Search results for "image"

showing 10 items of 6818 documents

Comparison of feature importance measures as explanations for classification models

2021

AbstractExplainable artificial intelligence is an emerging research direction helping the user or developer of machine learning models understand why models behave the way they do. The most popular explanation technique is feature importance. However, there are several different approaches how feature importances are being measured, most notably global and local. In this study we compare different feature importance measures using both linear (logistic regression with L1 penalization) and non-linear (random forest) methods and local interpretable model-agnostic explanations on top of them. These methods are applied to two datasets from the medical domain, the openly available breast cancer …

feature importanceComputer scienceGeneral Chemical EngineeringGeneral Physics and Astronomy02 engineering and technologyinterpretable modelstekoälyMachine learningcomputer.software_genreLogistic regressionDomain (software engineering)020204 information systems0202 electrical engineering electronic engineering information engineeringFeature (machine learning)General Materials ScienceGeneral Environmental Scienceluokitus (toiminta)explainable artificial intelligencebusiness.industrylogistic regressionGeneral EngineeringRandom forestkoneoppiminenTrustworthinessInjury dataGeneral Earth and Planetary Sciences020201 artificial intelligence & image processingArtificial intelligencebusinesscomputerrandom forestSN Applied Sciences
researchProduct

A Curvature Based Method for Blind Mesh Visual Quality Assessment Using a General Regression Neural Network

2016

International audience; No-reference quality assessment is a challenging issue due to the non-existence of any information related to the reference and the unknown distortion type. The main goal is to design a computational method to objectively predict the human perceived quality of a distorted mesh and deal with the practical situation when the reference is not available. In this work, we design a no reference method that relies on the general regression neural network (GRNN). Our network is trained using the mean curvature which is an important perceptual feature representing the visual aspect of a 3D mesh. Relatively to the human subjective scores, the trained network successfully asses…

feature learning[ INFO.INFO-TS ] Computer Science [cs]/Signal and Image ProcessingComputer sciencemedia_common.quotation_subjectFeature extractiondistorted meshGRNNmean curvature02 engineering and technologyMachine learningcomputer.software_genreCurvaturevisual aspect representation[INFO.INFO-TS]Computer Science [cs]/Signal and Image ProcessingDistortioncomputational method0202 electrical engineering electronic engineering information engineeringFeature (machine learning)computational geometrymean opinion scoresQuality (business)Polygon meshmedia_commonArtificial neural networkbusiness.industrycompetitive scores Author Keywords Blind mesh visual quality assessmentperceptual feature020207 software engineeringregression analysis INSPEC: Non-Controlled Indexing curvature based methodblind mesh visual quality assessmentno-reference quality assessmentvisual qualityVisualizationgeneral regression neural network traininggeneral regression neural networkmesh generationneural netssubject scoreshuman perceived quality predictionhuman subjective scores020201 artificial intelligence & image processinglearning (artificial intelligence)Artificial intelligencepredicted objective scoresbusiness3D meshcomputer
researchProduct

Multi-agent control architecture for RFID cyberphysical robotic systems initial validation of tagged objects detection and identification using Playe…

2016

International audience; The objective of this paper is to describe and validate a multi-agent architecture proposed to control RFID Cyber-Physical Robotic Systems. This environment may contain human operators, robots (mobiles, manipulators, mobile manipulators, etc.), places (workrooms, walls, etc.) and other objects (tables, chairs, etc.). The proposed control architecture is composed of two types of agents dispatched on two levels. We find at the Organization level a Supervisory agent to allow operators to configure, manage and interact with the overall control system. At the Control level, we distinguish the Robots agents, to each robot (mobiles, manipulators or mobile manipulators) is a…

fiducial tagsComputer sciencemultiagent control architectureReal-time computing02 engineering and technologyRobot kinematicscyber-physical systemsCyber-physical systemPlayer/Stagerobot vicinity[SPI]Engineering Sciences [physics]mobile robotsrobot agentradiofrequency identificationRobot sensing systems0202 electrical engineering electronic engineering information engineering[ SPI ] Engineering Sciences [physics][INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO]supervisory agentFiducial interfaceRFID cyber-physical robotic systemsmanipulatorsRFIDmobile manipulatorsRobot kinematicsbusiness.industryRFID readershuman operators020208 electrical & electronic engineering[ INFO.INFO-RB ] Computer Science [cs]/Robotics [cs.RO]intelligent robotsObject recognitionObject (computer science)fiducial finderscontrol levelRobot controlCentralized databaseIdentification (information)Control systemEmbedded systemfiducial systemMultirobot systemsRobot020201 artificial intelligence & image processingRFID tagsbusinessFiducial marker
researchProduct

El cuerpo materno en la red: entre el orden de lo ?deseable? y el de lo real

2019

El estilo de vida fitness se realiza en, y a través de, un entramado subjetivo y de poder en el que circulan y se ensamblan dispositivos empresariales, terapéuticos, sanitarios y de espectacularización de los cuerpos, tendientes a inscribir la vida de la población en las tramas conflictivas de un capitalismo heterosexista, neoliberal e informacional. En dicho marco, la temporalidad, metamorfosis y plasticidad del cuerpo materno irrumpe como problema y objeto de intervención de tecnologías diversas orientadas a ajustar su tamaño, forma y devenir a los parámetros de la vida activa y saludable. En escenarios digitales, dicha tensión se materializa, por un lado, a través de dispositivos visuale…

fitmomsanitaryproductivitythis tension is performedmaternidadesentre el orden de lo ?deseable? y el de lo real Landa [1137-7038 8537 Arxius de sociologia 558787 2019 41 7605686 El cuerpo materno en la red]tending to inscribe the life of the population in the conflictive plots of a heterosexistdispositive 135 156focusing mainly on the differences between a body that is presented as a fictional adaptation to the order of the ?desirable? and another that intends to make visible the constitutive vulnerability of the maternal process. The analysis shows how the management and exhibition of the maternal body in the net is marked by the imperatives of healthNúria The fitness lifestyle is carried out in and through a subjective and power framework in which businessMaría Inésembodied hyperbolically in the figure of the fit mom. And on the other handdispositivoUNESCO::SOCIOLOGÍAmaternidadmothers who question the hegemonic ?desirable? maternal figures in the digital scenario through exhibition of the realness of ther bodies and images of and ?self-acceptance?. The paper explores images and narratives in these two positions1137-7038 8537 Arxius de sociologia 558787 2019 41 7605686 El cuerpo materno en la red: entre el orden de lo ?deseable? y el de lo real Landa//purl.org/becyt/ford/5 [https]cuerpos realeshealth:SOCIOLOGÍA [UNESCO]bodymaternityand body spectacularization dispositifs circulate and are assembledas well as those of authenticity and extimacy cuerpofitnessCalafell Salatherapeuticand plasticity of the mother?s body arises as a problem and the object of the intervention of diverse technologies aimed at adjusting its size and shape to the parameters of an active and healthy life. In digital scenariossaludthrough visual and discursive dispositifs that promote the ideal of fitnessneoliberal and informational capitalism. Within this frameworkredes socialesthe temporalityand performance//purl.org/becyt/ford/5.8 [https]mutationon the one hand
researchProduct

Two-step cross correlation-based algorithm for motion estimation applied to fertilizer granules' motion during centrifugal spreading

2011

Imaging systems are progressing in both accuracy and ro- bustness, and their use in precision agriculture is increasing accordingly. One application of imaging systems is to understand and control the cen- trifugal fertilizing spreading process. Predicting the spreading pattern on the ground relies on an estimation of the trajectories and velocities of ejected granules. The algorithms proposed to date have shown low ac- curacy, with an error rate of a few pixels. But a more accurate estimation of the motion of the granules can be achieved. Our new two-step cross- correlation-based algorithm is based on the technique used in particle image velocimetry (PIV), which has yielded highly accurate…

fluid mechanicsImage processing01 natural sciences010305 fluids & plasmas010309 opticsmotion estimationMotion estimationcameras0103 physical sciencesComputer visionImage sensorMathematicsCross-correlationPixelbusiness.industrycentrifugesGeneral EngineeringfertilisersFluid mechanicsSubpixel renderingAtomic and Molecular Physics and Opticsimage processingParticle image velocimetryvelocimetersArtificial intelligencebusinessAlgorithm
researchProduct

El cineasta como "flâneur". trazado y formalización de la imagen documental en Guest (José Luis Guerin, 2010)

2015

flâneur&quotEl cineasta como "flâneur". trazado y formalización de la imagen documental en Guest (José Luis Guerin 2010) ArtículoUNESCO::HISTORIA::Historia por especialidades::Historia del arte. trazado y formalización de la imagen documental en Guest (José Luis Guerin 2010) Artículo:HISTORIA::Historia por especialidades::Historia del arte [UNESCO]El cineasta como &quot
researchProduct

Chronic neural probe for simultaneous recording of single-unit, multi-unit, and local field potential activity from multiple brain sites

2017

Drug resistant focal epilepsy can be treated by resecting the epileptic focus requiring a precise focus localisation using stereoelectroencephalography (SEEG) probes. As commercial SEEG probes offer only a limited spatial resolution, probes of higher channel count and design freedom enabling the incorporation of macro and microelectrodes would help increasing spatial resolution and thus open new perspectives for investigating mechanisms underlying focal epilepsy and its treatment. This work describes a new fabrication process for SEEG probes with materials and dimensions similar to clinical probes enabling recording single neuron activity at high spatial resolution.Polyimide is used as a bi…

focal epilepsyMaterials scienceFocus (geometry)SEEGBiomedical EngineeringPosterior parietal cortexFOS: Physical sciencesBiocompatible Materials02 engineering and technologyLocal field potentialchronic; depth neural probe; focal epilepsy; in vivo; SEEG; single unit activity; stereoelectroencephalography; Biomedical Engineering; Cellular and Molecular NeuroscienceStereoelectroencephalographystereoelectroencephalography03 medical and health sciencesCellular and Molecular Neuroscience0302 clinical medicineParietal LobeAnimalsMulti unitImage resolutionEvoked PotentialsNeuronsBrainElectroencephalography021001 nanoscience & nanotechnologyPhysics - Medical PhysicsMacaca mulattadepth neural probesingle unit activityElectrodes ImplantedchronicMicroelectrodein vivoQuantitative Biology - Neurons and CognitionFOS: Biological sciencesElectrodeFemaleNeurons and Cognition (q-bio.NC)Epilepsies PartialMedical Physics (physics.med-ph)0210 nano-technologyMicroelectrodes030217 neurology & neurosurgeryBiomedical engineering
researchProduct

A deep learning segmentation approach to calories and weight estimation of food images

2019

Master's thesis Information- and communication technology IKT590 - University of Agder 2019 Today’s generation is very aware of what they are eating and the amountof calories in their food. Eating too many calories can lead to increasedweight, which has become a big health issue. A study from 2016 states thatmore than 1,9 billion adults are overweight where almost one third of theseare obese. Statistics from Norway show that 1 of 4 men and 1 of 5 womenare obese.Artificial Intelligence in general and deep learning in particular can be usedto help understand the content of eaten food. In this master thesis, wepropose a network to estimate the weight of food from a single image. Thisis done in…

food classificationIKT590segmentationdeep learninginception networksVDP::Teknologi: 500::Informasjons- og kommunikasjonsteknologi: 550calorie estimationimage classification
researchProduct

Muscular Strength Imbalances Are not Associated with Skin Temperature Asymmetries in Soccer Players

2020

Although strength imbalances using isokinetic dynamometer have been examined for injury risk screening in soccer players, it is very expensive and time-consuming, making the evaluation of new methods appealing. The aim of the study was to analyze the agreement between muscular strength imbalances and skin temperature bilateral asymmetries as well as skin temperature differences in the hamstrings and quadriceps. The skin temperature of the anterior and posterior thigh of 59 healthy male soccer athletes was assessed at baseline using infrared thermography for the identification of hamstrings-quadriceps skin temperature differences and thermal asymmetries (&gt

footballmedicine.medical_specialtyinjurythermal imageConcentricPhysical strengthArticleGeneral Biochemistry Genetics and Molecular BiologyTemperatura corporal03 medical and health sciences0302 clinical medicinePhysical medicine and rehabilitationpreventionMedicineInjury riskEccentriclcsh:ScienceEcology Evolution Behavior and Systematics030222 orthopedicsintegumentary systembiologyAthletesbusiness.industryPaleontologySkin temperature030229 sport sciencesPosterior compartment of thighbiology.organism_classificationthermographySpace and Planetary ScienceIsokinetic dynamometerstrength imbalancelcsh:Qbusinesshuman activitiesLife
researchProduct

Monitoring deforestation and forest degradation linking high-resolution satellite data and field data in the context of REDD+. A case of Tanzania

2017

El principal objetivo de este doctorado es apoyar el desarrollo de un sistema nacional de monitoreo forestal en Tanzania para informar sobre las emisiones actuales e históricas derivadas de la deforestación y la degradación forestal. El marco de la tesis se centra específicamente en el emergente contexto internacional de la iniciativa REDD + (Reducción de Emisiones por Deforestación y Degradación) de las Naciones Unidas, bajo la cual los países pueden obtener subsidios financieros para demostrar que están reduciendo sus emisiones de carbono de tierras forestales con respecto a su práctica histórica reciente. La investigación se centró en cinco áreas de investigación: La parte (1) revisa los…

forest degradationnational monitoring:CIENCIAS DE LA TIERRA Y DEL ESPACIO [UNESCO]field dataUNESCO::FÍSICATanzaniaforestremote sensinghigh spatial resolution:FÍSICA [UNESCO]tree coversatellite imagedeforestationforest biomasscarbon emissionsREDD+UNESCO::CIENCIAS DE LA TIERRA Y DEL ESPACIOsatellite datavegetation classification
researchProduct