公开(公告)号：US12288392B2

公开(公告)日：2025-04-29

申请号：US17797034

申请日：2021-05-31

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Rui Zheng ,  Xiaojun Tang ,  Zhanfu An ,  Guangwei Huang

IPC: G06V10/82 ,  G06V10/77 ,  G06V10/80

Abstract: A method for training an object detection model includes: firstly obtaining M sample image sets; then, obtaining an initial object detection model; and finally, training the initial object detection model by using the M sample image sets to obtain the object detection model. A sample image set includes at least one sample image and object type(s) of object(s) in each sample image. An object type corresponds to one sample image set, and the M sample image sets correspond to N object types.

公开(公告)号：US11482013B2

公开(公告)日：2022-10-25

申请号：US16616962

申请日：2018-05-23

Applicant: BOE Technology Group Co., Ltd.

Inventor： Hongli Ding ,  Xiaojun Tang

IPC: G06V20/58 ,  G06T7/13 ,  B60R11/04 ,  G06K9/62 ,  G06V20/56

Abstract: The present application provides an object tracking method. The object tracking method includes obtaining an image of an area in front of a vehicle; dividing the image of the area in the front of the vehicle into a plurality of sub-images; determining a plurality of first sub-images that satisfy a plurality of threshold conditions out of the plurality of sub-images; selecting a plurality of target sub-images out of the plurality of first sub-images, at least one of the plurality of first sub-images is not selected as one of the plurality of target sub-images; and recognizing a target object in one of the plurality of target sub-images.

公开(公告)号：US11238296B2

公开(公告)日：2022-02-01

申请号：US16605752

申请日：2019-02-03

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Xiaojun Tang

IPC: G06K9/20 ,  G06K9/62 ,  G06K9/00

Abstract: The present disclosure discloses a sample acquisition method, a target detection model generation method, a target detection method, a computing device, and a computer readable medium. The sample acquisition method includes: adding a perturbation to a pre-marked sample original box in an original image to obtain a sample selection box, wherein an image framed by the sample original box contains a target; and extracting an image framed by the sample selection box as a sample. The technical solutions of the present disclosure can effectively increase the number of the samples that can be acquired in the original image, and adding a background to the samples can effectively improve the recognition accuracy of the trained target detection model.

公开(公告)号：US10311295B2

公开(公告)日：2019-06-04

申请号：US15553710

申请日：2017-03-17

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Xiaojun Tang ,  Jibo Zhao

IPC: G06K9/00 ,  G06T1/60 ,  G06T7/11 ,  G06T7/529 ,  G06T7/543

Abstract: A heuristic finger detection method based on a depth image is disclosed. The method includes the steps of: acquiring a hand connected region from a user's depth image; calculating the central point of the hand connected region; calculating a plurality of extremely far points in the hand connected region that have extremum 3D geodesic distances from the central point; detecting fingertips and finger regions from the plurality of calculated extremely far points; and outputting fingertip positions and the finger regions. The method calculates and detects fingertips of users by means of 3D geodesic distance, without extracting boundary contours of hand regions, which improves robustness of gesture detection and reduces detection error rates. The method has the advantages of higher finger detection accuracy and fast computing speed.

公开(公告)号：US11748702B2

公开(公告)日：2023-09-05

申请号：US17209493

申请日：2021-03-23

Applicant: BOE Technology Group Co., Ltd.

Inventor： Tong Liu ,  Xiaojun Tang

IPC: G06Q10/087 ,  G06T7/254

CPC classification number: G06Q10/087 ,  G06T7/254

Abstract: The present disclosure provides a method of determining a status of a position on a shelf, a shelf and a non-transitory computer-readable storage medium. The method includes: acquiring current detection data of the shelf in a current detection period; determining first status information of respective positions from the current detection data, in response to determining that the current detection data satisfies a preset condition; in response to the current detection data indicating that a target position whose first status information indicates the second status exists, acquiring first status information of the target position from previous detection data in a previous detection period; and determining second status information of the target position based on the first status information of the target position in the current detection data and the previous detection data respectively.

公开(公告)号：US11714921B2

公开(公告)日：2023-08-01

申请号：US16916872

申请日：2020-06-30

Applicant: BOE TECHNOLOGY GROUP CO., LTD. ,  PEKING UNIVERSITY

Inventor： Yigeng Fang ,  Xiaojun Tang ,  Yadong Mu

IPC: H04L9/32 ,  G06F21/64 ,  G06N3/08 ,  H04L9/06 ,  G06V10/764 ,  G06V10/82 ,  G06V10/44

CPC classification number: G06F21/64 ,  G06N3/08 ,  G06V10/454 ,  G06V10/764 ,  G06V10/82 ,  H04L9/0643 ,  H04L9/3247

Abstract: Provided are an image processing method, an image matching method, a device, and a storage medium. The image processing method includes: obtaining an image feature of an input image; determining a plurality of local image features of the image feature; determining a plurality of local feature vectors corresponding to the plurality of local image features respectively; determining a hash code of the input image based on the plurality of local feature vectors.

公开(公告)号：US11379988B2

公开(公告)日：2022-07-05

申请号：US16615644

申请日：2019-04-25

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Tong Liu ,  Xiangyang Liu ,  Xiaojun Tang

IPC: G06T7/13 ,  G06T7/55 ,  G06T7/12 ,  G06T5/00

Abstract: The embodiments of the present disclosure disclose a method and apparatus for extracting an edge of an object in a depth image, a storage medium, and a computer device, to provide a solution which is simple to implement and accurate in edge extraction. The method for extracting an edge of an object in a depth image includes: calculating at least two sets of edge pixels of a first depth image using at least two edge detection algorithms; and acquiring a final set of edge pixels by combining the at least two sets of edge pixels. The solution according to the embodiments of the present disclosure is simple to implement, and may obtain a more accurate set of edge pixels, thereby obtaining an accurate pixel-level segmentation result.

公开(公告)号：US11244418B2

公开(公告)日：2022-02-08

申请号：US16626564

申请日：2019-03-15

Applicant: BOE Technology Group Co., Ltd.

Inventor： Xiaojun Tang

IPC: G06T1/00

Abstract: A method of encoding a watermark into a digital image is provided. The method includes partitioning an image into a plurality of blocks of a same size; accumulating the plurality of blocks of the same size into a single block image; performing a Fourier transformation on the single block image to obtain a two-dimensional Fourier spectrum defined by Fourier coefficients at different positions of a Fourier domain; inserting a watermark into a frequency domain of the two-dimensional Fourier spectrum by modifying the two-dimensional Fourier spectrum as a function of watermarking coefficients in the watermark, to obtain a modified Fourier spectrum; performing an inverse Fourier transformation on the modified Fourier spectrum to obtain a watermarked image; copying the watermarked image horizontally and vertically into a plurality of copied watermarked images; and splicing the plurality of copied watermarked images into a reconstituted watermark image.

公开(公告)号：US20210256291A1

公开(公告)日：2021-08-19

申请号：US16960052

申请日：2019-05-09

Applicant: BOE Technology Group Co., Ltd.

Inventor： Huanhuan Zhang ,  Jingtao Xu ,  Xiaojun Tang ,  Hui Li

IPC: G06K9/46 ,  G06K9/32 ,  G06T7/60 ,  G06T7/90 ,  G06T7/11 ,  G06K9/62

Abstract: A computer-implemented method of detecting a foreign object on a background object in an image is provided. The computer-implemented method includes extracting image features of the image based on image characteristics of the background object and a suspected foreign object; detecting a salient region in the image based on a human visual attention model; generating a salient region mask based on detection of the salient region; obtaining the image features in a region surrounded by the salient region mask; perforating feature combination and feature vector length normalization on the image features of the region surrounded by the salient region mask to generate normalized feature vectors; and identifying the suspected foreign object using an image classifier, wherein the image classifier uses the normalized feature vectors as input for identifying the suspected foreign object.

公开(公告)号：US10832445B2

公开(公告)日：2020-11-10

申请号：US15966345

申请日：2018-04-30

Applicant: BOE TECHNOLOGY GROUP CO., LTD.

Inventor： Yu Gu ,  Xiaojun Tang

IPC: G06K9/00 ,  G06T7/80 ,  G06T7/70 ,  G06T7/20

Abstract: Described herein are a method, apparatus, terminal, and system for measuring a trajectory tracking accuracy of a target. Using each method, apparatus, terminal, and system to measure the trajectory tracking accuracy of the target includes determining a location information of the actual tracking trajectory of the target; comparing the location information of the actual tracking trajectory with a location information of the target trajectory to determine a variance between the location information of the actual tracking trajectory and the location information of the target trajectory; and determining the tracking accuracy of the target based on the variance.