Abstract:
An LED-based spectrophotometer uses a reconstruction algorithm, based on the spectral characteristics of the illumination source and the color sensing system, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. The reference database contains training samples that indicate reflectance spectra and their corresponding LED sensor output. A dynamic, Karhunen-Loeve-based (DKL) spectral reconstruction algorithm, used to reconstruct spectra, gives greater importance to the data from the training samples in the neighborhood of the color sample under measurement. This is done using linear operators and basis vectors.
Abstract:
Es werden ein System und ein Verfahren zum Identifizieren des Atmungsmusters eines Patienten zur kontaktlosen Bewertung der respiratorischen Funktion mit einem tiefenfähigen Bildgebungssystem offenbart. In einer Ausführungsform wird eine zeitveränderliche Tiefenkarten-Sequenz von einer Zielregion erfasst, wobei die Tiefenkarten verarbeitet werden, um ein Atmungssignal für den Probanden zu erhalten. Das Atmungssignal des Probanden umfasst eine zeitliche Sequenz von Momentanvolumina. Ein oder mehrere Segmente des Atmungssignals des Probanden werden dann mit einem oder mehreren Referenz-Atmungssignalen verglichen, die jeweils mit einem bekannten Atmungsmuster verknüpft sind. Als Resultat des Vergleichs wird ein Atmungsmuster für den Probanden identifiziert. Das identifizierte Atmungsmuster wird dann verwendet, um die respiratorische Funktion des Probanden zu bewerten. Die vorliegenden Lehren finden Verwendung in einer Vielzahl verschiedener medizinischen Anwendungen. Es werden verschiedene Ausführungsformen offenbart.
Abstract:
determinação de um número total de pessoas em uma imagem ir obtida através de um sistema de formação de imagem ir". a presente invenção refere-se a um sistema e método inovado- res para determinar o número de pessoas em uma imagem ir obtida com o uso de um sistema de formação de imagem ir. o presente metodo separa um humano do piano de fundo circundante através de um conjunto de ra- zões de intensidade particularmente formadas. as quantidades derivadas destas razões e valores limítrofes são usadas para classificar seletivamente se um pixei na imagem ir é de um humano ou de um não humano. com base na classificação dos vários pixeis na imagem ir, o número de pessoas é determinado. várias modalidades são reveladas.
Abstract:
Es werden ein neues System und ein Verfahren zur gleichzeitigen Spektralzerlegung offenbart, die für die Bildobjekterkennung und Kategorisierung von Szenen und zu analysierenden Objekten geeignet sind. Das vorliegende System nimmt unterschiedliche Spektralebenen gleichzeitig unter Anwendung eines Fabry-Perot-Mehrfachfiltergitters auf, wobei jeder Filter auf eine spezielle Wellenlänge abgestimmt ist. Ein Verfahren zum Klassifizieren von Pixel in dem aufgenommenen Bild wird ebenfalls bereitgestellt. Das vorliegende System und das Verfahren finden Verwendung in einer großen Bandbreite von Anwendungen, beispielsweise die Erkennung von besetzten Plätzen in einem Transportverwaltungssystem und bei der medizinischen Bildverarbeitung und Diagnose für die Gesundheitsverwaltung. Die hierin offenbarte Lehre findet auch Anwendung in Einsatzzwecken, in denen eine zweidimensionale Ansicht einer Szene aufgenommen werden muss und diese in spektrale Bänder zerlegt wird, so dass Objekte in dem Bild in geeigneter Weise erkannt werden können.
Abstract:
A full width array spectrophotometer for full width scanning color analysis of color test targets, with one or two substantially linear elongated arrays of closely spaced multiple LED illumination sources of plural different color emissions in a multiply repeated pattern of at least three or four different colors transversely spanning a printer paper path and sequentially illuminated to illuminate a transverse band across a printed sheet moving in the paper path, and a corresponding elongated low cost light imaging bar comprising a parallel and corresponding ly elongated array of multiple closely spaced different color sensitive (three or four rows of color-filtered) photodetectors, which imaging bar is positioned to detect and analyze light reflected from the transverse sequentially illuminated band.
Abstract:
An LED-based spectrophotometer uses a reconstruction algorithm, based on spectral information of an illumination source and a reference spectrophotometer, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. A non-linear model, such as a fuzzy inference system (FIS), is used to reconstruct spectra.
Abstract:
An LED-based spectrophotometer uses a reconstruction algorithm, based on the spectral characteristics of the illumination source and the color sensing system, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. The reference database contains training samples that indicate reflectance spectra and their corresponding LED sensor output. A dynamic, Karhunen-Loeve-based (DKL) spectral reconstruction algorithm, used to reconstruct spectra, gives greater importance to the data from the training samples in the neighborhood of the color sample under measurement. This is done using linear operators and basis vectors.
Abstract:
An LED-based spectrophotometer uses a reconstruction algorithm, based on the spectral characteristics of the illumination source and the color sensing system, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. The reference database contains training samples that indicate reflectance spectra and their corresponding LED sensor output. A dynamic, Karhunen-Loeve-based (DKL) spectral reconstruction algorithm, used to reconstruct spectra, gives greater importance to the data from the training samples in the neighborhood of the color sample under measurement. This is done using linear operators and basis vectors.
Abstract:
PROBLEM TO BE SOLVED: To provide a system and method for image-based determination of carbon dioxide (CO) concentration in exhaled breath.SOLUTION: An image of the exhaled airstream of a subject of interest is received. The image is captured using a mid-wave infrared camera system having an optical filter tuned to the infrared absorption band of CO. The image is preprocessed to isolate a region of pixels containing the exhaled airstream and intensity values of pixels in the identified region are normalized by a value of a known radiance such as that of the subject's nose or face. The image is analyzed to determine COconcentration levels of the exhaled airstream using a calibration curve which relates pixel intensity to COconcentrations. The calibration curve is derived using a physics-based parameterized model. The COconcentration levels are determined and communicated to a computer workstation 1328. Various embodiments are disclosed.