公开(公告)号：US20200311575A1

公开(公告)日：2020-10-01

申请号：US16754408

申请日：2018-08-09

申请人： Hitachi, Ltd.

发明人： Quan KONG ,  Yuki WATANABE ,  Naoto AKIRA ,  Tomokazu MURAKAMI

IPC分类号： G06N5/04 ,  G06N20/20 ,  G06K9/46

摘要： The online recognition apparatus includes a feature amount extraction unit that extracts a feature amount of input data, an identification result prediction unit that predicts an identification result based on the extracted feature amount, a prediction result evaluation unit that determines necessity of labeling from the predicted identification result unit, a correct answer assigning unit that assigns a correct answer to input data online from the determination result, a generator update unit that updates a parameter of a generator based on the input data with the correct answer, a pseudo-learning data generation unit that establishes a generator based on the parameter of the updated generator and generates pseudo-learning data, and an identifier update unit that online updates a parameter of an identifier prepared in advance based on the input data with the correct answer and the pseudo-learning data. The updated identifier is updated as a new identification result prediction unit.

公开(公告)号：US20230368508A1

公开(公告)日：2023-11-16

申请号：US18176519

申请日：2023-03-01

申请人： Hitachi, Ltd.

发明人： Ziwei DENG ,  Naoto AKIRA ,  Tomoaki YOSHINAGA

IPC分类号： G06V10/778 ,  G06V10/77 ,  G06V10/776 ,  G06V10/82 ,  G06V10/26 ,  G06V10/22 ,  G06V10/94

CPC分类号： G06V10/7788 ,  G06V10/7715 ,  G06V10/776 ,  G06V10/82 ,  G06V10/26 ,  G06V10/225 ,  G06V10/945

摘要： An area extraction method includes acquiring a captured image, generating a feature map using the captured image, extracting a candidate bounding box indicating an area in which an object exists in the captured image using the feature map, a segmentation step of generating a mask that is a bounding box indicating the area in which the object exists using the feature map and the candidate bounding box, acquiring input information which is input by a user regarding the object in the captured image, extracting the candidate bounding box using the feature map and the input information, and updating the feature map using the candidate bounding box extracted in the area estimation step and the input information. Wherein the candidate bounding box can be extracted further using the mask generated in the segmentation step, and the feature map can be updated using the mask generated in the segmentation step.

公开(公告)号：US20210239875A1

公开(公告)日：2021-08-05

申请号：US16950162

申请日：2020-11-17

申请人： Hitachi, Ltd.

发明人： Noboru OZEKI ,  Satoko HINOMIZU ,  Toru YASUMURA ,  Daisuke MATSUBARA ,  Koh KAJII ,  Naoto AKIRA ,  Junichi KIMURA ,  Hideki OKITA

IPC分类号： G01V5/00 ,  G01N23/04 ,  G06Q10/06

摘要： An alert output timing control apparatus is configured to hold an analysis target image and time information indicating a period of time time to be spent until alert is output, and output the analysis target image to a display apparatus. When the analysis target image is input to a model and it is determined that an alert target object is included in a photographic object of the analysis target image, the alert output timing control apparatus obtains a period of time time from the time information and performs control so that alert is output at timing that arrives when the obtained period of time time elapses since given timing after the analysis target image is output.

公开(公告)号：US20180060356A1

公开(公告)日：2018-03-01

申请号：US15556994

申请日：2015-03-13

申请人： Hitachi, Ltd.

发明人： Yuki WATANABE ,  Atsushi HIROIKE ,  Kenichi YONEJI ,  Nobuo NUKAGA ,  Naoto AKIRA ,  Tomoaki YOSHINAGA

IPC分类号： G06F17/30

CPC分类号： G06F16/5838 ,  G06F16/24578

摘要： An image search device searching an image from an image database storing a recorded image, one or more types of feature values, and propagation information that indicates a link between a same type of the feature values of the recorded image, the device acquires a first image and the one or more types of feature values of the first image, records the first image and the feature values in the image database, calculates a propagation intensity indicating similarity or sameness between the types of feature values of the recorded image and the types of feature values of the first image, and includes in the propagation information a link between the types of feature values of the first image and the types of feature values of the recorded image for which the propagation intensity between the types of feature values of the recorded image and the types of feature values of the first image is at a maximum value.

公开(公告)号：US20230043386A1

公开(公告)日：2023-02-09

申请号：US17792066

申请日：2020-08-28

申请人： Hitachi, Ltd

发明人： Toru YASUMURA ,  Noboru OZEKI ,  Satoko IWASAWA ,  Daisuke MATSUBARA ,  Koh KAJII ,  Naoto AKIRA ,  Junichi KIMURA ,  Hideki OKITA

IPC分类号： G01N23/10 ,  G01N23/04 ,  G01N23/083

摘要： An alert output apparatus is configured to hold: image information indicating one or more inspection objects included in a photographic object of an image for which determination is to be executed; alert target item information indicating an alert target item; and co-occurrence item information indicating alert co-occurrence items each determined in advance as an item to be included in the image at the same time as the alert target item, and to output alert when determining that at least one inspection object out of the one or more inspection objects matches to the alert target item and satisfies a given condition, and when determining that no inspection object out of the one or more inspection objects matches to the alert target item but at least one inspection object out of the one or more inspection objects matches to at least one of the alert co-occurrence items.

公开(公告)号：US20210264636A1

公开(公告)日：2021-08-26

申请号：US17253842

申请日：2019-06-11

申请人： HITACHI, LTD.

发明人： Naoto AKIRA ,  Atsushi HIROIKE ,  Tsutomu IMADA ,  Hiromu NAKAMAE

IPC分类号： G06T7/73 ,  H04N5/247 ,  G06K9/00

摘要： A video analysis apparatus is connected to an image capturing apparatus including a plurality of cameras, in which the video analysis apparatus, for a plurality of persons in images captured by the cameras, tracks at least one of the plurality of persons, detects a plurality of parts of the tracked person, and on the basis of information defining scores used to determine a best shot of each of the parts, computes a score for the part for each of frames of videos from security cameras, compares, for each of the parts, the scores for each part computed for each of the frames to determine a best shot of each of the parts, stores, for each of the parts, a feature value in association with the part, and compares the feature values of each of the parts of the plurality of persons to determine identity of the persons.