中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

35 records (took 0.017 seconds)

    LEARNING APPARATUS AND METHOD, PREDICTION APPARATUS AND METHOD, AND COMPUTER READABLE MEDIUM
    21.
    发明申请
    LEARNING APPARATUS AND METHOD, PREDICTION APPARATUS AND METHOD, AND COMPUTER READABLE MEDIUM 有权

    公开(公告)号：US20220128988A1

    公开(公告)日：2022-04-28

    申请号：US17431261

    申请日：2019-02-19

    Applicant: NEC Corporation

    Inventor： Azusa SAWADA ,  Takashi SHIBATA ,  Katsuhiko TAKAHASHI

    IPC: G05B23/02

    Abstract: A data-series group includes data series which is a series of data obtained by observing the same object at discrete times. Time labels are time information added to respective data included in the data-series group. State labels are added to some of the data included in the data-series group. A loss-function control unit determines a loss function to be used for learning based on the time labels and the state labels. A threshold is used to adjust a branch condition of the loss-function control unit. A regressor is a model, and is used to detect an abnormality or predict a remaining life span. A dictionary stores parameters of the regressor. A regressor training unit trains the regressor based on the loss function determined by the loss-function control unit.

    PATTERN RECOGNITION SYSTEM, PARAMETER GENERATION METHOD, AND PARAMETER GENERATION PROGRAM
    22.
    发明申请
    PATTERN RECOGNITION SYSTEM, PARAMETER GENERATION METHOD, AND PARAMETER GENERATION PROGRAM 有权

    公开(公告)号：US20210027110A1

    公开(公告)日：2021-01-28

    申请号：US17044425

    申请日：2018-05-10

    Applicant: NEC Corporation

    Inventor： Katsuhiko TAKAHASHI ,  Hiroyoshi MIYANO ,  Tetsuo lNOSHITA

    IPC: G06K9/62

    Abstract: The first parameter generation unit 811 generates a first parameter, which is a parameter of a first recognizer, using first learning data including a combination of data to be recognized, a correct label of the data, and domain information indicating a collection environment of the data. The second parameter generation unit 812 generates a second parameter, which is a parameter of a second recognizer, using second learning data including a combination of data to be recognized that is collected in a predetermined collection environment, a correct label of the data, and target domain information indicating the predetermined collection environment, based on the first parameter. The third parameter generation unit 813 integrates the first parameter and the second parameter to generate a third parameter to be used for pattern recognition of input data by learning using the first learning data.

    LEARNING DEVICE, INSPECTION SYSTEM, LEARNING METHOD, INSPECTION METHOD, AND PROGRAM
    23.
    发明申请
    LEARNING DEVICE, INSPECTION SYSTEM, LEARNING METHOD, INSPECTION METHOD, AND PROGRAM 审中-公开

    公开(公告)号：US20200334801A1

    公开(公告)日：2020-10-22

    申请号：US16769784

    申请日：2017-12-06

    Applicant: NEC Corporation

    Inventor： Katsuhiko TAKAHASHI ,  Takashi SHIBATA

    IPC: G06T7/00

    Abstract: In the present invention, a first image acquisition means 81 acquires a first image of an inspection target including an abnormal part. A second image acquisition means 82 acquires a second image of the inspection target captured earlier than the time when the first image is captured. A learning data generation means 83 generates learning data indicating that the second image includes an abnormal part. A learning means 84 learns a discrimination dictionary by using the learning data generated by the learning data generation means 83.

    ANOMALY DETECTION APPARATUS, ANOMALY DETECTION METHOD, AND COMPUTER-READABLE RECORDING MEDIUM
    24.
    发明申请
    ANOMALY DETECTION APPARATUS, ANOMALY DETECTION METHOD, AND COMPUTER-READABLE RECORDING MEDIUM 审中-公开

    公开(公告)号：US20200311894A1

    公开(公告)日：2020-10-01

    申请号：US16651649

    申请日：2017-09-29

    Applicant: NEC CORPORATION

    Inventor： Katsuhiko TAKAHASHI ,  Yuichi NAKATANI

    IPC: G06T7/00

    Abstract: An anomaly detection apparatus 100 includes an image transformation unit 103 that calculates an image transformation parameter, based on an inspection image in which an inspection object appears, a reference image indicating a normal state of the inspection object and a parameter for image transformation parameter calculation, and performs image transformation on the inspection image using the image transformation parameter, an image change detection unit 104 that collates the reference image and the image-transformed inspection image using a change detection parameter, and calculates an anomaly certainty factor indicating whether there is a change in a specific region of the inspection image, a change detection parameter learning unit 106 that learns the change detection parameter, based on a difference between a training image indicating a correct answer value of the change and the anomaly certainty factor, and an image transformation parameter learning unit 108 that learns the parameter for image transformation parameter calculation, based on a collection amount derived from the difference between the training image and the anomaly certainty factor and to be applied to the inspection image that has undergone image transformation.

    OBJECT DETECTION APPARATUS, OBJECT DETECTION METHOD, AND COMPUTER-READABLE RECORDING MEDIUM
    25.
    发明申请
    OBJECT DETECTION APPARATUS, OBJECT DETECTION METHOD, AND COMPUTER-READABLE RECORDING MEDIUM 审中-公开

    公开(公告)号：US20200242391A1

    公开(公告)日：2020-07-30

    申请号：US16753650

    申请日：2017-10-06

    Applicant: NEC CORPORATION

    Inventor： Katsuhiko TAKAHASHI

    IPC: G06K9/46 ,  G06K9/00 ,  G06K9/42

    Abstract: An object detection apparatus 100 is provided with: a fish-eye image acquisition unit 10 configured to acquire a time series fish-eye image; a horizontal panorama image generation unit 20 configured to, for each frame, perform conversion to a horizontal panorama image in which a vertical direction in a real space is expressed in a perpendicular direction of the frame, and an azimuth is expressed equiangularly in a horizontal direction of the frame; an edge pair extraction unit 30 configured to extract a pair of edges in the perpendicular direction from the horizontal panorama image; a change rate extraction unit 40 configured to extract a change rate of an inter-edge distance between the pair of edges; a lower end region extraction unit 50 configured to extract a region of a lower end of an object providing the pair of edges; a distance change rate extraction unit 60 configured to calculate a distance from the object to the fish-eye camera based on the position of the region of the lower end of the object in the horizontal panorama image, and extract a change rate of the distance; and an object detection unit 70 configured to determine whether or not the object exists based on the change rate of the inter-edge distance and the change rate of the distance.

    Image Recognition Apparatus and Storage Medium
    27.
    发明申请
    Image Recognition Apparatus and Storage Medium 有权
    图像识别装置和存储介质

    公开(公告)号：US20160155012A1

    公开(公告)日：2016-06-02

    申请号：US14904356

    申请日：2013-07-24

    Applicant: NEC SOLUTION INNOVATORS, LTD. ,  NEC CORPORATION

    Inventor： Katsuhiko TAKAHASHI ,  Masateru MORI

    IPC: G06K9/46 ,  G06K9/00

    CPC classification number: G06K9/4633 ,  G06K9/00986

    Abstract: A field-programmable gate array (FPGA) coarse Hough transform unit (102) performs on an FPGA a first coarse-precision Hough transform upon an image that has been applied as input to an image input unit (101), and supplies candidate location information obtained by the first Hough transform and the image to an external memory (103). A fine Hough transform unit (104) reads the candidate location information and the image stored in the external memory (103), uses the candidate location information to perform, on a general-purpose processor, a second detailed-precision Hough transform upon the image, and supplies detailed-precision location information of a two-dimensional pattern that is the object of recognition within the image to the external memory (103).

    Abstract translation: 现场可编程门阵列（FPGA）粗霍夫变换单元（102）在已经作为输入应用于图像输入单元（101）的图像上在FPGA上执行第一粗略精度霍夫变换，并且提供候选位置信息 通过第一霍夫变换获得的图像和外部存储器（103）。 精细霍夫变换单元（104）读取候选位置信息和存储在外部存储器（103）中的图像，使用候选位置信息在通用处理器上对图像执行第二详细精确霍夫变换 并且将作为图像内的识别对象的二维图案的详细精确位置信息提供给外部存储器（103）。

Patent Agency Ranking