phases – young and mature), our approach returned a mean absolute error (MAE) of 6.224 plants per image patch, a mean relative error (MRE) of 0.1038, precision and recall values of 0.856, and 0.905, respectively, and an F-measure equal to 0.876. These results were superior to the results from other deep networks (HRNet, Faster R-CNN, and RetinaNet) evaluated with the same task and dataset. For the plantation-row detection, our approach returned precision, recall, and F-measure scores of 0.913, 0.941, and 0.925, respectively. To test the robustness of our model with a different type of agriculture, we performed the same task in the citrus orchard dataset. It returned an MAE equal to 1.409 citrus-trees per patch, MRE of 0.0615, precision of 0.922, recall of 0.911, and F-measure of 0.965. For the citrus plantation-row detection, our approach resulted in precision, recall, and F-measure scores equal to 0.965, 0.970, and 0.964, respectively. The proposed method achieved state-of-the-art performance for counting and geolocating plants and plant-rows in UAV images from different types of crops. The method proposed here may be applied to future decision-making models and could contribute to the sustainable management of agricultural systems. Numéro de notice : A2021-205 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2021.01.024 Date de publication en ligne : 13/02/2021 En ligne : https://doi.org/10.1016/j.isprsjprs.2021.01.024 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=97171 [article]

Exemplaires(3)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2021041	SL	Revue	Centre de documentation	Revues en salle	Disponible
081-2021043	DEP-RECP	Revue	LASTIG	Dépôt en unité	Exclu du prêt
081-2021042	DEP-RECF	Revue	Nancy	Dépôt en unité	Exclu du prêt

Analysis of full-waveform LiDAR data for classification of an orange orchard scene / Karolina D. Fieber in ISPRS Journal of photogrammetry and remote sensing, vol 82 (August 2013)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 82 (August 2013) . - pp 63 - 82 Titre : Analysis of full-waveform LiDAR data for classification of an orange orchard scene Type de document : Article/Communication Auteurs : Karolina D. Fieber, Auteur ; Ian J. Davenport, Auteur ; James M. Ferryman, Auteur ; et al., Auteur Année de publication : 2013 Article en page(s) : pp 63 - 82 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] Citrus sinensis [Termes IGN] classification [Termes IGN] données lidar [Termes IGN] lidar à retour d'onde complète [Termes IGN] verger Résumé : (Auteur) Full-waveform laser scanning data acquired with a Riegl LMS-Q560 instrument were used to classify an orange orchard into orange trees, grass and ground using waveform parameters alone. Gaussian decomposition was performed on this data capture from the National Airborne Field Experiment in November 2006 using a custom peak-detection procedure and a trust-region-reflective algorithm for fitting Gauss functions. Calibration was carried out using waveforms returned from a road surface, and the backscattering coefficient Y was derived for every waveform peak. The processed data were then analysed according to the number of returns detected within each waveform and classified into three classes based on pulse width and Y. For single-peak waveforms the scatterplot of Y versus pulse width was used to distinguish between ground, grass and orange trees. In the case of multiple returns, the relationship between first (or first plus middle) and last return Y values was used to separate ground from other targets. Refinement of this classification, and further sub-classification into grass and orange trees was performed using the Y versus pulse width scatterplots of last returns. In all cases the separation was carried out using a decision tree with empirical relationships between the waveform parameters. Ground points were successfully separated from orange tree points. The most difficult class to separate and verify was grass, but those points in general corresponded well with the grass areas identified in the aerial photography. The overall accuracy reached 91%, using photography and relative elevation as ground truth. The overall accuracy for two classes, orange tree and combined class of grass and ground, yielded 95%. Finally, the backscattering coefficient Y of single-peak waveforms was also used to derive reflectance values of the three classes. The reflectance of the orange tree class (0.31) and ground class (0.60) are consistent with published values at the wavelength of the Riegl scanner (1550 nm). The grass class reflectance (0.46) falls in between the other two classes as might be expected, as this class has a mixture of the contributions of both vegetation and ground reflectance properties. Numéro de notice : A2013-412 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2013.05.002 En ligne : https://doi.org/10.1016/j.isprsjprs.2013.05.002 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=32550 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2013081	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Monitoring water stress and fruit quality in an orange orchard under regulated deficit irrigation using narrow-band structural and physiological remote sensing indices / S. Stagakis in ISPRS Journal of photogrammetry and remote sensing, vol 71 (July 2012)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 71 (July 2012) . - pp 47 - 61 Titre : Monitoring water stress and fruit quality in an orange orchard under regulated deficit irrigation using narrow-band structural and physiological remote sensing indices Type de document : Article/Communication Auteurs : S. Stagakis, Auteur ; V. Gonzales-Dugo, Auteur ; P. Cid, Auteur ; et al., Auteur Année de publication : 2012 Article en page(s) : pp 47 - 61 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] Citrus sinensis [Termes IGN] cultures irriguées [Termes IGN] drone [Termes IGN] image aérienne [Termes IGN] image multibande [Termes IGN] Séville [Termes IGN] stress hydrique [Termes IGN] verger Résumé : (Auteur) This paper deals with the monitoring of water status and the assessment of the effect of stress on citrus fruit quality using structural and physiological remote sensing indices. Four flights were conducted over a citrus orchard in 2009 using an unmanned aerial vehicle (UAV) carrying a multispectral camera with six narrow spectral bands in the visible and near infrared. Physiological indices such as the Photochemical Reflectance Index (PRI570), a new structurally robust PRI formulation that uses the 515 nm as the reference band (PRI515), and a chlorophyll ratio (R700/R670) were compared against the Normalized Difference Vegetation Index (NDVI), Renormalized Difference Vegetation Index (RDVI) and Modified Triangular Vegetation Index (MTVI) canopy structural indices for their performance in tracking water status and the effects of sustained water stress on fruit quality at harvest. The irrigation setup in the commercial orchard was compared against a treatment scheduled to satisfy full requirements (based on estimated crop evapotranspiration) using two regulated deficit irrigation (RDI) strategies. The water status of the trees throughout the experiment was monitored with frequent field measurements of stem water potential (?x), while titratable acidity (TA) and total soluble solids (TSS) were measured at harvest on selected trees from each irrigation treatment. The high spatial resolution of the multispectral imagery (30 cm pixel size) enabled identification of pure tree crown components, extracting the tree reflectance from shaded, sunlit and aggregated pixels. The physiological and structural indices were then calculated from each tree at the following levels: (i) pure sunlit tree crown, (ii) entire crown, aggregating the within-crown shadows, and (iii) simulating a lower resolution pixel, including tree crown, sunlit and shaded soil pixels. The resulting analysis demonstrated that both PRI formulations were able to track water status, except when water stress altered canopy structure. In such cases, PRI570 was more affected than PRI515 by the structural changes caused by sustained water stress throughout the season. Both PRI formulations were proven to serve as pre-visual water stress indicators linked to fruit quality TSS and TA parameters (r2 = 0.69 for PRI515 vs TSS; r2 = 0.58 vs TA). In contrast, the chlorophyll (R700/R670) and structural indices (NDVI, RDVI, MTVI) showed poor relationships with fruit quality and water status levels (r2 = 0.04 for NDVI vs TSS; r2 = 0.19 vs TA). The two PRI formulations showed strong relationships with the field-measured fruit quality parameters in September, the beginning of stage III, which appeared to be the period most sensitive to water stress and the most critical for assessing fruit quality in citrus. Both PRI515 and PRI570 showed similar performance for the two scales assessed (sunlit crown and entire crown), demonstrating that within-crown component separation is not needed in citrus tree crowns where the shaded vegetation component is small. However, the simulation conducted through spatial resampling on tree + soil aggregated pixels revealed that the physiological indices were highly affected by soil reflectance and between-tree shadows, showing that for TSS vs PRI515 the relationship dropped from r2 = 0.69 to r2 = 0.38 when aggregating soil + crown components. This work confirms a previous study that demonstrated the link between PRI570, water stress, and fruit quality, while also making progress in assessing the new PRI formulation (PRI515), the within-crown shadow effects on the physiological indices, and the need for high resolution imagery to target individual tree crowns for the purpose of evaluating the effects of water stress on fruit quality in citrus. Numéro de notice : A2012-347 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2012.05.003 En ligne : https://doi.org/10.1016/j.isprsjprs.2012.05.003 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=31793 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2012051	SL	Revue	Centre de documentation	Revues en salle	Disponible

Development of daily spatial heat unit mapping from monthly climatic surfaces for the Australian continent / Nicholas C. Coops in International journal of geographical information science IJGIS, vol 15 n° 4 (june 2001)  

Public [article]  inInternational journal of geographical information science IJGIS > vol 15 n° 4 (june 2001) . - pp 345 - 361 Titre : Development of daily spatial heat unit mapping from monthly climatic surfaces for the Australian continent Type de document : Article/Communication Auteurs : Nicholas C. Coops, Auteur ; A. Loughhead, Auteur ; P. Ryan, Auteur ; R. Hutton, Auteur Année de publication : 2001 Article en page(s) : pp 345 - 361 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Systèmes d'information géographique [Termes IGN] analyse spatiale [Termes IGN] Australie [Termes IGN] Citrus sinensis [Termes IGN] climat aride [Termes IGN] seuillage d'image [Termes IGN] système d'information géographique [Termes IGN] température de l'air Résumé : (Auteur) In absence of other limitations, the growth rate of a plant is dependent upon the amount of heat it receives. Each species, whether a crop, weed or disease organism, is adapted to grow at its optimum rate within a specific temperature range. Within this range, the growing degree days (GDD) is the heat accumulation above a given base temperature for a specific time period, such as a crop's growing season or phenological stage. In this paper we detail a methodology to predict GDD for synthetically generated average growing seasons derived from long term average climate data over the Australian continent. An application of these techniques has been made using the GEODATA 9 second DEM, with temperature threshold values estimated to characterize optimum growth in citrus (Citrus sinensis (L) Osbeek). Three major determinants of the annual growth cycle of Citrus sp. were established and predicted on a spatial basis including the starting day of the growing season, the GDD for a growing season, and the time required to accumulate an arbitrarily selected 2000 GDD from the estimated starting day. When these critical environmental factors are expressed on a spatial basis, covering the Australian continent, the combination can be used to identify locations where new crop varieties can most effectively be grown to maximize fruit quality and productivity, or to extend the harvest season. Likewise, new germplasm introduced to Australia from overseas can be horticulturally assessed at sites climatically matched to the source location. Numéro de notice : A2001-042 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/INFORMATIQUE Nature : Article DOI : 10.1080/13658810010011401 En ligne : https://doi.org/10.1080/13658810010011401 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=21744 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
079-01041	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Determination of frosts in orange groves from NOAA-9 AVHRR data / V. Caselles in Remote sensing of environment, vol 29 n° 2 (01/08/1989)  

Public [article]  inRemote sensing of environment > vol 29 n° 2 (01/08/1989) . - pp 135 - 146 Titre : Determination of frosts in orange groves from NOAA-9 AVHRR data Type de document : Article/Communication Auteurs : V. Caselles, Auteur ; J.A. Sobrino, Auteur Année de publication : 1989 Article en page(s) : pp 135 - 146 Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] Citrus sinensis [Termes IGN] emissivité [Termes IGN] Espagne [Termes IGN] gelée [Termes IGN] image NOAA-AVHRR [Termes IGN] verger Numéro de notice : A1989-360 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/0034-4257(89)90022-9 En ligne : https://doi.org/10.1016/0034-4257(89)90022-9 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=25319 [article]