中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

27 records (took 0.02 seconds)

    ENDOSCOPE POSITION SPECIFYING DEVICE, METHOD, AND PROGRAM
    11.
    发明申请
    ENDOSCOPE POSITION SPECIFYING DEVICE, METHOD, AND PROGRAM 审中-公开

    公开(公告)号：US20180263712A1

    公开(公告)日：2018-09-20

    申请号：US15868689

    申请日：2018-01-11

    Applicant: FUJIFILM Corporation

    Inventor： Yoshiro KITAMURA

    IPC: A61B34/20 ,  A61B5/06 ,  A61B1/267 ,  A61B1/04

    CPC classification number: A61B34/20 ,  A61B1/00009 ,  A61B1/041 ,  A61B1/2676 ,  A61B5/065 ,  A61B2034/2057 ,  A61B2576/02

    Abstract: A hole portion detection unit detects a hole portion of the bronchus from at least one of a first endoscope image or a second endoscope image temporally earlier than the first endoscope image. A first parameter calculation unit calculates a first parameter indicating the amount of parallel movement for matching the hole portions of the two endoscope images with each other. A second parameter calculation unit performs alignment between the two endoscope images based on the first parameter, and calculates a second parameter including the amount of enlargement and reduction. Based on the two parameters, a movement amount calculation unit calculates the amount of movement of the endoscope from the acquisition time of the second endoscope image to the acquisition time of the first endoscope image.

    THREE-DIMENSIONAL DATA PROCESSING SYSTEM, METHOD, AND PROGRAM, THREE-DIMENSIONAL MODEL, AND THREE-DIMENSIONAL MODEL SHAPING DEVICE
    13.
    发明申请
    THREE-DIMENSIONAL DATA PROCESSING SYSTEM, METHOD, AND PROGRAM, THREE-DIMENSIONAL MODEL, AND THREE-DIMENSIONAL MODEL SHAPING DEVICE 审中-公开

    公开(公告)号：US20170316619A1

    公开(公告)日：2017-11-02

    申请号：US15654981

    申请日：2017-07-20

    Applicant: FUJIFILM Corporation

    Inventor： Yoshiro KITAMURA

    IPC: G06T19/20 ,  G06T1/00 ,  B33Y50/00 ,  G09B23/30 ,  B33Y30/00

    CPC classification number: G06T19/20 ,  B29C64/386 ,  B33Y30/00 ,  B33Y50/00 ,  G06T1/0007 ,  G06T2219/2016 ,  G09B23/30

    Abstract: Three-dimensional data in which different three-dimensional patterns are respectively added to a plurality of positions of the three-dimensional data representing a three-dimensional object in a three-dimensional coordinate system is created, and the respective added three-dimensional patterns are stored in association with the positions in the three-dimensional data to which the three-dimensional patterns has been added. The three-dimensional model is shaped using the created three-dimensional data. A pattern is recognized in a captured image obtained by imaging the three-dimensional model that is shaped and of which a desired part is excised or incised, a three-dimensional pattern including the recognized pattern is searched for from the stored three-dimensional patterns, and the position in the three-dimensional data stored in association with the three-dimensional pattern that has been searched for is associated with a position on the captured image in which the pattern has been recognized.

    IMAGE CLASSIFYING APPARATUS, IMAGE CLASSIFYING METHOD, AND IMAGE CLASSIFYING PROGRAM
    14.
    发明申请
    IMAGE CLASSIFYING APPARATUS, IMAGE CLASSIFYING METHOD, AND IMAGE CLASSIFYING PROGRAM 审中-公开

    公开(公告)号：US20170277977A1

    公开(公告)日：2017-09-28

    申请号：US15464997

    申请日：2017-03-21

    Applicant: FUJIFILM Corporation

    Inventor： Yoshiro KITAMURA

    IPC: G06K9/62 ,  G06T7/00

    CPC classification number: G06K9/6267 ,  G06N3/08 ,  G06T7/0012 ,  G06T7/11 ,  G06T2200/04 ,  G06T2207/10072 ,  G06T2207/20084 ,  G06T2207/30004

    Abstract: [Objective]To enable a three dimensional image to be accurately classified into a plurality of classes with a small amount of calculations, in an image classifying apparatus, an image classifying method, and an image classifying program.[Constitution]A three dimensional image is classified into a plurality of classes by a convolutional neural network, in which a plurality of processing layers are connected hierarchically. The convolutional neural network includes: a convoluting layer that administers a convoluting process on each of a plurality of two dimensional images, which are generated by the neural network administering a projection process on the three dimensional image according to a plurality of processing parameters; and a pooling layer that pools the values of the same position within each of the plurality of two dimensional images which have undergone the convoluting process.

    LEARNING METHOD, LEARNING SYSTEM, LEARNED MODEL, PROGRAM, AND SUPER RESOLUTION IMAGE GENERATING DEVICE
    17.
    发明申请
    LEARNING METHOD, LEARNING SYSTEM, LEARNED MODEL, PROGRAM, AND SUPER RESOLUTION IMAGE GENERATING DEVICE 有权

    公开(公告)号：US20210374911A1

    公开(公告)日：2021-12-02

    申请号：US17400142

    申请日：2021-08-12

    Applicant: FUJIFILM Corporation

    Inventor： Akira KUDO ,  Yoshiro KITAMURA

    IPC: G06T3/40 ,  G06T5/00 ,  G06T5/20 ,  G06N3/04 ,  G06N3/08

    Abstract: Provided are a learning method and a learning system of a generative model, a program, a learned model, and a super resolution image generating device that can handle input data of any size and can suppress the amount of calculation at the time of image generation. A learning method according to an embodiment of the present disclosure is a learning method for performing machine learning of a generative model that estimates, from a first image, a second image including higher resolution image information than the first image, the method comprising using a generative adversarial network including a generator which is the generative model and a discriminator which is an identification model that identifies whether provided data is data of a correct image for learning or data derived from an output from the generator and implementing a self-attention mechanism only in a network of the discriminator among the generator and the discriminator.

Patent Agency Ranking