Suchergebnisse
Filter
13 Ergebnisse
Sortierung:
Enhancing Urban Brownfield Regeneration to Pursue Sustainable Community Outcomes through Dynamic Performance Governance
In: International journal of public administration, Band 44, Heft 2, S. 100-114
ISSN: 1532-4265
Insect counting through deep learning-based density maps estimation
In: Computers and electronics in agriculture: COMPAG online ; an international journal, Band 197, S. 106933
Deep learning-based segmentation of multiple species of weeds and corn crop using synthetic and real image datasets
In: Computers and Electronics in Agriculture, Band 194, S. 106719
Few-Shot Learning approach for plant disease classification using images taken in the field
In: Computers and Electronics in Agriculture, Band 175, S. 105542
Analysis of Few-Shot Techniques for Fungal Plant Disease Classification and Evaluation of Clustering Capabilities Over Real Datasets
Plant fungal diseases are one of the most important causes of crop yield losses. Therefore, plant disease identification algorithms have been seen as a useful tool to detect them at early stages to mitigate their effects. Although deep-learning based algorithms can achieve high detection accuracies, they require large and manually annotated image datasets that is not always accessible, specially for rare and new diseases. This study focuses on the development of a plant disease detection algorithm and strategy requiring few plant images (Few-shot learning algorithm). We extend previous work by using a novel challenging dataset containing more than 100,000 images. This dataset includes images of leaves, panicles and stems of five different crops (barley, corn, rape seed, rice, and wheat) for a total of 17 different diseases, where each disease is shown at different disease stages. In this study, we propose a deep metric learning based method to extract latent space representations from plant diseases with just few images by means of a Siamese network and triplet loss function. This enhances previous methods that require a support dataset containing a high number of annotated images to perform metric learning and few-shot classification. The proposed method was compared over a traditional network that was trained with the cross-entropy loss function. Exhaustive experiments have been performed for validating and measuring the benefits of metric learning techniques over classical methods. Results show that the features extracted by the metric learning based approach present better discriminative and clustering properties. Davis-Bouldin index and Silhouette score values have shown that triplet loss network improves the clustering properties with respect to the categorical-cross entropy loss. Overall, triplet loss approach improves the DB index value by 22.7% and Silhouette score value by 166.7% compared to the categorical cross-entropy loss model. Moreover, the F-score parameter obtained from the Siamese network with the triplet loss performs better than classical approaches when there are few images for training, obtaining a 6% improvement in the F-score mean value. Siamese networks with triplet loss have improved the ability to learn different plant diseases using few images of each class. These networks based on metric learning techniques improve clustering and classification results over traditional categorical cross-entropy loss networks for plant disease identification. ; This project was partially supported by the Spanish Government through CDTI Centro para el Desarrollo Tecnológico e Industrial project AI4ES (ref CER-20211030), by the University of the Basque Country (UPV/EHU) under grant COLAB20/01 and by the Basque Government through grant IT1229-19.
BASE
MRI Deep Learning-Based Solution for Alzheimer's Disease Prediction
Background: Alzheimer's is a degenerative dementing disorder that starts with a mild memory impairment and progresses to a total loss of mental and physical faculties. The sooner the diagnosis is made, the better for the patient, as preventive actions and treatment can be started. Al though tests such as the Mini-Mental State Tests Examination are usually used for early identification, diagnosis relies on magnetic resonance imaging (MRI) brain analysis. Methods: Public initiatives such as the OASIS (Open Access Series of Imaging Studies) collection provide neuroimaging datasets openly available for research purposes. In this work, a new method based on deep learning and image processing techniques for MRI-based Alzheimer's diagnosis is proposed and compared with previous literature works. Results: Our method achieves a balance accuracy (BAC) up to 0.93 for image-based automated diagnosis of the disease, and a BAC of 0.88 for the establishment of the disease stage (healthy tissue, very mild and severe stage). Conclusions: Results obtained surpassed the state-of-the-art proposals using the OASIS collection. This demonstrates that deep learning-based strategies are an effective tool for building a robust solution for Alzheimer's-assisted diagnosis based on MRI data. ; This work was partially supported by the SUPREME project. This project has received funding from the Basque Government's Industry Department HAZITEK program under agreement ZE-2019/00022. This research has also received funding from the Basque Government's Industry Department under the ELKARTEK program's project ONKOTOOLS under agreement KK-2020/00069
BASE
Ultrafast electron diffraction tomography for structure determination of the new zeolite ITQ-58
In this work a new ultrafast data collection strategy for electron diffraction tomography is presented that allows reducing data acquisition time by one order of magnitude. This methodology minimizes the radiation damage of beam-sensitive materials, such as microporous materials. This method, combined with the precession of the electron beam, provides high quality data enabling the determination of very complex structures. Most importantly, the implementation of this new electron diffraction methodology is easily affordable in any modern electron microscope. As a proof of concept, we have solved a new highly complex zeolitic structure named ITQ-58, with a very low symmetry (triclinic) and a large unit cell volume (1874.6 Å), containing 16 silicon and 32 oxygen atoms in its asymmetric unit, which would be very difficult to solve with the state of the art techniques. ; The authors gratefully acknowledge financial support of European Research Council (ERC-2014-AdG ref 671093 "MATching zeolite SYNthesis with CATalytic activity"), Spanish Government (MAT2015-71842-P (MINECO/FEDER), MAT2012-38567-C02-01 and Severo Ochoa SEV-2012-0267), and Generalitat Valenciana (Project Prometeo). E.M. was also supported by the Italian project FIR2013 Exploring the Nanoworld. ; Peer Reviewed
BASE
Ultrafast Electron Diffraction Tomography for Structure Determination of the New Zeolite ITQ-58
[EN] In this work a new ultrafast data collection strategy for electron diffraction tomography is presented that allows reducing data acquisition tithe by one order of Magnitude.,,This Methodology minimizes the radiation damage of, beam-sensitive materials, such as microporous materials. This method, combined with the precession of the electron beam, provides high quality data enabling the determination of very complex structures. Most importantly,, the implementation of this new, electron diffraction methodology is, easily affordable in any modem electron Microscope. As a proof of concept, we have solved a new highly complex zeolitic structure named ITQ-58, with a very low symmetry (triclinic) and a large unit cell volume (1874.61(3)), containing 16 silicon and 32 oxygen atoms in its asymmetric unit, which would be very difficult to solve with the state of the art techniques. ; The authors gratefully acknowledge financial support of European Research Council (ERC-2014-AdG ref 671093 "MATching zeolite SYNthesis with CATalytic activity"), Spanish Government (MAT2015-71842-P (MINECO/FEDER), MAT2012-38567-C02-01 and Severo Ochoa SEV-2012-0267), and Generalitat Valenciana (Project Prometeo). E.M. was also supported by the Italian project FIR2013 Exploring the Nanoworld. Authors thank beamline MSPD at Spanish Synchrotron ALBA for beam time allocation. We acknowledge JEOL Japan, JEOL Europe, and specially Junichi Morimoto from JEOL Europe for the assistance during the specific alignments of the microscope. Finally, authors specially thank the Electron Microscopy Service of the Universitat Politecnica de Valencia and, in particular, Manuel J. Planes and Jose L. Moya for their invaluable support for setting up the EDT data acquisition. Original data (collected EDT frames and integrated intensities) can be requested directly from corresponding authors. ; Simancas-Coloma, J.; Simancas Coloma, R.; Bereciartua-Pérez, PJ.; Jorda Moret, JL.; Rey Garcia, F.; Corma Canós, A.; Nicolopoulos ., S. (2016). Ultrafast Electron ...
BASE
Synthesis and structure determination via ultra-fast electron diffraction of the new microporous zeolitic germanosilicate ITQ-62
2122 2125 54 17 ; S Ch. Baerlocher , A.Hepp and W. M.Meier , DLS-76. Distance least squares refinement program , ETH Zurich Institut fur Kristallographie , Zurich, Switzerland , 1977 ; Sun, J., Bonneau, C., Cantín, Á., Corma, A., Díaz-Cabañas, M. J., Moliner, M., … Zou, X. (2009). The ITQ-37 mesoporous chiral zeolite. Nature, 458(7242), 1154-1157. doi:10.1038/nature07957 ; [EN] Here, we present the synthesis and structure determination of the new zeolite ITQ-62. Its structure was determined via ultra-fast electron diffraction tomography and refined using powder XRD data of the calcined material. This new zeolite contains a tridirectional channel system of highly distorted 8-rings, as well as a monodirectional 12-ring channel system. The authors gratefully acknowledge financial support from the Spanish Government (MAT2015-71842-P and MAT2015-71261-R MINECO/FEDER and Severo Ochoa SEV-2016-0683). The authors thank ALBA Light Source for beam allocation at the beamline MSPD, and specially thank the Electron Microscopy Service of the Universitat Politecnica de Valencia. Finally, the authors thank Dr Alejandro Vidal and Dr Teresa Blasco for helping in the NMR data discussion. Bieseki, L.; Simancas Coloma, R.; Jorda Moret, JL.; Bereciartua-Pérez, PJ.; Cantin Sanz, A.; Simancas-Coloma, J.; Pergher, SB. (2018). Synthesis and structure determination via ultra-fast electron diffraction of the new microporous zeolitic germanosilicate ITQ-62. Chemical Communications. 54(17):2122-2125. https://doi.org/10.1039/c7cc09240g Barrer, R. M., & Denny, P. J. (1961). 201. Hydrothermal chemistry of the silicates. Part IX. Nitrogenous aluminosilicates. Journal of the Chemical Society (Resumed), 971. doi:10.1039/jr9610000971 Kerr, G. T. (1966). Chemistry of Crystalline Aluminosilicates. II. The Synthesis and Properties of Zeolite ZK-4. Inorganic Chemistry, 5(9), 1537-1539. doi:10.1021/ic50043a015 Burton, A. W., & Zones, S. I. (2007). Organic Molecules in Zeolite Synthesis: Their Preparation and Structure-Directing Effects. ...
BASE
Synthesis and structure determination via ultra-fast electron diffraction of the new microporous zeolitic germanosilicate ITQ-62
[EN] Here, we present the synthesis and structure determination of the new zeolite ITQ-62. Its structure was determined via ultra-fast electron diffraction tomography and refined using powder XRD data of the calcined material. This new zeolite contains a tridirectional channel system of highly distorted 8-rings, as well as a monodirectional 12-ring channel system. ; The authors gratefully acknowledge financial support from the Spanish Government (MAT2015-71842-P and MAT2015-71261-R MINECO/FEDER and Severo Ochoa SEV-2016-0683). The authors thank ALBA Light Source for beam allocation at the beamline MSPD, and specially thank the Electron Microscopy Service of the Universitat Politecnica de Valencia. Finally, the authors thank Dr Alejandro Vidal and Dr Teresa Blasco for helping in the NMR data discussion. ; Bieseki, L.; Simancas Coloma, R.; Jorda Moret, JL.; Bereciartua-Pérez, PJ.; Cantin Sanz, A.; Simancas-Coloma, J.; Pergher, SB. (2018). Synthesis and structure determination via ultra-fast electron diffraction of the new microporous zeolitic germanosilicate ITQ-62. Chemical Communications. 54(17):2122-2125. https://doi.org/10.1039/c7cc09240g ; S ; 2122 ; 2125 ; 54 ; 17 ; Barrer, R. M., & Denny, P. J. (1961). 201. Hydrothermal chemistry of the silicates. Part IX. Nitrogenous aluminosilicates. Journal of the Chemical Society (Resumed), 971. doi:10.1039/jr9610000971 ; Kerr, G. T. (1966). Chemistry of Crystalline Aluminosilicates. II. The Synthesis and Properties of Zeolite ZK-4. Inorganic Chemistry, 5(9), 1537-1539. doi:10.1021/ic50043a015 ; Burton, A. W., & Zones, S. I. (2007). Organic Molecules in Zeolite Synthesis: Their Preparation and Structure-Directing Effects. Introduction to Zeolite Science and Practice, 137-179. doi:10.1016/s0167-2991(07)80793-2 ; Zones, S. I., Nakagawa, Y., Lee, G. S., Chen, C. Y., & Yuen, L. T. (1998). Searching for new high silica zeolites through a synergy of organic templates and novel inorganic conditions. Microporous and Mesoporous Materials, 21(4-6), 199-211. ...
BASE
280 one-opposition near-earth asteroids recovered by the EURONEAR with the Isaac Newton telescope
Context. One-opposition near-Earth asteroids (NEAs) are growing in number, and they must be recovered to prevent loss and mismatch risk, and to improve their orbits, as they are likely to be too faint for detection in shallow surveys at future apparitions. Aims. We aimed to recover more than half of the one-opposition NEAs recommended for observations by the Minor Planet Center (MPC) using the Isaac Newton Telescope (INT) in soft-override mode and some fractions of available D-nights. During about 130 h in total between 2013 and 2016, we targeted 368 NEAs, among which 56 potentially hazardous asteroids (PHAs), observing 437 INT Wide Field Camera (WFC) fields and recovering 280 NEAs (76% of all targets). Methods. Engaging a core team of about ten students and amateurs, we used the THELI, Astrometrica, and the FindOrb software to identify all moving objects using the blink and track-And-stack method for the faintest targets and plotting the positional uncertainty ellipse from NEODyS. Results. Most targets and recovered objects had apparent magnitudes centered around V ~ 22.8 mag, with some becoming as faint as V ~ 24 mag. One hundred and three objects (representing 28% of all targets) were recovered by EURONEAR alone by Aug. 2017. Orbital arcs were prolonged typically from a few weeks to a few years; our oldest recoveries reach 16 years. The O-C residuals for our 1854 NEA astrometric positions show that most measurements cluster closely around the origin. In addition to the recovered NEAs, 22 000 positions of about 3500 known minor planets and another 10 000 observations of about 1500 unknown objects (mostly main-belt objects) were promptly reported to the MPC by our team. Four new NEAs were discovered serendipitously in the analyzed fields and were promptly secured with the INT and other telescopes, while two more NEAs were lost due to extremely fast motion and lack of rapid follow-up time. They increase the counting to nine NEAs discovered by the EURONEAR in 2014 and 2015. Conclusions. Targeted projects to recover one-opposition NEAs are efficient in override access, especially using at least two-meter class and preferably larger field telescopes located in good sites, which appear even more efficient than the existing surveys.© ESO, 2018. ; I.O. acknowledges support from the European Research Council (ERC) in the form of Advanced Grant, cosmicism. R.T. acknowledges funding for her La Palma trip to Armagh Observatory, which is core-funded by the Northern Ireland Government. The research led by BTG, CJM, and NPGF has received funding from the European Research Council under the European Union's Seventh Framework Programme (FP/2007-2013) / ERC Grant Agreement No. 320964 (WDTracer). ; Peer Reviewed
BASE
First EURONEAR NEA discoveries from La Palma using the INT
© 2015 The Authors. Since 2006, the European Near Earth Asteroids Research (EURONEAR) project has been contributing to the research of near-Earth asteroids (NEAs) within a European network. One of the main aims is the amelioration of the orbits of NEAs, and starting in 2014 February we focus on the recovery of one-opposition NEAs using the Isaac Newton Telescope (INT) in La Palma in override mode. Part of this NEA recovery project, since 2014 June EURONEAR serendipitously started to discover and secure the first NEAs from La Palma and using the INT, thanks to the teamwork including amateurs and students who promptly reduce the data, report discoveries and secure new objects recovered with the INT and few other telescopes from the EURONEAR network. Five NEAs were discovered with the INT, including 2014 LU14, 2014 NL52 (one very fast rotator), 2014 OL339 (the fourth known Earth quasi-satellite), 2014 SG143 (a quite large NEA), and 2014 VP. Another very fast moving NEA was discovered but was unfortunately lost due to lack of follow-up time. Additionally, another 14 NEA candidates were identified based on two models, all being rapidly followed-up using the INT and another 11 telescopes within the EURONEAR network. They include one object discovered by Pan-STARRS, two Mars crossers, two Hungarias, one Jupiter trojan, and other few inner main belt asteroids (MBAs). Using the INT and Sierra Nevada 1.5 m for photometry, then the Gran Telescopio de Canarias for spectroscopy, we derived the very rapid rotation of 2014 NL52, then its albedo, magnitude, size, and its spectral class. Based on the total sky coverage in dark conditions, we evaluate the actual survey discovery rate using 2-m class telescopes. One NEA is possible to be discovered randomly within minimum 2.8 deg 2 and maximum 5.5 deg 2 . These findings update our past statistics, being based on double sky coverage and taking into account the recent increase in discovery. ; KK acknowledges support from the Polish Narodowe Centrum Nauki Grant UMO-2011/01/D/ST9/00427. LVM has been also supported by Grant AYA2011-30491-C02-01, co-financed by MICINN and FEDER funds, and the Junta de Andalucia (Spain) Grant TIC-114. SG was supported by the Slovak Grant Agency for Science VEGA, grant no. 1/0225/14. The research leading to these results has received funding from the European Research Council under the European Union's Seventh Framework Programme (FP/2007-2013)/ERC Grant Agreement no. 320964 (WDTracer). ; Peer Reviewed
BASE