Patents by Inventor Hiroki Nakano

Hiroki Nakano has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 11001263
    Abstract: A vehicular breaking force control system includes: a plurality of actuators capable of generating a braking force for a vehicle; a coasting state detection unit configured to detect that a coasting state has been established; a target braking force calculation unit configured to calculate a target braking force on the basis of a state of the vehicle when the coasting state detection unit detects that the coasting state has been established; and a braking force distribution control unit configured to determine a distribution braking force that is a braking force to be caused to be generated by each actuator, such that the distribution braking force is equal to or less than a braking force generable by the actuator and a sum of the distribution braking forces is equal to the target braking force, and to perform control of causing each actuator to generate the distribution braking force.
    Type: Grant
    Filed: February 11, 2019
    Date of Patent: May 11, 2021
    Assignee: TOYOTA JIDOSHA KABUSHIKI KAISHA
    Inventors: Masato Shimizu, Shun Sato, Katsumi Kono, Atsushi Ayabe, Noritake Mitsutani, Hiroki Nakano
  • Patent number: 10998134
    Abstract: This capacitor includes: a capacitor element; a case made of metal and configured to house the capacitor element; and a thermosetting resin that is filled in the case. The case includes a bottom face and a side face, the side face surrounding four sides of the bottom face. The side face has formed therein a plurality of slit parts, the plurality of slit parts extending from an end, of the side face, at an opposite side to the bottom face toward the bottom face side.
    Type: Grant
    Filed: September 21, 2018
    Date of Patent: May 4, 2021
    Assignee: PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO., LTD.
    Inventors: Hiroki Takeoka, Takashi Nakano
  • Publication number: 20210109036
    Abstract: Apparatus inspects the presence/absence of foreign substance on object having inspection region and non-inspection region arranged outside the inspection region. The apparatus includes sensor for illuminating the object and output, as image, result acquired by detecting light from region including the inspection region, and processor for detecting foreign substance based on inspection region image acquired by excluding non-inspection region image, which is image of the non-inspection region, from the image output from the sensor. The non-inspection region image includes first part generated by light from predetermined part of the non-inspection region of the inspected object and second part whose pixel value is continuous from pixel value of the first part and the processor specifies the second part based on fact that the pixel value of the second part is continuous from that of the first part.
    Type: Application
    Filed: October 2, 2020
    Publication date: April 15, 2021
    Inventors: Masayoshi Iino, Hiroki Nakano, Yasuhiro Yazawa, Kohei Maeda, Daisuke Nakajima
  • Publication number: 20210097686
    Abstract: Systems and methods are provided for recognizing pathological images captured by alternate image capturing devices. In embodiments, a computer-implemented method includes: obtaining, by a computing device, a classifier generated based on a plurality of standardized training pathology images associated with a known classification and generated by a first type of device; receiving, by the computing device, an alternate pathology image captured by a second type of device; standardizing, by the computing device, the alternate pathology image; determining, by the computing device, a classification of the alternate pathology image based on applying the generated classifier; and outputting, by the computing device, information regarding the determined classification to aid in diagnosis of a medical condition represented by the alternate pathology image.
    Type: Application
    Filed: December 10, 2020
    Publication date: April 1, 2021
    Inventors: Yusuke TAKEUCHI, Yoshinori KABEYA, Hiroki NAKANO, Issei OZAWA, Sho YONEZAWA
  • Patent number: 10936912
    Abstract: Image classification using a generated mask image is performed by generating a mask image that extracts a target area from an input image, extracting an image feature map of the input image by inputting the input image in a first neural network including at least one image feature extracting layer, masking the image feature map by using the mask image, and classifying the input image by inputting the masked image feature map to a second neural network including at least one classification layer.
    Type: Grant
    Filed: November 1, 2018
    Date of Patent: March 2, 2021
    Assignee: International Business Machines Corporation
    Inventors: Hiroki Nakano, Takuya Goto, Masaharu Sakamoto
  • Patent number: 10915810
    Abstract: A cascading convolutional neural network (CCNN) comprising a plurality of convolutional neural networks (CNNs) that are trained by weighting training data based on loss values of each training datum between CNNs of the CCN. The CCNN can receiving an input image from plurality of images, classify the input image using the CCNN, and present a classification of the input image.
    Type: Grant
    Filed: November 12, 2019
    Date of Patent: February 9, 2021
    Assignee: International Business Machines Corporation
    Inventors: Taro Sekiyama, Masaharu Sakamoto, Hiroki Nakano, Kun Zhao
  • Patent number: 10902590
    Abstract: Systems and methods are provided for recognizing pathological images captured by alternate image capturing devices. In embodiments, a computer-implemented method includes: obtaining, by a computing device, a classifier generated based on a plurality of standardized training pathology images associated with a known classification and generated by a first type of device; receiving, by the computing device, an alternate pathology image captured by a second type of device; standardizing, by the computing device, the alternate pathology image; determining, by the computing device, a classification of the alternate pathology image based on applying the generated classifier; and outputting, by the computing device, information regarding the determined classification to aid in diagnosis of a medical condition represented by the alternate pathology image.
    Type: Grant
    Filed: November 28, 2018
    Date of Patent: January 26, 2021
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Yusuke Takeuchi, Yoshinori Kabeya, Hiroki Nakano, Issei Ozawa, Sho Yonezawa
  • Publication number: 20200374986
    Abstract: A heater according to an embodiment includes: a tubular portion; a sealing portion provided in each of both end portions of the tubular portion; a conductive portion provided inside each sealing portion; a heating portion provided inside the tubular portion, extending along a tube axis of the tubular portion, and including carbons; an inner lead provided in each sealing portion so that one end portion side is connected to the conductive portion and the other end portion side is exposed into the tubular portion; and a connection portion connected to each of both end portions of the heating portion inside the tubular portion. A bent portion is provided in an end portion opposite to the conductive portion in each inner lead. The bent portion is bent in a direction in which the sealing portions face each other and is provided inside a hole of the connection portion.
    Type: Application
    Filed: February 27, 2020
    Publication date: November 26, 2020
    Applicant: Toshiba Lighting & Technology Corporation
    Inventor: Hiroki Nakano
  • Publication number: 20200311483
    Abstract: There is a desire to accurately learn a detection model. Provided is a computer-implemented method including acquiring an input image; acquiring an annotated image designating a region of interest in the input image; inputting the input image to a detection model that generates an output image showing a target region from the input image; calculating an error between the output image and the annotated image, using a loss function that weights an error inside the region of interest more heavily than an error outside the region of interest; and updating the detection model in a manner to reduce the error.
    Type: Application
    Filed: July 10, 2019
    Publication date: October 1, 2020
    Inventors: Takuya Goto, Hiroki Nakano, Masaharu Sakamoto
  • Publication number: 20200311479
    Abstract: There is a desire to accurately learn a detection model. Provided is a computer-implemented method including acquiring an input image; acquiring an annotated image designating a region of interest in the input image; inputting the input image to a detection model that generates an output image showing a target region from the input image; calculating an error between the output image and the annotated image, using a loss function that weights an error inside the region of interest more heavily than an error outside the region of interest; and updating the detection model in a manner to reduce the error.
    Type: Application
    Filed: March 28, 2019
    Publication date: October 1, 2020
    Inventors: Takuya Goto, Hiroki Nakano, Masaharu Sakamoto
  • Publication number: 20200271679
    Abstract: An automatic analyzer which accurately detects a liquid volume of a reagent irrespective of a shape of a reagent container is provided. The invention is directed to an automatic analyzer including: a reagent container that contains a reagent; an emission unit that is provided outside the reagent container and emits light so as to pass inside the reagent container; a light receiving unit that is provided outside the reagent container and receives the light emitted from the emission unit; and a determination unit that, based on the light received by the light receiving unit, detects a liquid level inside the reagent container, and determines whether a liquid volume in the reagent container becomes equal to or less than a predetermined value from the liquid level. A wavelength of the light is determined based on a material of the reagent container and a type of the reagent.
    Type: Application
    Filed: January 29, 2019
    Publication date: August 27, 2020
    Inventors: Hiroki NAKANO, Hidetsugu TANOUE, Yoichiro SUZUKI, Takenori OKUSA
  • Patent number: 10713783
    Abstract: Neural network classification may be performed by inputting a training data set into each of a plurality of first neural networks, the training data set including a plurality of samples, obtaining a plurality of output value sets from the plurality of first neural networks, each output value set including a plurality of output values corresponding to one of the plurality of samples, each output value being output from a corresponding first neural network in response to the inputting of one of the samples of the training data set, inputting the plurality of output value sets into a second neural network, and training the second neural network to output an expected result corresponding to each sample in response to the inputting of a corresponding output value set.
    Type: Grant
    Filed: June 1, 2017
    Date of Patent: July 14, 2020
    Assignee: International Business Machines Corporation
    Inventors: Hiroki Nakano, Masaharu Sakamoto
  • Publication number: 20200175324
    Abstract: An input image that includes a target area may be received. A first segment for extracting the target area from the input image may be generated using a first extracting model. A second segment for extracting the target area from the input image may be generated using a second extracting model. The first segment is compared to the second segment to determine a combined segment of at least the target area.
    Type: Application
    Filed: November 30, 2018
    Publication date: June 4, 2020
    Inventors: Koichi Takahashi, Hiroki Nakano, Masahiro Okawa
  • Publication number: 20200167910
    Abstract: Systems and methods are provided for recognizing pathological images captured by alternate image capturing devices. In embodiments, a computer-implemented method includes: obtaining, by a computing device, a classifier generated based on a plurality of standardized training pathology images associated with a known classification and generated by a first type of device; receiving, by the computing device, an alternate pathology image captured by a second type of device; standardizing, by the computing device, the alternate pathology image; determining, by the computing device, a classification of the alternate pathology image based on applying the generated classifier; and outputting, by the computing device, information regarding the determined classification to aid in diagnosis of a medical condition represented by the alternate pathology image.
    Type: Application
    Filed: November 28, 2018
    Publication date: May 28, 2020
    Inventors: Yusuke TAKEUCHI, Yoshinori KABEYA, Hiroki NAKANO, Issei OZAWA, Sho YONEZAWA
  • Publication number: 20200160230
    Abstract: A computer-implemented method is presented for automatically generating alerting rules. The method includes identifying, via offline analytics, abnormal patterns and normal patterns from history logs based on machine learning, statistical analysis and deep learning, the history logs stored in a history log database, automatically generating the alerting rules based on the identified abnormal and normal patterns, and transmitting the alerting rules to an alerting engine for evaluation. The method further includes receiving a plurality of online log messages from a plurality of computing devices connected to a network, augmenting the plurality of online log messages, and extracting information from the plurality of augmented online log messages to be provided to the alerting engine, the alerting engine configured to approve and enforce the alerting rules automatically generated by the offline analytics processing.
    Type: Application
    Filed: November 19, 2018
    Publication date: May 21, 2020
    Inventors: Yuan Wang, Lin Yang, Xiao Xi Liu, Fan Jing Meng, Jing Min Xu, William V. Da Palma, Sandhya Kapoor, Takayuki Kushida, Hiroki Nakano
  • Publication number: 20200143204
    Abstract: Image classification using a generated mask image is performed by generating a mask image that extracts a target area from an input image, extracting an image feature map of the input image by inputting the input image in a first neural network including at least one image feature extracting layer, masking the image feature map by using the mask image, and classifying the input image by inputting the masked image feature map to a second neural network including at least one classification layer.
    Type: Application
    Filed: November 1, 2018
    Publication date: May 7, 2020
    Inventors: Hiroki Nakano, Takuya Goto, Masaharu Sakamoto
  • Publication number: 20200110994
    Abstract: Methods and systems are provided for training a neural network with augmented data. A dataset comprising a plurality of classes is obtained for training a neural network. Prior to initiation of training, the dataset may be augmented by performing affine transformations of the data in the dataset, wherein the amount of augmentation is determined by a data augmentation variable. The neural network is trained with the augmented dataset. A training loss and a difference of class accuracy for each class is determined. The data augmentation variable is updated based on the total loss and class accuracy for each class. The dataset is augmented by performing affine transformations of the data in the dataset according to the updated data augmentation variable, and the neural network is trained with the augmented dataset.
    Type: Application
    Filed: October 4, 2018
    Publication date: April 9, 2020
    Inventors: Takuya Goto, Masaharu Sakamoto, Hiroki Nakano
  • Publication number: 20200097800
    Abstract: A cascading convolutional neural network (CCNN) comprising a plurality of convolutional neural networks (CNNs) that are trained by weighting training data based on loss values of each training datum between CNNs of the CCN. The CCNN can receiving an input image from plurality of images, classify the input image using the CCNN, and present a classification of the input image.
    Type: Application
    Filed: November 12, 2019
    Publication date: March 26, 2020
    Inventors: Taro Sekiyama, Masaharu Sakamoto, Hiroki Nakano, Kun Zhao
  • Patent number: 10599978
    Abstract: A cascading convolutional neural network (CCNN) comprising a plurality of convolutional neural networks (CNNs) that are trained by weighting training data based on loss values of each training datum between CNNs of the CCN. The CCNN can receiving an input image from plurality of images, classify the input image using the CCNN, and present a classification of the input image.
    Type: Grant
    Filed: November 3, 2017
    Date of Patent: March 24, 2020
    Assignee: International Business Machines Corporation
    Inventors: Taro Sekiyama, Masaharu Sakamoto, Hiroki Nakano, Kun Zhao
  • Patent number: 10599977
    Abstract: A method includes: training a first neural network using a first training dataset; inputting each test data of a first test dataset to the first neural network; calculating output data of the first neural network for each test data of the first test dataset; composing a second training dataset of training data from the first test dataset that causes the first neural network to output data within a first range; and training a second neural network using the second training dataset.
    Type: Grant
    Filed: August 23, 2016
    Date of Patent: March 24, 2020
    Assignee: International Business Machines Corporation
    Inventors: Hiroki Nakano, Masaharu Sakamoto