公开(公告)号：GB2522558A

公开(公告)日：2015-07-29

申请号：GB201501343

申请日：2013-06-27

Applicant: UNIV TSUKUBA ,  HAMAMATSU PHOTONICS KK

Inventor： KUDO HIROYUKI ,  SAITO NAOYA ,  UEDA YUKIO ,  YOSHIMOTO KENJI ,  YAMASHITA YUTAKA

IPC: A61B5/00

Abstract: This measurement data selection method: treats a vector comprising measurement data y1 to yN1 obtained for each combination among a plurality of light exit positions, a plurality of light detection positions, and a plurality of time slices in a time-resolved waveform as (1); treats a vector, which has pixel values of learning-use image data prepared beforehand as examples of internal image data as components, as x; and treats a system matrix for calculating the internal image data on the basis of the measurement data y1 to yN1 as A1. Then, a vector y which satisfies the following conditional expressions (2) and (3) or which satisfies the following conditional expression (4) is obtained by back-calculation. When measuring a subject, the internal image data are created using only measurement data corresponding to non-zero components of the vector y.

公开(公告)号：DE112013003284T5

公开(公告)日：2015-04-30

申请号：DE112013003284

申请日：2013-06-27

Applicant: HAMAMATSU PHOTONICS KK ,  UNIV TSUKUBA

Inventor： KUDO HIROYUKI ,  SAITO NAOYA ,  YOSHIMOTO KENJI ,  YAMASHITA YUTAKA ,  UEDA YUKIO

IPC: A61B10/00 ,  G01N21/17

Abstract: Messdaten-Auswahlverfahren für eine biometrische Vorrichtung, Lichtaustrittspositions-Ermittlungsverfahren für eine biometrische Vorrichtung und biometrische Vorrichtung Wenn bei dem Messdaten-Auswahlverfahren ein Vektor, der sich aus Messdaten y1 bis yN1 zusammensetzt, die für entsprechende Kombinationen einer Vielzahl von Lichtemissionspositionen, einer Vielzahl von Lichterfassungspositionen und einer Vielzahl von Auflösungszeiten in einer zeitaufgelösten Wellenform gewonnen worden sind, gegeben ist durch [Formel 1]ein Vektor, bei dem Bildpunktwerte von Lernbilddaten, die im Voraus als Beispiel für Innenbilddaten bereitgestellt werden, Komponenten sind, durch x gegeben ist und eine Systemmatrix zum Berechnen der Innenbilddaten aus den Messdaten y1 bis yN1 durch A1 gegeben ist, wird der Vektor y, der die folgenden bedingten Ausdrücke (2) und (3) erfüllt [Formel 2] min∥y∥00 (2)[Formel 3] ∥x – A1 Ty∥22 ≤ &egr;2 (3)oder der den folgenden bedingten Ausdruck (4) erfüllt [Formel 4] min(∥y∥00 + &bgr;∥x – A1 Ty∥22 )(4) ...

公开(公告)号：EP2730231A4

公开(公告)日：2015-03-25

申请号：EP12807701

申请日：2012-06-28

Applicant: HAMAMATSU PHOTONICS KK

Inventor： KUDO HIROYUKI ,  SAITO NAOYA ,  UEDA YUKIO ,  YOSHIMOTO KENJI ,  YAMASHITA YUTAKA

IPC: A61B5/00 ,  A61B6/00 ,  A61B6/03 ,  A61B8/15 ,  A61B10/00 ,  G01N21/17 ,  G01N21/47 ,  G06T11/00

CPC classification number: A61B5/7203 ,  A61B5/0073 ,  A61B5/0082 ,  A61B6/5247 ,  A61B8/15 ,  A61B8/5207 ,  A61B10/0041 ,  G01N21/4795 ,  G06T11/006

Abstract: The bioinstrumentation apparatus 10 includes a light irradiation unit irradiating a measurement region B with light, a light detection unit detecting diffused light from the measurement region, and a computing unit 14 generating a reconstructed image for the interior of the measurement region. The computing unit 14 calculates J coefficients w j set for every pixel of the reconstructed image and more than 0 and not more than 1 (where J is the number of pixels of the reconstructed image) and carries out successive approximation computation by the following iterative formula x j k + 1 = x j k + w j �¢ d j k (where k is an integer from 1 to N, N is the number of times of iterative computation, x j (k) is a pixel value of the jth pixel on the kth iterative computation, and d j (k) is an update amount of the jth pixel on the kth iterative computation) to generate the reconstructed image. Thereby, there are provided a bioinstrumentation apparatus and an image generating method capable of suppressing a difference in spatial resolution and noise characteristics depending on a position inside the measurement region to generate an image which is uniform to a greater extent.

公开(公告)号：JP2014027960A

公开(公告)日：2014-02-13

申请号：JP2012150793

申请日：2012-07-04

Applicant: Univ Of Tsukuba ,  国立大学法人 筑波大学 ,  Hamamatsu Photonics Kk ,  浜松ホトニクス株式会社

Inventor： KUDO HIROYUKI ,  SAITO NAOYA ,  UEDA YUKIO ,  YOSHIMOTO KENJI ,  YAMASHITA YUTAKA

IPC: A61B10/00

CPC classification number: G06K9/0014 ,  A61B5/0073 ,  A61B2576/00 ,  G06T7/70 ,  G06T2207/10101

Abstract: PROBLEM TO BE SOLVED: To shorten the creation time for image data by decreasing the number of measurement data required fro creating the image data.SOLUTION: In this measurement data selection method, when measurement data obtained for every combination of a plurality of light emission positions, a plurality of photo detecting positions and a plurality of resolution times in the time resolution waveform is taken as y, a vector, the component of which is a pixel value of learning image data previously provided as an example of internal image data is taken as an x, and a system matrix for calculating the internal image data from the measurement data y is taken as A, the following conditional expression is satisfied, or a vector y satisfying the following conditional expression is obtained by inverse operation, and in measuring a subject, internal image data is created using only measurement data corresponding to the component of the vector y that is not zero.

Abstract translation: 要解决的问题：通过减少创建图像数据所需的测量数据的数量来缩短图像数据的创建时间。解决方案：在该测量数据选择方法中，当对于多个发光位置的每个组合获得的测量数据 将时间分辨率波形中的多个光检测位置和多个分辨率时间取为y，将其分量作为内部图像数据的示例的先前提供的学习图像数据的像素值的矢量作为 将x和用于从测量数据y计算内部图像数据的系统矩阵作为A，满足以下条件表达式，或者通过反向操作获得满足以下条件表达式的向量y，并且在测量对象 ，仅使用对应于不为零的矢量y的分量的测量数据来创建内部图像数据。

公开(公告)号：JP2013019696A

公开(公告)日：2013-01-31

申请号：JP2011151086

申请日：2011-07-07

Applicant: Univ Of Tsukuba ,  国立大学法人 筑波大学 ,  Hamamatsu Photonics Kk ,  浜松ホトニクス株式会社

Inventor： KUDO HIROYUKI ,  SAITO NAOYA ,  UEDA YUKIO ,  YOSHIMOTO KENJI ,  YAMASHITA YUTAKA

IPC: G01N21/17 ,  A61B6/03 ,  A61B8/00 ,  G01N22/02 ,  G01N29/00 ,  G01N29/44

CPC classification number: A61B5/7203 ,  A61B5/0073 ,  A61B5/0082 ,  A61B6/5247 ,  A61B8/15 ,  A61B8/5207 ,  A61B10/0041 ,  G01N21/4795 ,  G06T11/006

Abstract: PROBLEM TO BE SOLVED: To provide a biological measurement apparatus and an image formation method that can form an image which is as uniform as possible by suppressing differences in spatial resolution and noise characteristic with positions in a region to be measured.SOLUTION: A biological measurement apparatus 10 includes a light irradiation part which irradiates a region B to be measured with light, an optical detection part which detects diffused light from the region to be measured, and an arithmetic part 14 which forms a reconstructed image associated with the interior of the region to be measured. The operation part 14 forms the reconstructed image by calculating J coefficients w(J is the number of pixels of the reconstructed image) which are set by the pixels of the reconstructed image, and larger than 0 and equal to or less than 1, and performing a sequential approximation operation using an iterative expression (where (k) is an integer of 1 to N, N is the number of times of iterative operation, xis a pixel value in a (k)th iterative operation of a (j)th pixel, and dis an update quantity in the (k)th iterative operation of the (j)th pixel).

Abstract translation: 解决的问题：提供一种可以通过抑制在待测区域中的位置的空间分辨率和噪声特性的差异来形成尽可能均匀的图像的生物测量装置和图像形成方法。 解决方案：生物测量装置10包括用光照射待测区域B的光照射部分，检测来自待测区域的漫射光的光检测部分和形成重建的运算部分14的运算部分14。 与要测量的区域的内部相关联的图像。 操作部分14通过计算由重构图像的像素设置的J个系数，从而形成重构图像，并且 大于0且等于或小于1，并且使用迭代表达式（其中（k）是1到N的整数）执行顺序近似运算，N是迭代操作的次数，x 是第（j）个像素的（k）次迭代运算中的像素值，d 是第（j）个像素的（k）次迭代运算中的更新量。 版权所有（C）2013，JPO＆INPIT