摘要:
L'invention concerne un procédé de restauration d'une élocution altérée par traitement d'un signal électrique, provenant d'un microphone, représentatif de cette élocution, et comportant les principales étapes suivantes: - extraction et séparation de l'excitation originale et des paramètres de l'articulation, - restauration de l'excitation originale à partir d'éléments issus de l'élocution altérée et d'éléments issus d'une élocution saine, et - reconstruction d'un signal de parole restauré à partir des paramètres de l'articulation et de l'excitation originale restaurée afin de produire un signal acoustique représentatif de l'élocution reconstruite.
摘要:
The method is useful for quantitative evaluation of bead-based lab-on-a-chip affinity assay fluorescence measurements. Fluorescent beads coated with probe molecules are immobilized (11) on a substrate. The fluorescence of the immobilized beads is measured (12) and represents the number of beads. The beads are then perfused with an analyte solution such that analyte molecules are bound to the beads. The analyte molecules are labeled with a second, different fluorescent marker which is used in a second fluorescence measurement (14) to determine the number of analyte molecules bound to the beads. A calculation (15), e.g., a quotient formation, is performed with the first (12) and the second (14) fluorescence measurement.
摘要:
A method for replication (recombination by applying the replication in step and repeat mode) of three-dimensional microstructures with high accuracy and high replication predictability is described. The method uses UV-curable materials or thermoplastic polymers to replicate a single element on an arbitrarily curved surface. The high replication predictability is achieved by using a multistep process. A coarse replication is made in a first step. In subsequent steps replications with high accuracy can be made by using the first coarse replica as a raw preform. The method is suited for the replication of macroscopic structures in three dimensions (from micrometer to some mm or more in relief height) which cannot be achieved by the usual, well known replication technologies for quasi two-dimensional microstructures.
摘要:
The pixel for use in an image sensor comprises a low-doped semiconductor substrate (A). On the substrate (A), an arrangement of a plurality of floating areas, e.g., floating gates (FG1-FG7), is provided. Neighboring floating gates are electrically isolated from each other yet capacitively coupled to each other. By applying a voltage (V2 - V1) to two of the floating gates (FG1, FG7), a lateral steplike electric field is generated. Photogenerated charge carriers move along the electric-field lines to the point of highest potential energy, where a floating diffusion (D) accumulate the photocharges. The charges accumulated in the various pixels are sequentially read out with a suitable circuit known from image-sensor literature, such as a source follower or a charge amplifier with row and column select mechanisms. The pixel of offers at the same time a large sensing area, a high photocharge-detection sensitivity and a high response speed, without any static current consumption.
摘要:
The optical proximity detector (10) comprises a flexible substrate (1) on which a plurality of light sources (6) and photodetectors (8) are arranged, consisting of thin, flexible layers of organic semiconductors (3.1, 3.2) and transparent electrodes (2.1, 2.2, 4.1, 4.2). The light sources (6) emit continuous or pulsed light (7) through the transparent electrodes (2.1, 2.2, 4.1, 4.2). An object (9) close to the light-emitting surface of the detector (10) partially reflects this light (7), and a fraction of it reaches an adjacent photodetector (8) that converts this light into an electrical signal. This signal contains information about the proximity of the object (9), and the information can be extracted with an electronic circuit that is either measuring the amount of detected continuous light or demodulating the detected light pulses.
摘要:
The invention concerns a cell comprising a photosensitive element (1), first addressable means (2) for deactivating said photosensitive element after an exposure phase, second addressable means (9, 10) for, during said exploration phase, retrieving from the cell the luminance signal generated by the photosensitive element, third addressable means (6) for transferring the luminance signal from the photosensitive element to the second addressable means, storage means (cp1 to cp3) for preserving the luminance signal between the exposure phase and the time when the cell is addressed for activating the second addressable means and fourth addressable means (14) for unloading said storage means after the cell has been read. The invention is characterised in that the third (6) and fourth (14) means are placed in a casing arranged in the substrate and having a type of conductivity opposite to that of the substrate.
摘要:
A method for replication (recombination by applying the replication in step and repeat mode) of three-dimensional microstructures with high accuracy and high replication predictability is described. The method uses UV-curable materials or thermoplastic polymers to replicate a single element on an arbitrarily curved surface. The high replication predictability is achieved by using a multistep process. A coarse replication is made in a first step. In subsequent steps replications with high accuracy can be made by using the first coarse replica as a raw preform. The method is suited for the replication of macroscopic structures in three dimensions (from micrometer to some mm or more in relief height) which cannot be achieved by the usual, well known replication technologies for quasi two-dimensional microstructures.
摘要:
Les instructions de programme utilisées dans ce dispositif comprennent un code d'élaboration (CE) des adresses pour l'accès à une mémoire (7) associée au processeur. A chacun des registres d'adresse (ax0 à ax3; ay0 à ay3) sont associés un registre d'offset (ox0 à ox3; oy0 à oy3) et un registre de modulo (mx0 à mx3; my0 à my3). Un circuit de calcul d'adresses (28) calcule chaque adresse d'accès à la mémoire (7) sur la base de codes d'opération désignés par le code d'élaboration d'adresses (CE) des instructions et du contenu d'un registre d'adresse (ax0 à ax3; ay0 àay3) et de ses registres associés (0x0 à ox3; oy0 à oy3; mx0 à mx3; my0 à my3). Chaque code d'élaboration (CE) définit l'un des codes d'opération à envoyer au circuit de calcul (28) et désigne le registre d'adresse dont le contenu doit subir l'opération de calcul. Des moyens de commande assurent les transferts des contenus des registres et du code d'opération vers le circuit de calcul (28). Selon l'invention, chaque registre d'adresse (ax0 à ax3; ay0 à ay3) est associé à un registre de configuration (cx0 à cx3; cy0 à cy3) qui est désigné en même temps que ce registre d'adresse par le code d'élaboration d'adresses (CE), et chaque registre de configuration (cx0 à cx3; cy0 à cy3) contient au moins un jeu de codes d'opération prédéfinis, chaque code d'opération étant à même de commander une opération de calcul prédéterminée dans le circuit de calcul (28).