Abstract: An optical fibre, for use in the field of distributed measurement of temperature or deformation by optical reflectometry in the frequency domain using the Rayleigh backscattered signal in the fibre, includes a core doped with nanoparticles for example formed from gold particles covered with zirconium oxide, and can be subjected to high temperatures during the measurement. A method for producing the optical fibre includes a step of heat treatment during which the optical fibre is subjected, for a duration of at least one hour, to a heat treatment temperature higher than a maximum temperature to which it will be subjected during a measurement.

Abstract: Systems, methods, and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: Systems, methods, and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: A radiology device includes a support capable of translational movement about an axis of translation relative to a frame of the device, the support being intended to support an object that is to be imaged, an ionizing-ray generator and a detector configured to detect the rays emitted by the generator, the generator and the detector facing one another. The generator comprises several sources each considered to emit from a focal point, the focal points of the various sources being distributed along a sources axis, the detector extending along a detector axis, the sources axis and the detector axis extending in the form of mutually intertwined helices about the axis of translation.

Abstract: The present invention relates to a method of increasing the foaming properties of detergent compositions comprising at least one surfactant by incorporation of saccharide isomerate.

Abstract: A method for calculating during use the geometric parameters of an x-ray imaging system, an object or a patient to be observed being placed between the x-ray source and a detector of x-rays having passed through the object or patient, wherein it includes at least the following steps: detecting at least one marker on the object or the patient or in proximity to the object, the marker being of unknown 3D position, acquiring a plurality of 2D images for a plurality of viewpoints of the imaging system, detecting the position of at least one marker in each of the acquired 2D images, estimating the projection matrices corresponding to the projections of the object at various viewing angles and reconstructing in 3D the position of a marker on the basis of the estimation of the projection matrices.

Abstract: A disconnect clutch comprises a passive centrifugal valve that is utilized to disconnect parasitic loads from an engine during the initial start-up and to automatically connect the parasitic loads once the engine speed is sufficient. The passive centrifugal valve operates dependent on a rotational speed of a flywheel assembly of the clutch and is configured to switch at a threshold rotational speed of the flywheel assembly.

Abstract: Systems, methods, and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: Systems, methods and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: The present invention relates to topical compositions comprising a saccharide isomerate and at least one acryloyldimethyltaurate polymer comprising alkali metal ions and/or ammonium ions, characterized in that the ratio of alkali metal ions to ammonium ions in the acryloyldimethyltaurate moieties in the at least one polymer is at least 1 as well as to the use of an alkali metal acryloyldimethyltaurate polymer to suppress discoloration in topical compositions comprising a saccharide isomerate and an ammonium acryloyldimethyltaurate polymer.

Abstract: A method for calibrating a device D includes at least one radiation source and a detector, the radiation source and the detector being installed on at least one moving support, comprising at least the following elements: at least one first sensor positioned close to the radiation source and at least one second sensor positioned close to the detector, the two first and second sensors being configured to estimate through calculation a position Ps of the source and a position Pd of the detector, and a sensor for sensing the angular position of the moving support, a synchronization module configured to synchronously trigger the measurements of the sensors, a module for pre-processing the measurements of the sensors, the processing module comprising an input receiving an operating model M of the device and a data merging algorithm taking into account at least the two measurements of the sensors and the model M in order to estimate an accurate position value for the source Ps and for the detector Pd.

Abstract: A method for calculating during use the geometric parameters of an x-ray imaging system, an object or a patient to be observed being placed between the x-ray source and a detector of x-rays having passed through the object or patient, wherein it includes at least the following steps: detecting at least one marker on the object or the patient or in proximity to the object, the marker being of unknown 3D position, acquiring a plurality of 2D images for a plurality of viewpoints of the imaging system, detecting the position of at least one marker in each of the acquired 2D images, estimating the projection matrices corresponding to the projections of the object at various viewing angles and reconstructing in 3D the position of a marker on the basis of the estimation of the projection matrices.

Abstract: A disconnect clutch comprises a passive centrifugal valve that is utilized to disconnect parasitic loads from an engine during the initial start-up and to automatically connect the parasitic loads once the engine speed is sufficient. The passive centrifugal valve operates dependent on a rotational speed of a flywheel assembly of the clutch and is configured to switch at a threshold rotational speed of the flywheel assembly.

Abstract: A method for calibrating a device D includes at least one radiation source and a detector, the radiation source and the detector being installed on at least one moving support, comprising at least the following elements: at least one first sensor positioned close to the radiation source and at least one second sensor positioned close to the detector, the two first and second sensors being configured to estimate through calculation a position Ps of the source and a position Pd of the detector, and a sensor for sensing the angular position of the moving support, a synchronization module configured to synchronously trigger the measurements of the sensors, a module for pre-processing the measurements of the sensors, the processing module comprising an input receiving an operating model M of the device and a data merging algorithm taking into account at least the two measurements of the sensors and the model M in order to estimate an accurate position value for the source Ps and for the detector Pd.

Abstract: Systems, methods and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: A test pattern geometrically calibrates an x-ray imaging device to generate three-dimensional images of an object by reconstruction based on two-dimensional projections of the object, the calibrating test pattern comprising a volume support with markers having a radiological absorbance providing contrast to the volume support, the markers distributed in a three-dimensional pattern, in subsets substantially in parallel respective straight lines wherein sequences of cross-ratios are constructed from the respective subsets of markers. Each sequence of cross-ratios comprises a single cross-ratio for each quadruplet of markers in which quadruplet the markers are ordered depending on rank number of respective markers along the straight line they are aligned in a predefined first direction, the order being common to all cross-ratios.

Abstract: A method of manufacturing a thermoelectric material including: providing a half-Heusler compound of MgCuSn nanoparticles, obtaining a powder by mechanical alloying by using Mg chips, Si fine powder, Sn fine powder and Sb powder, the half-Heusler compound of MgCuSn nanoparticles and cyclohexane solution,wherein the weight percent V, of the cyclohexane solution is comprised between 0.5 wt % and 4.0 wt % and wherein the volume percent VHH of the Half-Heusler compound of MgCuSn nanoparticles satisfies: 1.4 vol %<VHH<2.0 vol %.

Abstract: A test pattern geometrically calibrates an x-ray imaging device to generate three-dimensional images of an object by reconstruction based on two-dimensional projections of the object, the calibrating test pattern comprising a volume support with markers having a radiological absorbance providing contrast to the volume support, the markers distributed in a three-dimensional pattern, in subsets substantially in parallel respective straight lines wherein sequences of cross-ratios are constructed from the respective subsets of markers. Each sequence of cross-ratios comprises a single cross-ratio for each quadruplet of markers in which quadruplet the markers are ordered depending on rank number of respective markers along the straight line they are aligned in a predefined first direction, the order being common to all cross-ratios.

Abstract: The invention relates to a sintered product containing: in mass percent based on oxides, more than 99.5% of a material having a cubic crystallographic structure for more than 95% of the mass thereof, known as a material with a cubic structure, and having a refractive index of less than 2.75 in the wavelength range between 0.2 ?m and 5 ?m; and more than 50 ppma of dopants which must contain titanium oxide TiO2 and at least one additional dopant selected from ZrO2, CaO, and MgO, in which said at least one additional dopant is different from the oxide(s) forming the material with a cubic structure, but can be MgO when the material with a cubic structure is spinel MgAl2O4.

Abstract: A sintered yttrium oxide product comprises: • more than 99.5% of yttrium oxide, as a percentage by weight; and • more than 50 ppma and less than 1000 ppma of a dopant selected from Zr&Ogr;2 and/or HfO2; the mean grain size of the doped yttrium oxide product being more than 0.03 ?m and less than 5 ?m.