公开(公告)号：US12062192B2

公开(公告)日：2024-08-13

申请号：US17486441

申请日：2021-09-27

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Yu Gu

IPC: G06V40/10 ,  G06T7/246 ,  G06V10/44 ,  G06V10/75 ,  G06V40/16

CPC classification number: G06T7/248 ,  G06V10/457 ,  G06V10/751 ,  G06V40/10 ,  G06V40/168 ,  G06T2207/30201 ,  G06T2207/30241

Abstract: A method, an apparatus, a computing device, and a computer-readable storage medium for correcting pedestrian trajectory are disclosed. The method includes obtaining face image frames and body image frames, determining a face identifier for a face area in each of the face image frames, determining an uncorrected face trajectory for the face identifier, determining a body identifier for a body area in each of the body image frames, determining an uncorrected body trajectory for the body identifier, for each of the face image frames and each of the body image frames that are at a same moment, establishing a set of matching relationship between the face identifier and the body identifier, and correcting at least one of the uncorrected face trajectory and the uncorrected body trajectory based on a plurality of sets of matching relationship of a plurality of same moments.

公开(公告)号：US20240212336A1

公开(公告)日：2024-06-27

申请号：US18293704

申请日：2022-07-08

Applicant: NUCTECH COMPANY LIMITED ,  TSINGHUA UNIVERSITY

Inventor： Zhiqiang CHEN ,  Li ZHANG ,  Yunda SUN ,  Juan ZHENG ,  Lu WANG ,  Tao YANG ,  Dong LI

IPC: G06V10/84 ,  G06T3/40 ,  G06T11/00 ,  G06V10/44 ,  G06V10/48

CPC classification number: G06V10/84 ,  G06T3/40 ,  G06T11/006 ,  G06V10/457 ,  G06V10/48 ,  G06T2211/428 ,  G06T2211/441 ,  G06V2201/05 ,  G06V2201/07 ,  G06V2201/12

Abstract: A method and an apparatus of identifying at least one target object for a security inspection CT are provided. The method includes: performing a dimension reduction on three-dimensional CT data to generate a plurality of two-dimensional dimension-reduced views (S10); performing a target identification on a plurality of two-dimensional views to obtain a set of two-dimensional semantic descriptions of the at least one target object, where the plurality of two-dimensional views include the plurality of two-dimensional dimension-reduced views (S20); and performing a dimension increase on the set of two-dimensional semantic descriptions to obtain a three-dimensional recognition result of the at least one target object (S30).

公开(公告)号：US12008797B2

公开(公告)日：2024-06-11

申请号：US17383181

申请日：2021-07-22

Applicant: Huawei Technologies Co., Ltd.

Inventor： Zhi Tian ,  Tong He ,  Chunhua Shen ,  Youliang Yan ,  Songcen Xu ,  Yiren Zhou ,  Xiaofei Wu ,  Jianzhuang Liu

IPC: G06V10/44 ,  G06F18/2135 ,  G06F18/2137 ,  G06F18/214 ,  G06F18/25 ,  G06N3/08 ,  G06T3/40 ,  G06V10/77 ,  G06V10/82

CPC classification number: G06V10/457 ,  G06F18/2135 ,  G06F18/2137 ,  G06F18/214 ,  G06F18/253 ,  G06N3/08 ,  G06T3/40 ,  G06V10/454 ,  G06V10/7715 ,  G06V10/82

Abstract: This application discloses an image segmentation method in the field of artificial intelligence. The method includes: obtaining an input image and a processing requirement; performing multi-layer feature extraction on the input image to obtain a plurality of feature maps; downsampling the plurality of feature maps to obtain a plurality of feature maps with a reference resolution, where the reference resolution is less than a resolution of the input image; fusing the plurality of feature maps with the reference resolution to obtain at least one feature map group; upsampling the feature map group by using a transformation matrix W, to obtain a target feature map group; and performing target processing on the target feature map group based on the processing requirement to obtain a target image.

公开(公告)号：US20240087132A1

公开(公告)日：2024-03-14

申请号：US18272885

申请日：2022-01-18

Applicant: KONINKLIJKE PHILIPS N.V.

Inventor： ZHANGPENG HAO

IPC: G06T7/12 ,  G06T7/11 ,  G06T7/136 ,  G06T7/50 ,  G06T19/00 ,  G06V10/44 ,  G06V10/762 ,  G06V10/764

CPC classification number: G06T7/12 ,  G06T7/11 ,  G06T7/136 ,  G06T7/50 ,  G06T19/00 ,  G06V10/457 ,  G06V10/762 ,  G06V10/764 ,  G06T2200/24 ,  G06T2207/10116 ,  G06T2207/30101 ,  G06T2210/41 ,  G06V2201/03

Abstract: In an embodiment, a method (100) is described. The method comprises estimating (102) a pair of threshold values for being compared to radiographic imaging values to classify a region derived from radiographic imaging data of a subject based on whether or not a part of the region comprises a radiographic imaging value indicative of presence of a vessel. The method further comprises identifying (104), within a cross-section of the region, a boundary of the vessel based on a set of radiographic imaging values classified according to the pair of threshold values. The method further comprises determining (106) a segment shape of the vessel that connects the identified boundary of the vessel with another boundary of the vessel identified in a cross section of a further region adjacent to the region.

公开(公告)号：US20240078822A1

公开(公告)日：2024-03-07

申请号：US18388996

申请日：2023-11-13

Applicant: Panasonic Intellectual Property Management Co., Ltd.

Inventor： Shoichi ARAKI

IPC: G06V20/60 ,  G06V10/44 ,  G06V10/764 ,  G06V10/774

CPC classification number: G06V20/60 ,  G06V10/457 ,  G06V10/764 ,  G06V10/774

Abstract: An image identification device includes a first determiner that determines whether each of one or more areas in an image includes a joint between two first materials, and an identifier that identifies the type of a first material by using a first discriminative model and outputs type information indicating the type of the first material, the type of the first material being identified by the first discriminative model receiving input on an area that has been determined by the first determiner as an area without the joint between the two first materials, out of the one or more areas in the image, the first discriminative model being created on the basis of first learning images that are images of different types of materials including the first material, the first learning images being images without a joint between two first materials.

公开(公告)号：US20240046611A1

公开(公告)日：2024-02-08

申请号：US18106105

申请日：2023-02-06

Applicant: Seetrue Screening Ltd.

Inventor： Koren BEISER ,  Yuval AMSTERDAM ,  Assaf FRENKEL

IPC: G06V10/74 ,  G06V10/44 ,  G06T7/00 ,  G06T7/11

CPC classification number: G06V10/761 ,  G06V10/457 ,  G06T7/0004 ,  G06T7/11 ,  G06T2207/30112 ,  G06T2207/20076

Abstract: There are provided systems and methods comprising obtaining an image of at least a part of an item, obtaining data informative of a pixel intensity of the part of the item in the image, obtaining first data informative of a pixel intensity in an image of a part of a first item associated with a prohibited material, wherein the part of the first item meets a similarity criterion with the part of the item, obtaining second data informative of a pixel intensity in an image of a part of a second item which is not associated with a prohibited material, wherein the part of the second item meets the similarity criterion with the part of the item, and using the first data, the second data and the data to determine whether the part of the item is associated with a concealed prohibited material in the image.

公开(公告)号：US11893768B2

公开(公告)日：2024-02-06

申请号：US17422720

申请日：2020-08-18

Applicant: CENTRAL SOUTH UNIVERSITY

Inventor： Zhonggang Wang ,  Xifeng Liang ,  Chong Shi ,  Wei Zhou ,  Can Cui ,  Wei Xiong ,  Xinxin Wang

IPC: G06V10/40 ,  G06V10/44 ,  H04N23/695 ,  G06T7/00 ,  G06T7/60

CPC classification number: G06V10/40 ,  G06T7/001 ,  G06T7/60 ,  G06V10/457 ,  H04N23/695 ,  G06T2207/20036

Abstract: The present invention discloses a method and system for recognizing a geometric regularity image of a honeycomb structure. The method includes the steps of image acquisition, image processing, vertex extraction, cell reconstruction, and quality evaluation, wherein a step of binaryzation is set between the step of image processing and the step of vertex extraction, and is to set a pixel value of a background in the image to be 0 and set a pixel value of a honeycomb skeleton in the image to be 1 to form a binary image, and the step of quality evaluation is to calculate angular deviation values of all the cells and an average thereof as well as linear deviation values and an average thereof based on the reconstructed cell image, and determine whether the honeycomb structure is qualified or not by comparing with a set tolerance zone.

公开(公告)号：US11756211B2

公开(公告)日：2023-09-12

申请号：US17411386

申请日：2021-08-25

Applicant: 7-ELEVEN, INC.

Inventor： Sailesh Bharathwaaj Krishnamurthy ,  Shahmeer Ali Mirza

IPC: G06T7/246 ,  G06T7/292 ,  G06T7/223 ,  G06T7/215 ,  G06V20/52 ,  G06V10/25 ,  G06V10/44 ,  G06V10/147 ,  G06V20/64 ,  G06V10/24

CPC classification number: G06T7/246 ,  G06T7/215 ,  G06T7/223 ,  G06T7/292 ,  G06V10/147 ,  G06V10/25 ,  G06V10/457 ,  G06V20/52 ,  G06V20/653 ,  G06T2207/10021 ,  G06T2207/30232 ,  G06V10/245

Abstract: An object tracking system includes a first sensor, a second sensor, and a tracking system. The first sensor is configured to capture a first frame of a global plane for at least a first portion of a space. The second sensor is configured to capture a second frame of at least a second portion of the space. The tracking system is configured to determine the object is within an overlap region with the second sensor based on a first pixel location. The tracking system is further configured to determine a first coordinate in the global plane for the object, to determine a second pixel location in the second frame for the object based on the first coordinate, and to store the second pixel location with an object identifier a tracking list associated with the second sensor.

公开(公告)号：US20230260246A1

公开(公告)日：2023-08-17

申请号：US18137660

申请日：2023-04-21

Applicant: Mitsubishi Electric Corporation

Inventor： Shunya OSAWA ,  Ryota SEKIGUCHI

IPC: G06V10/44 ,  G06V10/26 ,  G06V10/20 ,  G06V20/50

CPC classification number: G06V10/457 ,  G06V10/267 ,  G06V10/255 ,  G06V20/50

Abstract: A rail detecting device includes: a starting-region rail-starting-point redesignating unit that shifts a starting region for detecting rails to set a modified starting region and designates modified starting points of two rails at the lower edge of the modified starting region when portions of the two rails are not detected based on rail segments in an image; a rail-candidate detecting unit that detects multiple lines in the modified starting region as multiple modified rail candidates, the lines being candidates of edges of the two rails; a starting-region-rail assigning unit that assigns modified rail candidates whose lowest ends are closest to the corresponding modified starting points out of the multiple modified rail candidates as tentative modified rail segments; and a starting-region-rail disappearance determining unit that establishes the tentative modified rail segments as modified rail segments when distances between the lowest ends of the tentative modified rail segments.

公开(公告)号：US20230222788A1

公开(公告)日：2023-07-13

申请号：US17574458

申请日：2022-01-12

Applicant: WOVEN ALPHA, INC.

Inventor： José Felix RODRIGUES

IPC: G06V20/10 ,  G06V20/13 ,  G06V10/44 ,  G06V10/82 ,  G06T7/60 ,  G06V10/26 ,  G06V10/80 ,  G01C21/00

CPC classification number: G06V20/182 ,  G01C21/3852 ,  G06T7/60 ,  G06V10/80 ,  G06V10/82 ,  G06V10/267 ,  G06V10/457 ,  G06V20/13 ,  G06T2207/10032 ,  G06T2207/20044 ,  G06T2207/30181 ,  G06T2207/30242

Abstract: A method of determining a roadway map includes receiving an image from above a roadway. The method further includes generating a skeletonized map based on the received image, wherein the skeletonized map comprises a plurality of roads. The method includes identifying intersections based on joining of multiple roads of the plurality of roads in the skeletonized map. The method includes partitioning the skeletonized map based on the identified intersections, wherein partitioning the skeletonized map defines a roadway data set and an intersection data set. The method includes analyzing the roadway data set to determine a number of lanes in each roadway of the plurality of roads. The method further includes analyzing the intersection data set to lane connections in the identified intersections. The method further includes merging results of the analyzed road data set and the analyzed intersection data set to generate the roadway map.