Éditeur(s) : HAL CCSD Maney Publishing
Résumé : International audience
The strict environmental legislations and increasing ecological awareness among scientists have led to the development of “green” alternatives to mitigate corrosion. In the present work, literature on green corrosion inhibitors has been reviewed, and the salient features of our work on green corrosion inhibitors have been highlighted. Among the studied leaves, extract Andrographis paniculata showed better inhibition performance (98%) than the other leaves extract. Strychnos nuxvomica showed better inhibition (98%) than the other seed extracts. Moringa oleifera is reflected as a good corrosion inhibitor of mild steel in 1 M HCl with 98% inhibition efficiency among the studied fruits extract. Bacopa monnieri showed its maximum inhibition performance to be 95% at 600 ppm among the investigated stem extracts. All the reported plant extracts were found to inhibit the corrosion of mild steel in acid media.
ISSN: 1478-422X

hal-01097106
https://hal-agroparistech.archives-ouvertes.fr/hal-01097106
DOI : 10.1016/j.corsci.2010.10.006

Éditeur(s) : HAL CCSD
Résumé : International audience
This paper deals with the study of the kinematic dimension of scribbling activities executed by kindergarten children aged from 3 to 6 years old from three grades. For this purpose, three sigma-lognormal features, six classical ones and one hybrid feature related to visuo-motor skills are extracted from scribbles realized using five different Type Grib. The statistical analysis of these data illustrates that sigma-lognormal modeling can satisfactorily reconstruct kindergarten children’s scribbles. Moreover, this preliminary study confirms that there are significant differences between grades with respect to six of the features we studied, regardless of the nature of the scribbling movement made by children. These features are related to rapidity, fluidity and precision of linear and curvilinear movements used for scribbling tasks.
17th Biennial Conference of the International Graphonomics Society
Pointe-à-Pitre, Guadeloupe

hal-01165910
https://hal.univ-antilles.fr/hal-01165910
https://hal.univ-antilles.fr/hal-01165910/document
https://hal.univ-antilles.fr/hal-01165910/file/IGS_2015_submission_28.pdf

Éditeur(s) : Wiley
Résumé : This paper investigates some economic consequences of creating a marine reserve on both fishing and ecotourism, when the range of controllability of fishing effort is limited and the impact of the reserve on ecosystem is considered. The issue is illustrated by the example of creating a no-take zone in part of a region where fishing is managed through a limited entry license system, and which is inhabited by two interacting stocks: a stock of prey (fish) and a stock of predators (seals). While the former is targeted by commercial fishing, the latter is not subject to harvest but is a potential basis for a commercial non-extractive activity (seal watching). Analysis is conducted with the help of a bioeconomic model combining the features of marine reserve modeling and of multispecies modeling. Following a description of the model, results of several simulation runs are presented. These show that creating a marine reserve has more complex economic implications than predicted in studies focused exclusively on one stock and/or commercial fisheries. More specifically, the model shows that the dynamics of the two interacting stocks reduces the benefits of the no-take zone for the fishing industry, while it makes the creation of this zone provide an opportunity for the development of ecotourism. Due to this dynamics, the model suggests that the optimal size of the reserve is larger when ecotourism is taken into account along with fishing activities.
Natural Resource Modeling (1939-7445) (Wiley), 2002-12 , Vol. 15 , N. 4 , P. 387–411

Droits : 2002 Rocky Mountain Mathematics Consortium
http://archimer.ifremer.fr/doc/00244/35506/34010.pdf
DOI:10.1111/j.1939-7445.2002.tb00095.x
http://archimer.ifremer.fr/doc/00244/35506/

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : 17 pages, 11 figures.
International audience
The problem addressed in this paper is the automatic extraction of names from a document image. Our approach relies on the combination of two complementary analyses. First, the image-based analysis exploits visual clues to select the regions of interest in the document. Second, the textual-based analysis searches for name patterns and low-level word textual features. Both analyses are then combined at the word level through a neural network fusion scheme. Reported results on degraded documents such as facsimile and photocopied technical journals demonstrate the interest of the combined approach.
ISSN: 1433-7541

hal-00329459
https://hal.archives-ouvertes.fr/hal-00329459
DOI : 10.1007/s10044-006-0038-6

Éditeur(s) : HAL CCSD IGI Global
Résumé : International audience
Standard data mining techniques have been applied and adapted for eliciting knowledge from social networks, by achieving classical tasks such as classification, search for frequent patterns or link prediction. Most works have exploited only the network topological structure, and therefore cannot be used to answer questions involving nodes features. For instance, the frequent pattern discovery task generally refers to the search for sub-networks frequently found in a single network or in a set of networks. In the same area, this paper focuses on the concept of frequent link that stands as a regularity found in a network on links between node groups that share common characteristics. The extraction of such links from a social network is a particularly challenging and computationally intensive problem, since it is much dependent on the number of links and attributes. In this study, we propose a solution for reducing the search space of frequent links, by filtering the nodes features on a criterion of frequency. We make the assumption that frequent links occur between sets of features that are themselves frequent. This property is used to reduce the search space and speed up the extraction process. We empirically show that it is well founded. We discuss the efficiency of the solution in terms of computation time and number of frequent patterns found depending on several frequency thresholds.
ISSN: 1947-8186

hal-00786166
https://hal.archives-ouvertes.fr/hal-00786166

Éditeur(s) : HAL CCSD MDPI
Résumé : International audience
Malaria remains one of the most common vector-borne diseases in the world and the definition of novel control strategies can benefit from the modeling of transmission processes. However, data-driven models are often difficult to build, as data are very often incomplete, heterogeneous in nature and in quality, and/or biased. In this context, a knowledge-based approach is proposed to build a robust and general landscape-based hazard index for malaria transmission that is tailored to the Amazonian region. A partial knowledge-based model of the risk of malaria transmission in the Amazonian region, based on landscape features and extracted from a systematic literature review, was used. Spatialization of the model was obtained by generating land use and land cover maps of the cross-border area between French Guiana and Brazil, followed by computing and combining landscape metrics to build a set of normalized landscape-based hazard indices. An empirical selection of the best index was performed by comparing the indices in terms of adequacy with the knowledge-based model, intelligibility and correlation with P. falciparum incidence rates. The selected index is easy to interpret and successfully represents the current knowledge about the role played by landscape patterns in malaria transmission within the study area. It was significantly associated with P. falciparum incidence rates, using the Pearson and Spearman correlation coefficients (up to 0.79 and 0.75, respectively; p-value < 0.001), and the linear regression coefficient of determination (reaching 0.63; p-values < 0.001). This study establishes a spatial knowledge-driven, landscape-based hazard malaria index using remote sensing that can be easily produced on a regular basis and might be useful for malaria prediction, surveillance, and control. Keywords: remote sensing; land use and land cover; landscape metric; knowledge-based hazard modeling; malaria; cross-border area between French Guiana and Brazil
ISSN: 2072-4292

hal-01355255
https://hal.archives-ouvertes.fr/hal-01355255
https://hal.archives-ouvertes.fr/hal-01355255/document
https://hal.archives-ouvertes.fr/hal-01355255/file/010066865.pdf
IRD : fdi:010066865

Éditeur(s) : HAL CCSD
Résumé : 6 pages, 4 figures. In ICML 2013 Workshop on Machine Learning for Bioacoustics
International audience
The work presented in this paper is part of a global framework which long term goal is to design a wireless sensor network able to support the observation of a population of endangered birds. We present the first stage for which we have conducted a knowledge discovery approach on a sample of acoustical data. We use MFCC features extracted from bird songs and we exploit two knowledge discovery techniques. One that relies on clustering-based approaches, that highlights the homogeneity in the songs of the species. The other, based on predictive modeling, that demonstrates the good performances of various machine learning techniques for the identification process. The knowledge elicited provides promising results to consider a widespread study and to elicit guidelines for designing a first version of the automatic approach for data collection based on acoustic sensors.
ICML 2013 Workshop on Machine Learning for Bioacoustics
Atlanta, United States

hal-00840741
https://hal.archives-ouvertes.fr/hal-00840741
ARXIV : 1306.5349

Éditeur(s) : HAL CCSD Elsevier
Résumé : Primary arc melts may form through fluxed or adiabatic decompression melting in the mantle wedge, or via a combination of both processes. Major limitations to our understanding of the formation of primary arc melts stem from the fact that most arc lavas are aggregated blends of individual magma batches, further modified by differentiation processes in the sub-arc mantle lithosphere and overlying crust. Primary melt generation is thus masked by these types of second-stage processes. Magma-hosted peridotites sampled as xenoliths in subduction zone magmas are possible remnants of sub-arc mantle and magma generation processes, but are rarely sampled in active arcs. Published studies have emphasised the predominantly harzburgitic lithologies with particularly high modal orthopyroxene in these xenoliths; the former characteristic reflects the refractory nature of these materials consequent to extensive melt depletion of a lherzolitic protolith whereas the latter feature requires additional explanation.Here we present major and minor element data for pristine, mantle-derived, lava-hosted spinel-bearing harzburgite and dunite xenoliths and associated primitive melts from the active Kamchatka and Bismarck arcs. We show that these peridotite suites, and other mantle xenoliths sampled in circum-Pacific arcs, are a distinctive peridotite type not found in other tectonic settings, and are melting residues from hydrous melting of silica-enriched mantle sources. We explore the ability of experimental studies allied with mantle melting parameterisations (pMELTS, Petrolog3) to reproduce the compositions of these arc peridotites, and present a protolith (‘hybrid mantle wedge’) composition that satisfies the available constraints. The composition of peridotite xenoliths recovered from erupted arc magmas plausibly requires their formation initially via interaction of slab-derived components with refractory mantle prior to or during the formation of primary arc melts. The liquid compositions extracted from these hybrid sources are higher in normative quartz and hypersthene (i.e., they have a more silica-saturated character) in comparison with basalts derived from prior melt-depleted asthenospheric mantle beneath ridges. These primary arc melts range from silica-rich picrite to boninite and high-Mg basaltic andesite along a residual spinel harzburgite cotectic. Silica enrichment in the mantle sources of arc-related, subalkaline picrite-boninite-andesite suites coupled with the amount of water and depth of melting, are important for the formation of medium-Fe (‘calc-alkaline’) andesite-dacite-rhyolite suites, key lithologies forming the continental crust.
ISSN: 0016-7037

hal-01553520
https://hal.archives-ouvertes.fr/hal-01553520
DOI : 10.1016/j.gca.2016.09.030

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
We report new data on the geochemical evolution, metasomatic and deformation processes in the lower layers of cratonic lithospheric mantle based on a detailed study of extraordinarily fresh and large deformed peridotite xenoliths from the Udachnaya kimberlite pipe, Siberia. Based on our T-P estimates, the deformed peridotites are localized in a depth range of 170 to 220 km near the base of cratonic mantle. The degree of deformation is not correlated with the depth and equilibration temperatures of the xenoliths. The deformed peridotites are depleted in magmaphile major elements indicating their residual nature, but are enriched in incompatible trace elements, Fe and K. The deformed peridotites experienced a complex evolution, i.e., they were formed initially as high-degree melt extraction residues and later were subjected to three main stages of metasomatic modification. (1) An old, mostly cryptic metasomatism by melt/fluid of carbonatitic composition, which formed garnets with sinusoidal REE patterns (Smn/Ern > 1). (2) Silicate metasomatism, which led to the most significant changes in mineralogical and chemical composition of the deformed peridotites. (3) Fe and Ti metasomatism just before the entrainment of the deformed rocks into kimberlite magma. Enrichment of the peridotites in large-ion lithophile elements (K, Rb, Ba) is related to the formation of kelyphitic rims around garnet. The nature of the metasomatic agent of silicate metasomatism was evaluated from mass-balance of measured and calculated whole-rock compositions, ratios of highly incompatible elements (D < 0.1) and fractional crystallization modeling. All this evidence indicates that the agent of silicate metasomatism had a composition intermediate between that of kimberlites and HIMU OIB. The metasomatic processes responsible for the formation of deformed peridotites and precipitation of megacryst suites cannot be widespread at the base of the cratonic mantle. Rather, they are local features only existing in the cratonic mantle below kimberlite fields and localized along the metasomatic vein system.
ISSN: 0024-4937

hal-00941010
https://hal.archives-ouvertes.fr/hal-00941010
DOI : 10.1016/j.lithos.2012.11.014

Éditeur(s) : HAL CCSD
Résumé : International audience
In this paper, we present a novel method for recognition of handwritten sketches. Unlike previous approaches, we focus on online retrieval and ability to build our model incrementally, thus we do not need to know all the data in advance and we can achieve very good recognition results after as few as 15 samples. The method is composed of two main parts: feature representation and learning and recognition. In feature representation part, we utilize SIFT-like feature descriptors in combination with soft response Bag-of-Words techniques. Descriptors are extracted locally using our novel sketch-specific sampling strategy and for support regions we follow patch-based approach. For learning and recognition, we use a novel technique based on fuzzy-neural networks, which has shown good performance in incremental learning. The experiments on state-of-the-art benchmarks have shown promising results.
17th Biennial Conference of the International Graphonomics Society
Pointe-à-Pitre, Guadeloupe

hal-01166497
https://hal.univ-antilles.fr/hal-01166497
https://hal.univ-antilles.fr/hal-01166497/document
https://hal.univ-antilles.fr/hal-01166497/file/IGS_2015_submission_37.pdf

Éditeur(s) : HAL CCSD
Résumé : National audience
Designing systems able to interact with students in a natural manner is a complex and far from solved problem. A key aspect of natural interaction is the ability to understand and appropriately respond to human emotions. This paper details our response to the continuous Audio/Visual Emotion Challenge (AVEC'12) whose goal is to predict four affective signals describing human emotions. The proposed method uses Fourier spectra to extract multi-scale dynamic descriptions of signals characterizing face appearance, head movements and voice. We perform a kernel regression with very few representative samples selected via a supervised weighted-distance-based clustering, that leads to a high generalization power. We also propose a particularly fast regressor-level fusion framework to merge systems based on different modalities. Experiments have proven the efficiency of each key point of the proposed method and our results on challenge data were the highest among 10 international research teams.
IHM'14, 26e conférence francophone sur l'Interaction Homme-Machine
Lille, France

hal-01089628
https://hal.archives-ouvertes.fr/hal-01089628
https://hal.archives-ouvertes.fr/hal-01089628/document
https://hal.archives-ouvertes.fr/hal-01089628/file/p44-nicole.pdf

Éditeur(s) : HAL CCSD
Résumé : International audience
We instrumented a car simulator by gathering low level data and fed it to an artificial neural network in order to perform blood alcohol content (BAC) estimations. The results depend on the quality of the data extraction and processing, and also on the selected inputs. We explain our data extraction and processing methodology, and how we used it to generate reliable and comparable features. At last, we describe the performances of individual features and how they combine. In the end, the prototype was able to accurately estimate the BAC value of a subject after being trained with driving samples of this subject with various BAC values.
Proceedings of ACHI 2013, The Sixth International Conference on Advances in Computer-Human Interactions
Nice, France

hal-00850137
https://hal.archives-ouvertes.fr/hal-00850137

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
A laser ablation ICP-MS U-Pb age of 17.84 +/- 0.12 Ma (late Burdigalian) was obtained from monazites separated from a leucocratic diatexite collected in close proximity to a small peridotite massif incorporated into the lower crustal sequence of the Edough Massif (north-eastern Algeria), a southern segment of the peri-Mediterranean Alpine Belt. Monazites extracted from a neighbouring deformed leucogranite intruding early Paleozoic phyllites yield a consistent age of 17.4 +/- 1.3 Ma. Zircons occurring in the same leucogranite, with magmatic characteristics, have an age of 308 +/- 7 Ma interpreted as dating magmatic crystallisation of the leucogranite and reflecting partial melting during the Hercynian orogeny. Low Th/U domains (Th/U<0.10) from the same grains substantiate recrystallisation during a younger metamorphic event whose upper age limit is 286 +/- 11 Ma. These results emphasize the polycyclic evolution of basement rocks preserved in the crystalline units of the western Mediterranean and indicate that part of their metamorphic features were inherited from older, Hercynian, events. Taken together with published Ar-Ar dates, the late Burdigalian age of monazites indicates a rapid cooling rate of c. 370 degrees C/Ma and is regarded as closely approximating the emplacement of the peridotites into the Hercynian basement. The monazite ages are significantly younger than those recorded for orogenic peridotites from the Betic-Rif oroctine and for the timing of lithospheric extension forming the Alboran sea. It is also younger than rifting and back-arc extension opening the Liguro-Provencal basin. The late Burdigalian age is interpreted as dating the incipient rifting event that opened the Algerian basin, which is consequently not a continuation of the Liguro-Provencal basin. At the scale of the western Mediterranean, these observations concur with current models supporting slab roll-back and an eastwards migration of extension in the western Mediterranean, but suggest that the Algerian basin opened as a result of torsion and stretching of the Thethyan slab due to its steepening under the Alboran microplate.
ISSN: 0009-2541

hal-00424511
https://hal.archives-ouvertes.fr/hal-00424511
DOI : 10.1016/j.chemgeo.2008.11.016

Éditeur(s) : HAL CCSD IEEE
Résumé : International audience
This paper proposes a novel automated approach for the categorization of cells in fluorescence microscopy images. Our supervised classification method aims at recognizing patterns of unlabeled cells based on an annotated dataset. First, the cell images need to be indexed by encoding them in a feature space. For this purpose, we propose tailored bio-inspired features relying on the distribution of contrast information. Then, a supervised learning algorithm is proposed for classifying the cells. We carried out experiments on cellular images related to the diagnosis of autoimmune diseases, testing our classification method on the HEp-2 Cells dataset of Foggia et al (CBMS 2010). Results show classification precision larger than 96% on average, thus confirming promising application of our approach to the challenging application of cellular image classification for computer-aided diagnosis.
ICPR 2012
Tsukuba, Japan

hal-00958856
https://hal.archives-ouvertes.fr/hal-00958856
https://hal.archives-ouvertes.fr/hal-00958856/document
https://hal.archives-ouvertes.fr/hal-00958856/file/icpr_2012.pdf

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : International audience
Grain deviations and high extractives content are common features of many tropical woods. We aimed at clarifying their respective impact on vibrational properties, referring to African Padauk (Pterocarpus soyauxii Taub.), a species selected for its interlocked grain, high extractives content, and uses in xylophones. Specimens were cut parallel to the trunk axis (L), and local variations in grain angle (GA), microfibril angle (MFA), specific Young's modulus (E'L/ρ, where r stands for specific gravity) and damping coefficient (tanδL), were measured. GA dependence was analyzed by a mechanical model which allowed to identify the specific Young's modulus (E'3/ρ) and shear modulus (G'/ρ) along the grain (3), as well as their corresponding damping coefficients (tanδ3, tanδG). This analysis was done for native and then for extracted wood. Interlocked grain resulted in 0-25° GA and in variations of a factor 2 in E'L/ρ and tanδL. Along the grain, Padauk wood was characterized, as compared to typical hardwoods, by a somewhat lower E'3/ρ and elastic anisotropy (E'/G'), due to a wide microfibril angle plus a small weight effect of extracts; and a very low tanδ3 and moderate damping anisotropy (tanδG/tanδ3). Extraction affected mechanical parameters in the order: tanδ3 ≈ tanδG > G'/ρ >> E'3/ρ. That is, extractives' effects were nearly isotropic on damping but clearly anisotropic on storage moduli.
ISSN: 0043-7719

hal-00544382
https://hal.archives-ouvertes.fr/hal-00544382
https://hal.archives-ouvertes.fr/hal-00544382/document
https://hal.archives-ouvertes.fr/hal-00544382/file/Identification_anisotropic_vibrational_properties_Bremaud_al.pdf
DOI : 10.1007/s00226-010-0348-0

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
The Maghrebides, Betics and some parts of the Calabrian, NE Sicilian and Tuscan allochtons constitute dismembered fragments of the Alpine belt that resulted from the Cenozoic collision between Africa and Eurasia and the opening of the Western Mediterranean basin. Mineral and whole-rock geochemical analyses have been performed on three distinct outcrops of mantle material from the Edough Massif of NE Algeria, namely the Bou Maiza (BM) gabbros, the La Voile Noire (LVN) amphibolites and the Sidi Mohamed (SM) peridotites. In all samples, Sr isotopes are largely affected by seawater alteration (87Sr/86Sract. > 0.70384 and up to 0.70888) and cannot be used to evaluate the nature of the source reservoirs. SM peridotites display variable depleted mantle Nd isotopic signatures (εNdact. from + 7.0 to + 12.2) and geochemical features suggesting no significant chemical depletion as a result of partial melting and melt extraction (Mg# < 90; slightly LREE-depleted patterns with La/YbN = 0.33-0.39). These rocks are interpreted as parts of the subcontinental lithospheric mantle incorporated into the crustal units of the Edough Massif during the early stages of opening of the Algerian basin. BM gabbros and LVN amphibolites show geochemical signatures indicating derivation from a common Depleted Mantle reservoir (εNd > + 7.9) and are likely cogenetic, but without filiation with the SM peridotites. Pb isotopes indicate a contribution of sediments in the source reservoir, which is attributed to contamination solely by hydrous fluids released from a sedimentary component. This observation, together with a LILE-enrichment, suggests a back-arc basin environment. These results indicate that BM and LVN units were obducted onto the North African margin and subsequently fragmented, probably during doming and exhumation of the lower continental crust of the Edough massif. Doming resulted in opposite movements of the overlying oceanic units, southward for the BM units and northward for LVN amphibolites.
ISSN: 0040-1951

hal-01122777
https://hal.archives-ouvertes.fr/hal-01122777
DOI : 10.1016/j.tecto.2014.03.031

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
These last years, many analysis methods have been proposed to extract knowledge from social networks. As for the traditional data mining domain, these network-based approaches can be classified according to two main families. The approaches based on predictive modelling, which encompass the techniques that analyse current and historical facts to make predictive assumptions about future or unknown events. The approaches based on descriptive modelling, which cover the set of techniques that aim to summarize the data by identifying some relevant features in order to describe how things organize and actually work. In this paper, we review the main descriptive modelling methods of social networks and show for each of them the resulting useful knowledge on a running example. We particularly emphasize on the most recent methods that combine information available on both the network structure and the node attributes in order to provide original description models taking into account the context.
EISSN: 1877-0509

hal-01312383
https://hal.univ-antilles.fr/hal-01312383
DOI : 10.1016/j.procs.2015.05.505

Résumé : Cette thèse traite de la problématique de l'analyse automatique de tracés manuscrits réalisés à mains levée, à caractère géométrique et acquis en ligne. Lorsque l'on analyse un tracé en ligne, on peut le faire selon plusieurs angles. Comme pour les tracés hors ligne, on peu s'atteler à reconnaitre la forme produite. Mais le caractère en ligne de l'acquisition permet également d'autres niveaux d'analyse, comme l'analyse de la procédure utilisée par le dessinateur lors de la production de son dessin. Nous avons dans le cadre de ce travail élaboré des outils permettant de telles analyses de la production de tracé. Dans une première partie nous traitons des pré traitements à effectuer sur le tracé afin de pouvoir procéder aux analyses spatio-graphique et procédurale. Ces pré traitements sont le filtrage, la segmentation mixte à base d'arcs de cercles et de segments de droite, la sélection des points d'intérêt ainsi que l'identification de la nature de ces points. La seconde partie concerne l'analyse visuo-spatiale. Nous abordons alors deux aspects: la question de la reconnaissance des formes et celle de l'évaluation automatique de leur qualité. Notre contribution consiste en une méthode de reconnaissance basée sur l'apparence utilisant des descripteurs locaux. La dernière partie traite de l'analyse procédurale. Nous proposons une méthode d'extraction et de modélisation de la procédure mise en œuvre lors du dessin d'une forme. Nous montrons comment en utilisant cette modélisation, nous pouvons non seulement déterminer la procédure favorite d'un utilisateur mais également reconnaitre un dessinateur par l'analyse de sa procédure.
This thesis handles the problem of the automatic analysis of online hand drawn geometric sketches. An online sketch can be analysed according to several points of view. As for offline sketching, we can try to recognize the produced shape. However, online sketching allows other levels of analysis. For example the analysis of the behavior of the drawer during the production of the sketch. In this thesis, we have tried to develop tools allowing a multi level analysis, including both shape and behavior analysis. The first part of our work deals with the pre treatments that must be performed on the sketch in order to allow upper level analysis. Those pre treatments are filtering, mixed segmentation and feature points detection and labelisation. In the second part, we approach shape analysis in two aspects: shape recognition and evaluation. We have developed a appearance based method which use local descriptors to allows both recognition and evaluation of the quality of a produced shape compared to the model . in the last part we propose a method for drawer's behavior extraction and modeling. Then we show how we can not only determine the favorite procedure of a drawer but also recognize the drawer by analyzing his behavior.

http://www.theses.fr/2010AGUY0394/document

Éditeur(s) : HAL CCSD
Résumé : International audience
Indic scripts are among few scripts in the world that have had continuous development for more than two millennia. The modern forms of the scripts are the result of infinitesimal changes in handwriting being accumulated over centuries. They present us with a unique opportunity to understand various changes occurring in handwriting behavior. We have taken four major Indic scripts in six different stages of evolution and extracted features quantifying their handwriting behavior. We have derived these features by applying the principles of handwriting production and gesture analysis on a paleographic data set. We present various trends and behaviors that occurred during script development and discuss our interpretation of the results in terms of evolution of handwriting behavior. We then briefly discuss the detailed analyses that will be performed on the dataset in the future. We also consider the applications of these results in digital paleography and handwriting-driven systems.
17th Biennial Conference of the International Graphonomics Society
Pointe-à-Pitre, Guadeloupe

hal-01165770
https://hal.univ-antilles.fr/hal-01165770
https://hal.univ-antilles.fr/hal-01165770/document
https://hal.univ-antilles.fr/hal-01165770/file/IGS_2015_submission_8.pdf

Éditeur(s) : HAL CCSD IEEE
Résumé : Object classification is a challenging task in computer vision. Many approaches have been proposed to extract meaningful descriptors from images and classifying them in a supervised learning framework. In this paper, we revisit the classic k-nearest neighbors (k-NN) classification rule, which has shown to be very effective when dealing with local image descriptors. However, k-NN still features some major drawbacks, mainly due to the uniform voting among the nearest prototypes in the feature space. In this paper, we propose a generalization of the classic k-NN rule in a supervised learning (boosting) framework. Namely, we redefine the voting rule as a strong classifier that linearly combines predictions from the k closest prototypes. To induce this classifier, we propose a novel learning algorithm, MLNN (Multiclass Leveraged Nearest Neighbors), which gives a simple procedure for performing prototype selection very efficiently. We tested our method on 12 categories of objects, and observed significant improvement over classic k-NN in terms of classification performances.
International Workshop on Image Analysis for Multimedia Interactive Services
Desenzano del Garda, Italy

hal-00481725
https://hal.archives-ouvertes.fr/hal-00481725
https://hal.archives-ouvertes.fr/hal-00481725/document
https://hal.archives-ouvertes.fr/hal-00481725/file/pbnn_wiamis_10.pdf