Détection de mouvements dans des séquences d’images basée sur la dynamique de supraconductivité

Le but de ce travail est d’étudier la faisabilité de la détection de mouvements dans des séquences d’images en utilisant l’équation de continuité et la dynamique de supraconductivité. Notre approche peut être motivée par le fait que l’équation de continuité apparait dans plusieurs techniques qui estiment le flot optique. Un grand nombre de techniques qui utilisent les flots optiques utilisent une contrainte appelée contrainte de l’invariance lumineuse. La dynamique de supraconductivité nous permet de nous affranchir de la contrainte de l’invariance lumineuse. Les expériences se feront avec la base de données de séquences d’images CDNET 2014. Pour obtenir les résultats numériques en terme de score F1, une combinaison sera faite par la suite entre la dynamique de supraconductivité et un méchanisme d’attention qui est un résumé des vérites de terrain.

USO Y COBERTURA DE LA TIERRA EN COSTA RICA PARA 1992: UNA APLICACIÓN DE TELEDETECCIÓN ESPACIAL Y SISTEMAS DE INFORMACION GEOGRÁFICA

En el presente artículo se describe el procedimiento utilizado para actualizar el mapa de uso y cobertura de la tierra de Costa Rica para el año 1992. En el proceso se integró la cartografía análoga existente para 1985 a escala 1:200.000 e imágenes digitales del Mapeador Temático de LANDSAT para los años 1991 y 1993. Para comprobar la exactitud de la clasificación se usaron 1.372 puntos obtenidos de fotos aéreas de 1992 y trabajo de campo empleando un Sistema de Posicionamiento Global (SPG). Los resultados obtenidos indican que existe un 46.6% del país bajo pastos, un 32,9 bajo bosques y un 8,5% bajo uso agrícola. Un 7,8% del área se incluyó en una categoría denominada <<no clasificada, usos mezclados, deforestada>>. La exactitud global de la clasificación fue de un 74%; con una confusión entre pasto y bosque de un 19%.  SUMMARY The objective of this paper is to describe the process used by the authors to update the preliminary 1985 land use-land cover map of Costa Rica. Paper maps of 1985 at scale 1:200.000 we digitized, rasterized and use to label the output of a nonsupervised classification carried out using 1991-93 digital data from LANDSAT 5 (Thematic Mapper). Aerial photography and field work aided by a Global Positional System (GPS) was used to gather ground-truth data. A total of 1372 stratified points were used to test the accuracy of the final map. Our results showed that 46,6% of the country is under pasture, 32,9% under forest and 8.5% under agriculture. The nonclassified areas were lumped into one category that accounted for 7,8% of the country. Global accuracy of the classification was 74% with the confusion between forest and pasture accounting for 19% of this error.

Field-oriented assessment of agricultural crops through temporal segmentation of modis VI data.

Monitoring agricultural crops constitutes a vital task for the general understanding of land use spatio-temporal dynamics. This paper presents an approach for the enhancement of current crop monitoring capabilities on a regional scale, in order to allow for the analysis of environmental and socio-economic drivers and impacts of agricultural land use. This work discusses the advantages and current limitations of using 250m VI data from the Moderate Resolution Imaging Spectroradiometer (MODIS) for this purpose, with emphasis in the difficulty of correctly analyzing pixels whose temporal responses are disturbed due to certain sources of interference such as mixed or heterogeneous land cover. It is shown that the influence of noisy or disturbed pixels can be minimized, and a much more consistent and useful result can be attained, if individual agricultural fields are identified and each field's pixels are analyzed in a collective manner. As such, a method is proposed that makes use of image segmentation techniques based on MODIS temporal information in order to identify portions of the study area that agree with actual agricultural field borders. The pixels of each portion or segment are then analyzed individually in order to estimate the reliability of the temporal signal observed and the consequent relevance of any estimation of land use from that data. The proposed method was applied in the state of Mato Grosso, in mid-western Brazil, where extensive ground truth data was available. Experiments were carried out using several supervised classification algorithms as well as different subsets of land cover classes, in order to test the methodology in a comprehensive way. Results show that the proposed method is capable of consistently improving classification results not only in terms of overall accuracy but also qualitatively by allowing a better understanding of the land use patterns detected. It thus provides a practical and straightforward procedure for enhancing crop-mapping capabilities using temporal series of moderate resolution remote sensing data.

Perceiving the 3d world from single images

Depth represents a crucial piece of information in many practical applications, such as obstacle avoidance and environment mapping. This information can be provided either by active sensors, such as LiDARs, or by passive devices like cameras. A popular passive device is the binocular rig, which allows triangulating the depth of the scene through two synchronized and aligned cameras. However, many devices that are already available in several infrastructures are monocular passive sensors, such as most of the surveillance cameras. The intrinsic ambiguity of the problem makes monocular depth estimation a challenging task. Nevertheless, the recent progress of deep learning strategies is paving the way towards a new class of algorithms able to handle this complexity. This work addresses many relevant topics related to the monocular depth estimation problem. It presents networks capable of predicting accurate depth values even on embedded devices and without the need of expensive ground-truth labels at training time. Moreover, it introduces strategies to estimate the uncertainty of these models, and it shows that monocular networks can easily generate training labels for different tasks at scale. Finally, it evaluates off-the-shelf monocular depth predictors for the relevant use case of social distance monitoring, and shows how this technology allows to overcome already existing strategies limitations.

Synthetic data generation for the assessment of antimicrobial resistance through machine learning

As a consequence of the diffusion of next generation sequencing techniques, metagenomics databases have become one of the most promising repositories of information about features and behavior of microorganisms. One of the subjects that can be studied from those data are bacteria populations. Next generation sequencing techniques allow to study the bacteria population within an environment by sampling genetic material directly from it, without the needing of culturing a similar population in vitro and observing its behavior. As a drawback, it is quite complex to extract information from those data and usually there is more than one way to do that; AMR is no exception. In this study we will discuss how the quantified AMR, which regards the genotype of the bacteria, can be related to the bacteria phenotype and its actual level of resistance against the specific substance. In order to have a quantitative information about bacteria genotype, we will evaluate the resistome from the read libraries, aligning them against CARD database. With those data, we will test various machine learning algorithms for predicting the bacteria phenotype. The samples that we exploit should resemble those that could be obtained from a natural context, but are actually produced by a read libraries simulation tool. In this way we are able to design the populations with bacteria of known genotype, so that we can relay on a secure ground truth for training and testing our algorithms.

Depth Estimation in Stereo Biomedical Images via Proxy-Supervised Deep Learning

In order to estimate depth through supervised deep learning-based stereo methods, it is necessary to have access to precise ground truth depth data. While the gathering of precise labels is commonly tackled by deploying depth sensors, this is not always a viable solution. For instance, in many applications in the biomedical domain, the choice of sensors capable of sensing depth at small distances with high precision on difficult surfaces (that present non-Lambertian properties) is very limited. It is therefore necessary to find alternative techniques to gather ground truth data without having to rely on external sensors. In this thesis, two different approaches have been tested to produce supervision data for biomedical images. The first aims to obtain input stereo image pairs and disparities through simulation in a virtual environment, while the second relies on a non-learned disparity estimation algorithm in order to produce noisy disparities, which are then filtered by means of hand-crafted confidence measures to create noisy labels for a subset of pixels. Among the two, the second approach, which is referred in literature as proxy-labeling, has shown the best results and has even outperformed the non-learned disparity estimation algorithm used for supervision.

A deep learning model for controlling the galvanizing process in a production line

In the industry of steelmaking, the process of galvanizing is a treatment which is applied to protect the steel from corrosion. The air knife effect (AKE) occurs when nozzles emit a steam of air on the surfaces of a steel strip to remove excess zinc from it. In our work we formalized the problem to control the AKE and we implemented, with the R&D dept.of MarcegagliaSPA, a DL model able to drive the AKE. We call it controller. It takes as input the tuple : a tuple of the physical conditions of the process line (t,h,s) with the target value of the zinc coating (c); and generates the expected tuple of (pres and dist) to drive the mechanical nozzles towards the (c). According to the requirements we designed the structure of the network. We collected and explored the data set of the historical data of the smart factory. Finally, we designed the loss function as sum of three components: the minimization between the coating addressed by the network and the target value we want to reach; and two weighted minimization components for both pressure and distance. In our solution we construct a second module, named coating net, to predict the coating of zinc resulting from the AKE when the conditions are applied to the prod. line. Its structure is made by a linear and a deep nonlinear “residual” component learned by empirical observations. The predictions made by the coating nets are used as ground truth in the loss function of the controller. By tuning the weights of the different components of the loss function, it is possible to train models with slightly different optimization purposes. In the tests we compared the regularization of different strategies with the standard one in condition of optimal estimation for both; the overall accuracy is ± 3 g/m^2 dal target for all of them. Lastly, we analyze how the controller modeled the current solutions with the new logic: the sub-optimal values of pres and dist can be optimize of 50% and 20%.

Gli strumenti di profilazione nella società dell’informazione

La nostra società è sempre più interconnessa al punto che difficilmente riusciamo a distinguere tra la vita online e quella offline: “onlife” è il termine coniato dal filosofo Luciano Floridi e inserito nel 2019 tra le voci della Treccani proprio per indicare questo nuova forma di comportamento. Il virtuale diventa reale grazie alle nuove tecnologie e le modalità di comunicazione, vivendo in un ambiente che è contemporaneamente analogico e digitale. Il soggetto a cui i dati si riferiscono perde il controllo su di essi, mentre questi vengono raccolti e processati a costi nulli e, spesso, senza l’adeguato consenso. È all’interno di questo contesto che nasce la regolamentazione per tutelare la vita privata delle persone ed evitare trattamenti dei dati non autorizzati, i quali possano avere anche un’influenza sulla capacità di scelta e determinazione dell’utente. In questo elaborato di tesi saranno trattati i problemi e rischi della profilazione, quale caso specifico di trattamento automatizzato dei dati personali, delineando un quadro normativo di rifermento. Verrano analizzate anche le informative privacy di alcune società dell’informazione per prendere coscienza di quali (e quanti) dati vengono condivisi durante l’utilizzo dei loro servizi gratuiti. Infine, verrà fornita una descrizione di quali possano essere gli strumenti software, le azioni da compiere per lasciare meno impronte digitali possibili.

The church, the pillar and ground of the truth,

Mode of access: Internet.

Site Specific Spin Dynamics In Bafe2as2: Tuning The Ground State By Orbital Differentiation.

The role of orbital differentiation on the emergence of superconductivity in the Fe-based superconductors remains an open question to the scientific community. In this investigation, we employ a suitable microscopic spin probe technique, namely Electron Spin Resonance (ESR), to investigate this issue on selected chemically substituted BaFe2As2 single crystals. As the spin-density wave (SDW) phase is suppressed, we observe a clear increase of the Fe 3d bands anisotropy along with their localization at the FeAs plane. Such an increase of the planar orbital content is interestingly independent of the chemical substitution responsible for suppressing the SDW phase. As a consequence, the magnetic fluctuations in combination with this particular symmetry of the Fe 3d bands are propitious ingredients for the emergence of superconductivity in this class of materials.

Qualidade de luz na biometria e anatomia foliar de plântulas de Cattleya loddigesii L. (Orchidaceae) micropropagadas

A qualidade de luz pode alterar a morfogênese das plantas por meio de uma série de processos mediados por receptores de luz, principalmente na região do vermelho e azul. O objetivo do presente estudo foi verificar alterações anatômicas foliares e características biométricas de Cattleya loddigesii 'Tipo', cultivadas in vitro, sob diferentes malhas coloridas com nível de radiação de 50% de sombreamento. Plântulas oriundas de autopolinização e sementes germinadas in vitro, com aproximadamente 1,0cm de comprimento e com raízes, foram inoculadas em meio WPM e submetidas a diferentes condições de incubação. Testou-se o efeito de sombrites coloridos (vermelho e azul) sobre os frascos cultivados em casa de vegetação (CV) e sala de crescimento (SC), além dos tratamentos, nos dois ambientes, sem utilização das telas coloridas. A avaliação foi efetuada 180 dias após inoculação. Com os resultados obtidos, observou-se que o ambiente de cultivo promove alterações anatômicas e biométricas em plântulas de Cattleya loddigesii 'Tipo' micropropagadas. As alterações promovidas pelo cultivo em luz natural evidenciam maior capacidade fotossintética, por meio de maior diferenciação dos tecidos clorofilianos, promovendo uma superfície foliar anatomicamente adaptada à fase de aclimatização.

A comparison of lower limb EMG and ground reaction forces between barefoot and shod gait in participants with diabetic neuropathic and healthy controls

Background: It is known that when barefoot, gait biomechanics of diabetic neuropathic patients differ from nondiabetic individuals. However, it is still unknown whether these biomechanical changes are also present during shod gait which is clinically advised for these patients. This study investigated the effect of the participants own shoes on gait biomechanics in diabetic neuropathic individuals compared to barefoot gait patterns and healthy controls. Methods: Ground reaction forces and lower limb EMG activities were analyzed in 21 non-diabetic adults (50.9 +/- 7.3 yr, 24.3 +/- 2.6 kg/m(2)) and 24 diabetic neuropathic participants (55.2 +/- 7.9 yr, 27.0 +/- 4.4 kg/m(2)). EMG patterns of vastus lateralis, lateral gastrocnemius and tibialis anterior, along with the vertical and antero-posterior ground reaction forces were studied during shod and barefoot gait. Results: Regardless of the disease, walking with shoes promoted an increase in the first peak vertical force and the peak horizontal propulsive force. Diabetic individuals had a delay in the lateral gastrocnemius EMG activity with no delay in the vastus lateralis. They also demonstrated a higher peak horizontal braking force walking with shoes compared to barefoot. Diabetic participants also had a smaller second peak vertical force in shod gait and a delay in the vastus lateralis EMG activity in barefoot gait compared to controls. Conclusions: The change in plantar sensory information that occurs when wearing shoes revealed a different motor strategy in diabetic individuals. Walking with shoes did not attenuate vertical forces in either group. Though changes in motor strategy were apparent, the biomechanical did not support the argument that the use of shoes contributes to altered motor responses during gait.

The B0.5 IVe CoRoT target HD 49330 II. Spectroscopic ground-based observations

Context. We present spectroscopic ground-based observations of the early Be star HD 49330 obtained simultaneously with the CoRoT-LRA1 run just before the burst observed in the CoRoT data. Aims. Ground-based spectroscopic observations of the early Be star HD 49330 obtained during the precursor phase and just before the start of an outburst allow us to disantangle stellar and circumstellar contributions and identify modes of stellar pulsations in this rapidly rotating star. Methods. Time series analysis (TSA) is performed on photospheric line profiles of He I and Si III by means of the least squares method. Results. We find two main frequencies f1 = 11.86 c d(-1) and f2 = 16.89 c d(-1) which can be associated with high order p-mode pulsations. We also detect a frequency f3 = 1.51 c d(-1) which can be associated with a low order g-mode. Moreover we show that the stellar line profile variability changed over the spectroscopic run. These results are in agreement with the results of the CoRoT data analysis, as shown in Huat et al. (2009). Conclusions. Our study of mid-and short-term spectroscopic variability allows the identification of p-and g-modes in HD 49330. It also allows us to display changes in the line profile variability before the start of an outburst. This brings new constraints for the seimic modelling of this star.

The pulsations of the B5IVe star HD 181231 observed with CoRoT and ground-based spectroscopy

Context. HD 181231 is a B5IVe star, which has been observed with the CoRoT satellite during similar to 5 consecutive months and simultaneously from the ground in spectroscopy and spectropolarimetry. Aims. By analysing these data, we aim to detect and characterize as many pulsation frequencies as possible, to search for the presence of beating effects possibly at the origin of the Be phenomenon. Our results will also provide a basis for seismic modelling. Methods. The fundamental parameters of the star are determined from spectral fitting and from the study of the circumstellar emission. The CoRoT photometric data and ground-based spectroscopy are analysed using several Fourier techniques: CLEAN-NG, PASPER, and TISAFT, as well as a time-frequency technique. A search for a magnetic field is performed by applying the LSD technique to the spectropolarimetric data. Results. We find that HD 181231 is a B5IVe star seen with an inclination of similar to 45 degrees. No magnetic field is detected in its photosphere. We detect at least 10 independent significant frequencies of variations among the 54 detected frequencies, interpreted in terms of non-radial pulsation modes and rotation. Two longer-term variations are also detected: one at similar to 14 days resulting from a beating effect between the two main frequencies of short-term variations, the other at similar to 116 days due either to a beating of frequencies or to a zonal pulsation mode. Conclusions. Our analysis of the CoRoT light curve and ground-based spectroscopic data of HD 181231 has led to the determination of the fundamental and pulsational parameters of the star, including beating effects. This will allow a precise seismic modelling of this star.

A possible role of ground-based microorganisms on cloud formation in the atmosphere

The formation of clouds is an important process for the atmosphere, the hydrological cycle, and climate, but some aspects of it are not completely understood. In this work, we show that microorganisms might affect cloud formation without leaving the Earth's surface by releasing biological surfactants (or biosurfactants) in the environment, that make their way into atmospheric aerosols and could significantly enhance their activation into cloud droplets. In the first part of this work, the cloud-nucleating efficiency of standard biosurfactants was characterized and found to be better than that of any aerosol material studied so far, including inorganic salts. These results identify molecular structures that give organic compounds exceptional cloud-nucleating properties. In the second part, atmospheric aerosols were sampled at different locations: a temperate coastal site, a marine site, a temperate forest, and a tropical forest. Their surface tension was measured and found to be below 30 mN/m, the lowest reported for aerosols, to our knowledge. This very low surface tension was attributed to the presence of biosurfactants, the only natural substances able to reach to such low values. The presence of strong microbial surfactants in aerosols would be consistent with the organic fractions of exceptional cloud-nucleating efficiency recently found in aerosols, and with the correlations between algae bloom and cloud cover reported in the Southern Ocean. The results of this work also suggest that biosurfactants might be common in aerosols and thus of global relevance. If this is confirmed, a new role for microorganisms on the atmosphere and climate could be identified.