中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

18 records (took 0.02 seconds)

    IMAGE PROCESSING METHOD, IMAGE PROCESSING APPARATUS, AND RECORDING MEDIUM
    2.
    发明公开
    IMAGE PROCESSING METHOD, IMAGE PROCESSING APPARATUS, AND RECORDING MEDIUM 审中-公开
    图像处理方法，图像处理装置和记录介质

    公开(公告)号：EP3217358A1

    公开(公告)日：2017-09-13

    申请号：EP17158268.7

    申请日：2017-02-28

    Applicant: Panasonic Intellectual Property Corporation of America

    Inventor： KOZUKA, Kazuki ,  TANIGAWA, Toru ,  SAITO, Masahiko

    IPC: G06T7/20 ,  G06K9/00 ,  G06K9/62

    CPC classification number: G06K9/4604 ,  G06K9/00362 ,  G06K9/00805 ,  G06K9/6259 ,  G06T11/00 ,  G06T11/60 ,  G08G1/04 ,  G08G1/166 ,  H04N7/183

    Abstract: An image processing method includes acquiring consecutive time-series images captured by an onboard camera and including at least one image having a first annotation indicating a first region; determining, for each of the images, in reverse chronological order from an image of the last time point, whether the first region exists in the image based on whether the first annotation is attached; identifying the first image of a first time point for which the first region is determined not to exist, and setting a second region including a partial region of an object in the identified first image, indicating the moving object that is obstructed by the object before appearing on the path, and having dimensions based on dimensions of the first region in an image of a second time point immediately after the first time point; and attaching a second annotation to the image corresponding to the second time point, the second annotation indicating the second region.

    Abstract translation: 一种图像处理方法，包括：获取由车载照相机拍摄的并且包括具有指示第一区域的第一注释的至少一个图像的连续时间序列图像; 根据上一时间点的图像以反向时间顺序为每个图像确定第一区域是否存在于图像中，基于是否附加了第一注释; 识别第一区域被确定为不存在的第一时间点的第一图像，并且设置包括被识别的第一图像中的物体的部分区域的第二区域，其指示出现之前被物体阻挡的移动物体 在所述路径上并且具有基于紧接在所述第一时间点之后的第二时间点的图像中的所述第一区域的尺寸的尺寸; 以及将第二注释附加到对应于第二时间点的图像，第二注释指示第二区域。

    LEARNING METHOD, LEARNING DEVICE, AND PROGRAM
    3.
    发明公开
    LEARNING METHOD, LEARNING DEVICE, AND PROGRAM 审中-实审

    公开(公告)号：EP4350612A1

    公开(公告)日：2024-04-10

    申请号：EP22811341.1

    申请日：2022-05-25

    Applicant: Panasonic Intellectual Property Corporation of America

    Inventor： FUJIYOSHI, Hironobu ,  YAMASHITA, Takayoshi ,  HIRAKAWA, Tsubasa ,  KOZUKA, Kazuki

    IPC: G06T7/00 ,  G06V10/774

    Abstract: In this learning method, a first image is generated by applying a noise to a first region, a second image is generated by applying a noise to a second region, a composite image is generated by performing weighted addition of the first image and the second image, a first teacher label (y1) with respect to the first image is generated, a second teacher label (y2) with respect to the second image is generated, a composite teacher label (y) is generated by performing weighted addition of the first teacher label (y1) and the second teacher label (y2), and a learning model is generated by using the composite image and the composite teacher label (y) to perform machine learning.

    DATA AUGMENTATION METHOD, LEARNING DEVICE, AND PROGRAM
    4.
    发明公开
    DATA AUGMENTATION METHOD, LEARNING DEVICE, AND PROGRAM 审中-公开

    公开(公告)号：EP4242928A1

    公开(公告)日：2023-09-13

    申请号：EP21889261.0

    申请日：2021-11-05

    Applicant: Panasonic Intellectual Property Corporation of America

    Inventor： ISHIZAKA, Shun ,  KOZUKA, Kazuki ,  TSUKIZAWA, Sotaro ,  GUDOVSKIY, Denis

    IPC: G06N3/08

    Abstract: First optimization processing (S11) for optimizing parameters of a DNN and second optimization processing (S12) for optimizing hyperparameters for each sample used in data augmentation processing are alternately performed. The first optimization processing includes causing the DNN to predict a first augmentation label from a first augmented sample obtained by performing data augmentation processing on a first sample included in the training data set, calculating a first error function between the first augmentation label and a first correct label for the first sample, and updating the parameters in accordance with the first error function. The second optimization processing includes acquiring a second sample from an evaluation data set that is similar in distribution to a test data set, causing the DNN after the updating of the parameters to predict a second label from the second sample, calculating a second error function between the second label and a second correct label for the second sample, and updating the hyperparameter in accordance with a gradient obtained by differentiation of the second error function with respect to the hyperparameter.

    INFORMATION PRESENTATION CONTROL APPARATUS, AUTONOMOUS VEHICLE, AND AUTONOMOUS-VEHICLE DRIVING SUPPORT SYSTEM
    5.
    发明公开
    INFORMATION PRESENTATION CONTROL APPARATUS, AUTONOMOUS VEHICLE, AND AUTONOMOUS-VEHICLE DRIVING SUPPORT SYSTEM 审中-公开
    信息显示控制装置，自主车辆和自主车辆驾驶辅助系统

    公开(公告)号：EP3232289A1

    公开(公告)日：2017-10-18

    申请号：EP17165722.4

    申请日：2017-04-10

    Applicant: Panasonic Intellectual Property Corporation of America

    Inventor： ISHII, Yasunori ,  KOZUKA, Kazuki ,  FUJIMURA, Ryota

    IPC: G05D1/00 ,  G05D1/02

    CPC classification number: G06K9/00791 ,  B60R1/00 ,  B60R2300/60 ,  G01C21/26 ,  G01S19/42 ,  G05D1/0061 ,  G05D1/0088 ,  G05D1/021 ,  G05D1/0246 ,  G06F3/04883

    Abstract: An information presentation control apparatus includes a selection information obtainer and a presentation controller. The selection information obtainer obtains selection information representing a selection status of a plurality of recognizers that recognize different targets in surroundings of an autonomous vehicle. The presentation controller causes a presentation device mounted in the autonomous vehicle to present driving information in accordance with the selection information, the driving information being based on at least one of control that is executable by the autonomous vehicle and control that is not executable by the autonomous vehicle and being information about at least one of driving by an automated driving system of the autonomous vehicle and driving by a driver.

    Abstract translation: 信息呈现控制装置包括选择信息获取器和呈现控制器。 选择信息获取器获得选择信息，该选择信息表示识别自动车辆周围的不同目标的多个识别器的选择状态。 呈现控制器使安装在自主车辆中的呈现装置根据选择信息呈现驾驶信息，驾驶信息基于可由自主车辆执行的控制和不能由自主执行的控制中的至少一个 并且是关于由自动车辆的自动驾驶系统驾驶和由驾驶员驾驶中的至少一个的信息。

    RISK PREDICTION METHOD
    6.
    发明公开
    RISK PREDICTION METHOD 审中-公开
    风险预测方法

    公开(公告)号：EP3217332A1

    公开(公告)日：2017-09-13

    申请号：EP17160068.7

    申请日：2017-03-09

    Applicant: Panasonic Intellectual Property Corporation of America

    Inventor： KOZUKA, Kazuki ,  ISHII, Yasunori ,  SAITO, Masahiko ,  FUCHIKAMI, Tetsuji

    IPC: G06N3/04

    CPC classification number: G06N3/08 ,  G06K9/00805 ,  G06K9/66 ,  G06N3/0454 ,  G06N5/04 ,  G08G1/166 ,  H04N7/183

    Abstract: A risk prediction method executed by a computer of a risk predictor using a convolutional neural network, the method including making the convolutional neural network acquire an input image taken by an in-vehicle camera installed on a vehicle, making the convolutional neural network estimate a risk area and a feature of the risk area, the risk area being in the acquired input image, the risk area having a possibility that a moving object may appear from the risk area into a travelling path of the vehicle and the moving object may collide with the vehicle in a case where the vehicle simply continues running, and making the convolutional neural network output the estimated risk area and the estimated feature of the risk area as a risk predicted for the input image.

    Abstract translation: 由使用卷积神经网络的风险预测器的计算机执行的风险预测方法，所述方法包括使卷积神经网络获取由安装在车辆上的车载摄像机拍摄的输入图像，使得卷积神经网络估计风险 区域和危险区域的特征，所述危险区域位于所获取的输入图像中，所述危险区域可能存在移动物体从危险区域出现到车辆的行驶路径中并且移动物体可能与 在车辆简单地继续行驶并且使得卷积神经网络输出估计的风险区域和风险区域的估计特征作为针对输入图像预测的风险的情况下的车辆。

Patent Agency Ranking