公开(公告)号：US20140376817A1

公开(公告)日：2014-12-25

申请号：US14479235

申请日：2014-09-05

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: G06T7/00 ,  G06K9/68

CPC classification number: G06T7/30 ,  A61B1/00009 ,  F04C2270/0421 ,  G06F19/00 ,  G06F19/321 ,  G06K9/00751 ,  G06K9/68

Abstract: An image processing device includes an image sequence acquisition section that acquires an image sequence that includes a plurality of images, and a processing section that performs an image summarization process that deletes some of the plurality of images included in the acquired image sequence to acquire a summary image sequence, the processing section selecting a reference image and a determination target image from the plurality of images, calculating a coverage ratio of the determination target image by the reference image based on deformation information about the reference image and the determination target image, and determining whether or not the determination target image can be deleted based on the coverage ratio.

Abstract translation: 一种图像处理装置，包括：图像序列获取部，其获取包含多个图像的图像序列;以及处理部，执行图像聚合处理，所述图像聚合处理删除所获取的图像序列中包含的多个图像中的一些，以获得总结 图像序列，所述处理部分从所述多个图像中选择参考图像和确定目标图像，基于关于所述参考图像和所述确定目标图像的变形信息，通过所述参考图像计算所述确定目标图像的覆盖率，以及确定 是否可以基于覆盖率来删除确定目标图像。

公开(公告)号：US20140376792A1

公开(公告)日：2014-12-25

申请号：US14480570

申请日：2014-09-08

Applicant: OLYMPUS CORPORATION

Inventor： Hiroshi MATSUZAKI ,  Yoichi YAGUCHI

IPC: G06K9/00 ,  A61B1/00 ,  A61B1/04

CPC classification number: G06T7/0012 ,  A61B1/00009 ,  A61B1/041 ,  G06K9/00751 ,  G06K9/00765 ,  G06K9/52 ,  G06K9/6215 ,  G06K9/6289 ,  G11B27/031

Abstract: An image summarization device includes a first image summarization section that performs a first image summarization process based on a similarity between a plurality of images to acquire a first summary image sequence, a second image summarization section that performs a second image summarization process based on a target object/scene recognition process on each image among the plurality of images to acquire a second summary image sequence, and an integration processing section that performs an integration process on the first summary image sequence and the second summary image sequence, or performs an integration process on the first image summarization process and the second image summarization process to acquire an output summary image sequence.

Abstract translation: 图像聚合装置包括：第一图像汇总部，其基于多个图像之间的相似度进行第一图像聚合处理，以获取第一摘要图像序列;第二图像汇总部，其基于目标进行第二图像汇总处理 多个图像中的每个图像上的对象/场景识别处理，以获取第二摘要图像序列;以及积分处理部，对第一概要图像序列和第二概要图像序列执行积分处理，或者对 第一图像汇总处理和第二图像汇总处理以获取输出汇总图像序列。

公开(公告)号：US20200349445A1

公开(公告)日：2020-11-05

申请号：US16929805

申请日：2020-07-15

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: G06N3/08 ,  G06N3/04

Abstract: A data processing system includes a learning unit that optimizes an optimization target parameter of a neural network on the basis of a comparison between output data that is output by execution of a process according to a neural network on learning data and ideal output data for the learning data. The learning unit optimizes a slope ratio parameter indicating a ratio of a slope when an input value is in a positive range and a slope when the input value is in a negative range in an activation function of the neural network, as one of optimization parameters.

公开(公告)号：US20190099060A1

公开(公告)日：2019-04-04

申请号：US16208651

申请日：2018-12-04

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: A61B1/00 ,  A61B1/317 ,  G06T7/00

CPC classification number: A61B1/00039 ,  A61B1/00009 ,  A61B1/041 ,  A61B1/317 ,  G06T7/0012

Abstract: An image processing apparatus includes a processor to: sequentially acquire plural subject images; select a detection target image from the plural subject images; detect the location of interest in the detection target image; generate spatial information of a superimposed object indicating the location of interest on one of the plural subject images; select a superimposition target image to be superimposed, from the plural subject images; generate inter-image correspondence information between the detection target image and the superimposition target image; correct the spatial information of the superimposed object, based on the inter-image correspondence information; and superimpose the superimposed object on the superimposition target image. The processor is configured to repeatedly execute the selecting the superimposition target image, the generating the inter-image correspondence information, the correcting the spatial information, and the superimposing the superimposed object, until the spatial information of the superimposed object is generated for the detection target image.

公开(公告)号：US20180342079A1

公开(公告)日：2018-11-29

申请号：US16035745

申请日：2018-07-16

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: G06T7/73 ,  G06T7/00

Abstract: An endoscope image processing apparatus comprising a processor comprising hardware. The processor executes: acquiring subject images; generating space information of a superimposed object to be superimposed on the subject image, and to be arranged so as to correspond to a location of interest detected from a detection target subject image different from a superimposition target subject image; deciding the superimposition target subject image to be displayed together with the superimposed object, according to a time at which the space information has generated; generating entire image correspondence information that estimates correspondence between the detection target subject image and the superimposition target subject image, using the entire detection target subject image; correcting space information of the superimposed object based on the information that estimates correspondence between the detection target subject image and the superimposition target subject image; and superimposing the corrected superimposed object on the decided superimposition target subject image.

公开(公告)号：US20210182679A1

公开(公告)日：2021-06-17

申请号：US17185825

申请日：2021-02-25

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: G06N3/08

Abstract: A data processing system includes: a neural network processing unit that performs a process determined by a neural network including an input layer, one or more intermediate layers, and an output layer; and a learning unit that trains the neural network by optimizing an optimization parameter of the neural network based on a comparison between output data output when the neural network processing unit subjects learning data to the process determined by the neural network and ideal output data for the learning data. The neural network processing unit performs, in a learning process, a coefficient process of multiplying intermediate data representing input data input to an intermediate layer element constituting the intermediate layer of an M-th layer (M is an integer equal to or larger than 1) or representing output data from the intermediate layer element by a coefficient the absolute value of which increases monotonically in accordance with progress of learning.

公开(公告)号：US20210182678A1

公开(公告)日：2021-06-17

申请号：US17185810

申请日：2021-02-25

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: G06N3/08 ,  G06N3/04 ,  G06K9/62

Abstract: A data processing system includes: a processor including hardware, wherein the processor performs a process determined by a neural network. An optimization parameter of the neural network is optimized based on a comparison between output data output when learning data is subject to the process and ideal output data for the learning data. The processor is configured to: output a feature map having the same width and height as the intermediate data by applying, in an M-th (M is an integer equal to or larger than 1) intermediate layer, an operation to intermediate data representing input data input to the M-th intermediate layer, the operation including a convolutional operation that uses a convolutional kernel comprised of the optimization parameter; multiply the intermediate data and the feature map mutually at each corresponding coordinate, the intermediate data being input to the M-th intermediate layer, and the feature map being output by inputting the intermediate data to the M-th intermediate layer; and execute a pooling process in an (M+1)-th intermediate layer on the intermediate data output by executing multiplication.

公开(公告)号：US20200349444A1

公开(公告)日：2020-11-05

申请号：US16929746

申请日：2020-07-15

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: G06N3/08 ,  G06N3/04

Abstract: A data processing system includes a learning unit that optimizes optimization target parameters of a neural network on the basis of a comparison between output data that is output by execution of a process according to a neural network on learning data and ideal output data for the learning data. An activation function f(x) of the neural network is defined, when a first parameter is C and a second parameter being a non-negative value is W, as a function in which an output value for an input value is a value continuous within a range of C±W, the output value for the input value is uniquely determined, and a graph of the function is point-symmetric with respect to a point corresponding to f(x)=C. The learning unit optimizes the optimization target parameters that include the first parameter and the second parameter.

公开(公告)号：US20170039709A1

公开(公告)日：2017-02-09

申请号：US15331139

申请日：2016-10-21

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI

IPC: G06T7/00 ,  A61B1/04 ,  G06K9/52 ,  G11B27/031 ,  G06K9/00 ,  G06K9/62

CPC classification number: G06T7/0012 ,  A61B1/00009 ,  A61B1/041 ,  G06K9/00751 ,  G06K9/00765 ,  G06K9/52 ,  G06K9/6215 ,  G06K9/6289 ,  G11B27/031

Abstract: An image sequence acquisition section acquires an image sequence including a plurality of images. A processing section performs an image summarization process that acquires a summary image sequence based on first and second deletion determination processes that delete some of the images included in the acquired image sequence. The processing section sets an attention image sequence including one at least one attention image included in the plurality of images, selects a first reference image from the attention image sequence, selects a first determination target image from the plurality of images, and performs the first deletion determination process that determines whether the first determination target image can be deleted based on first deformation information that represents deformation between the first reference image and the first determination target image. The processing section sets a partial image sequence from the image sequence, a plurality of images that have been determined to be allowed to remain by the first deletion determination process being consecutively arranged in the partial image sequence. The processing section selects a second reference image and a second determination target image from the partial image sequence, and performs the second deletion determination process that determines whether the second determination target image can be deleted based on second deformation information that represents deformation between the second reference image and the second determination target image.

Abstract translation: 图像序列获取部分获取包括多个图像的图像序列。 处理部分执行基于删除所获取的图像序列中包括的一些图像的第一和第二删除确定处理来获取概要图像序列的图像聚合处理。 所述处理部设置包括所述多个图像中包括的至少一个关注图像的注意图像序列，从所述注意图像序列中选择第一参考图像，从所述多个图像中选择第一确定对象图像，并执行所述第一删除 确定处理，其基于表示第一参考图像和第一确定对象图像之间的变形的第一变形信息来确定是否可以删除第一确定对象图像。 处理部分从图像序列设置部分图像序列，通过连续排列在部分图像序列中的第一删除确定处理被确定为允许保留的多个图像。 处理部分从部分图像序列中选择第二参考图像和第二确定目标图像，并且执行第二删除确定处理，该第二删除确定处理基于表示第二参考的变形的第二变形信息来确定是否可以删除第二确定目标图像 图像和第二确定目标图像。

公开(公告)号：US20150228056A1

公开(公告)日：2015-08-13

申请号：US14697272

申请日：2015-04-27

Applicant: OLYMPUS CORPORATION

Inventor： Yoichi YAGUCHI ,  Ronee ASAD

IPC: G06T3/40 ,  G06T7/00 ,  A61B1/04

CPC classification number: G06T3/4038 ,  A61B1/00009 ,  A61B1/041 ,  G06F17/3028 ,  G06F17/30843 ,  G06T7/0012 ,  G06T2207/10068 ,  G06T2207/20221 ,  G06T2207/30196 ,  G11B27/031 ,  G11B27/105 ,  G11B27/28

Abstract: An image processing device includes an image sequence acquisition section that acquires an input image sequence that includes first to N-th images, and a processing section that performs an image summarization process that deletes some of the first to Nth images to generate a summary image sequence, the processing section selecting an s-th (s is an integer that satisfies 0≦s≦N+1) image to be a provisional summary image, selecting a t-th (t is an integer that satisfies 0≦t≦s−1) image to be a provisional preceding summary image, selecting a u-th (u is an integer that satisfies t

Abstract translation: 一种图像处理装置，包括获取包括第一至第N图像的输入图像序列的图像序列获取部，以及执行删除第一至第N图像中的一些以生成概要图像序列的图像聚合处理的处理部 ，选择第t（t是满足0＆nlE; t＆nlE; N + 1）的图像的s（s是满足0＆nlE; s＆nlE; N + 1）的整数作为临时摘要图像， 1）图像作为临时前一概要图像，选择第u（u是满足t