Search results for "Transfer of learning"

showing 10 items of 32 documents

Convolutional Neural Networks for Cloud Screening: Transfer Learning from Landsat-8 to Proba-V

2018

Cloud detection is a key issue for exploiting the information from Earth observation satellites multispectral sensors. For Proba-V, cloud detection is challenging due to the limited number of spectral bands. Advanced machine learning methods, such as convolutional neural networks (CNN), have shown to work well on this problem provided enough labeled data. However, simultaneous collocated information about the presence of clouds is usually not available or requires a great amount of manual labor. In this work, we propose to learn from the available Landsat −8 cloud masks datasets and transfer this learning to solve the Proba-V cloud detection problem. CNN are trained with Landsat images adap…

010504 meteorology & atmospheric sciencesComputer sciencebusiness.industryMultispectral image0211 other engineering and technologiesPattern recognitionCloud computing02 engineering and technologySpectral bands01 natural sciencesConvolutional neural networkData modelingKey (cryptography)Artificial intelligencebusinessTransfer of learning021101 geological & geomatics engineering0105 earth and related environmental sciencesIGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium
researchProduct

Transferring deep learning models for cloud detection between Landsat-8 and Proba-V

2020

Abstract Accurate cloud detection algorithms are mandatory to analyze the large streams of data coming from the different optical Earth observation satellites. Deep learning (DL) based cloud detection schemes provide very accurate cloud detection models. However, training these models for a given sensor requires large datasets of manually labeled samples, which are very costly or even impossible to create when the satellite has not been launched yet. In this work, we present an approach that exploits manually labeled datasets from one satellite to train deep learning models for cloud detection that can be applied (or transferred) to other satellites. We take into account the physical proper…

010504 meteorology & atmospheric sciencesExploitComputer sciencebusiness.industryDeep learning0211 other engineering and technologiesCloud detectionCloud computing02 engineering and technologyEarth observation satellitecomputer.software_genre01 natural sciencesConvolutional neural networkAtomic and Molecular Physics and OpticsComputer Science ApplicationsSatelliteData miningArtificial intelligenceComputers in Earth SciencesbusinessTransfer of learningEngineering (miscellaneous)computer021101 geological & geomatics engineering0105 earth and related environmental sciencesISPRS Journal of Photogrammetry and Remote Sensing
researchProduct

Biometric Fish Classification of Temperate Species Using Convolutional Neural Network with Squeeze-and-Excitation

2019

Our understanding and ability to effectively monitor and manage coastal ecosystems are severely limited by observation methods. Automatic recognition of species in natural environment is a promising tool which would revolutionize video and image analysis for a wide range of applications in marine ecology. However, classifying fish from images captured by underwater cameras is in general very challenging due to noise and illumination variations in water. Previous classification methods in the literature relies on filtering the images to separate the fish from the background or sharpening the images by removing background noise. This pre-filtering process may negatively impact the classificat…

0106 biological sciencesBiometricsComputer sciencebusiness.industry010604 marine biology & hydrobiologyPattern recognitionSharpening010603 evolutionary biology01 natural sciencesConvolutional neural networkBackground noiseA priori and a posterioriArtificial intelligenceUnderwaterbusinessTransfer of learningClassifier (UML)
researchProduct

Temperate Fish Detection and Classification: a Deep Learning based Approach

2021

A wide range of applications in marine ecology extensively uses underwater cameras. Still, to efficiently process the vast amount of data generated, we need to develop tools that can automatically detect and recognize species captured on film. Classifying fish species from videos and images in natural environments can be challenging because of noise and variation in illumination and the surrounding habitat. In this paper, we propose a two-step deep learning approach for the detection and classification of temperate fishes without pre-filtering. The first step is to detect each single fish in an image, independent of species and sex. For this purpose, we employ the You Only Look Once (YOLO) …

0106 biological sciencesFOS: Computer and information sciencesComputer Science - Machine LearningComputer scienceComputer Vision and Pattern Recognition (cs.CV)Computer Science - Computer Vision and Pattern Recognition010603 evolutionary biology01 natural sciencesConvolutional neural networkVDP::Matematikk og Naturvitenskap: 400::Informasjons- og kommunikasjonsvitenskap: 420Machine Learning (cs.LG)Artificial IntelligenceClassifier (linguistics)FOS: Electrical engineering electronic engineering information engineeringbusiness.industry010604 marine biology & hydrobiologyDeep learningImage and Video Processing (eess.IV)Process (computing)Pattern recognitionElectrical Engineering and Systems Science - Image and Video ProcessingObject detectionA priori and a posterioriNoise (video)Artificial intelligenceTransfer of learningbusiness
researchProduct

Automatic sleep scoring: A deep learning architecture for multi-modality time series

2020

Background: Sleep scoring is an essential but time-consuming process, and therefore automatic sleep scoring is crucial and urgent to help address the growing unmet needs for sleep research. This paper aims to develop a versatile deep-learning architecture to automate sleep scoring using raw polysomnography recordings. Method: The model adopts a linear function to address different numbers of inputs, thereby extending model applications. Two-dimensional convolution neural networks are used to learn features from multi-modality polysomnographic signals, a “squeeze and excitation” block to recalibrate channel-wise features, together with a long short-term memory module to exploit long-range co…

0301 basic medicineProcess (engineering)Computer sciencePolysomnographyPolysomnographyMachine learningcomputer.software_genreuni (lepotila)03 medical and health sciencesDeep Learning0302 clinical medicinepolysomnographymedicineHumansBlock (data storage)Sleep Stagesmedicine.diagnostic_testArtificial neural networksignaalinkäsittelybusiness.industryunitutkimusGeneral NeuroscienceDeep learningdeep learningsignaalianalyysiElectroencephalographyautomatic sleep scoringmulti-modality analysiskoneoppiminen030104 developmental biologyMemory moduleSleep StagesArtificial intelligenceSleepTransfer of learningbusinesscomputer030217 neurology & neurosurgeryJournal of Neuroscience Methods
researchProduct

Constraints representing a meta-stable régime facilitate exploration during practice and transfer of learning in a complex multi-articular task

2018

Previous investigations have shown that inducing meta-stability\ud in behavior can be achieved by overlapping affordances through constraint\ud manipulation, allowing cooperative and competitive tendencies to\ud functionally coexist. The purpose of this paper was to test a number of\ud conditions applying these design principles on performance during skills\ud practice and transfer. Of additional interest, was whether the existing\ud skill level interacted with the environmental properties of the\ud experimental tasks (varying indoor climbing routes). Two skill groups\ud practised on three routes per session over four separate sessions. At the\ud end of the final session, climbers undertook…

AdultComputer scienceTransfer PsychologyaffordancesBiophysicsseinäkiipeilyExperimental and Cognitive Psychologysiirto050105 experimental psychologySession (web analytics)Task (project management)Young Adult03 medical and health sciences0302 clinical medicinemeta-stabilityHuman–computer interactionTransfer (computing)HumansLearningharjoittelu0501 psychology and cognitive sciencesOrthopedics and Sports MedicineAffordanceta315skillComputingMilieux_MISCELLANEOUSlajitaidotModels Statisticalbusiness.industry[SCCO.NEUR]Cognitive science/Neuroscience05 social sciences030229 sport sciencesGeneral MedicinekäytäntöHandpracticeTest (assessment)Constraint (information theory)AthletesTouchmotorinen oppiminenClimbingExploratory BehaviorArtificial intelligencebusinessTransfer of learningconstraintstransfer
researchProduct

Educar

2015

El presente artículo describe una experiencia de formación en alternancia en un contexto de educación superior; en concreto, a partir de la iniciativa desarrollada por Florida Universitaria, centro adscrito a la Universidad de Valencia. La formación en alternancia implica alternar el aprendizaje entre períodos desarrollados en el centro educativo y momentos desarrollados en el contexto laboral. El objetivo del presente trabajo es identificar una iniciativa que, de forma empírica, muestre cómo estos periodos impactan en el aprendizaje de competencias de los estudiantes. La metodología que se ha aplicado en la experiencia de formación en alternancia se basa en el aprendizaje basado en problem…

Aprendizaje basado en problemassolución de problemasProblem-based learningUniversity-industry relationsCompetenciasaprendizaje basado en problemasformación alternadaCompetenciesEducationTransferencia del aprendizajeTransfer of learningAlternance trainingProfessional trainingAprenentatgeAprenentatge basat en problemesL7-991Col·laboració empresa-universitatFormació en alternançaCommunicationRelación universidad-empresaaprendizaje profesionalrelación universidad-empresaEducation (General)competenciasHuman-Computer InteractionRelació universitat-empresatransferencia del aprendizajeAprenentatge professional-- FormacióCompetènciesTransferència de l'aprenentatgeAprendizaje profesionalFormación alternada
researchProduct

Machine Learning VS Transfer Learning - Smart Camera Implementation for Face Authentication

2018

The aim of this paper is to highlight differences between classical machine learning and transfer learning applied to low cost real-time face authentication. Furthermore, in an access control context, the size of biometric data should be minimized so it can be stored on a remote personal media. These constraints have led us to compare only lightest versions of these algorithms. Transfer learning applied on Mobilenet v1 raises to 85% of accuracy, for a 457Ko model, with 3680s and 1.43s for training and prediction tasks. In comparison, the fastest integrated method (Random Forest) shows accuracy up to 90% for a 7,9Ko model, with a fifth of a second to be trained and a hundred of microseconds …

AuthenticationComputer sciencebusiness.industry05 social sciencesContext (language use)Access controlMachine learningcomputer.software_genre050105 experimental psychologyRandom forest03 medical and health sciences0302 clinical medicineFace (geometry)0501 psychology and cognitive sciencesArtificial intelligenceBiometric dataSmart camerabusinessTransfer of learningcomputer[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing030217 neurology & neurosurgeryComputingMilieux_MISCELLANEOUS[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing
researchProduct

Benefits of computer screen-based simulation in learning cardiac arrest procedures

2010

Medical Education 2010: 44: 716–722 Objectives  What is the best way to train medical students early so that they acquire basic skills in cardiopulmonary resuscitation as effectively as possible? Studies have shown the benefits of high-fidelity patient simulators, but have also demonstrated their limits. New computer screen-based multimedia simulators have fewer constraints than high-fidelity patient simulators. In this area, as yet, there has been no research on the effectiveness of transfer of learning from a computer screen-based simulator to more realistic situations such as those encountered with high-fidelity patient simulators. Methods  We tested the benefits of learning cardiac arre…

Class (computer programming)medicine.medical_specialtybusiness.industrymedicine.medical_treatmentLearning environmenteducationGeneral MedicineEducationTest (assessment)Basic skillsMedicineMedical physicsCardiopulmonary resuscitationClinical competencebusinessPatient simulationTransfer of learningMedical Education
researchProduct

Emergent Collective Behaviors in a Multi-agent Reinforcement Learning Pedestrian Simulation: A Case Study

2015

In this work, a Multi-agent Reinforcement Learning framework is used to generate simulations of virtual pedestrians groups. The aim is to study the influence of two different learning approaches in the quality of generated simulations. The case of study consists on the simulation of the crossing of two groups of embodied virtual agents inside a narrow corridor. This scenario is a classic experiment inside the pedestrian modeling area, because a collective behavior, specifically the lanes formation, emerges with real pedestrians. The paper studies the influence of different learning algorithms, function approximation approaches, and knowledge transfer mechanisms on performance of learned ped…

Collective behaviorFunction approximationbusiness.industryComputer scienceBellman equationVector quantizationProbabilistic logicReinforcement learningArtificial intelligencebusinessTransfer of learningKnowledge transferSimulation
researchProduct