公开(公告)号：US12118736B2

公开(公告)日：2024-10-15

申请号：US17836316

申请日：2022-06-09

申请人： Swallow Incubate Co., Ltd. ,  Panasonic Holdings Corporation

发明人： Toshikazu Ohno

IPC分类号： G06V40/18 ,  G06T7/536 ,  G06T7/60 ,  G06T7/73 ,  G06V10/28

CPC分类号： G06T7/536 ,  G06T7/60 ,  G06T7/73 ,  G06V10/28 ,  G06V40/18 ,  G06T2207/30201

摘要： A viewing distance estimation method includes: acquiring a first image captured by an image capturing device and including a face of a person who watches a target; detecting a size of an iris of the person from the first image; calculating a first value indicating a pixel number for the detected size of the iris; acquiring a resolution of the first image; calculating, based on the first value and a second value indicating a predetermined inherent dimension for the size of the iris, a third value indicating an actual dimension of one pixel; estimating a viewing distance corresponding to the acquired resolution and the calculated third value, based on relational information representing a relation among the resolution, the third value, and the viewing distance; and outputting estimative information including the estimated viewing distance.

公开(公告)号：US12112516B2

公开(公告)日：2024-10-08

申请号：US17498566

申请日：2021-10-11

申请人： ADLINK Technology Inc.

发明人： Chien-Chung Lin ,  Wei-Jyun Tu ,  Yu-Yen Chen

IPC分类号： G06V10/44 ,  G06T5/30 ,  G06T7/13 ,  G06T7/62 ,  G06T7/73 ,  G06V10/28 ,  G06V10/46 ,  G06V10/75 ,  G06V30/10

CPC分类号： G06V10/44 ,  G06T5/30 ,  G06T7/13 ,  G06T7/62 ,  G06T7/73 ,  G06V10/28 ,  G06V10/471 ,  G06V10/752 ,  G06T2207/10016 ,  G06T2207/20044 ,  G06T2207/20164 ,  G06T2207/20224 ,  G06V30/10 ,  G06V2201/02

摘要： A non-intrusive detection method for detecting at least one pop-up window button of the pop-up window includes the following steps: retrieving a screen image on a display device; comparing the screen image with a preset screen image and generating a differential image area according the screen image and the preset screen image; determining the differential image area as the pop-up window when the differential image area is greater than an image area threshold value; selecting a plurality of contour lengths of the pop-up window matching up with a contour length threshold value by Canny edge detector; and analyzing the contour lengths according to Douglas-Peucker algorithm and an amount of endpoints to generate a contour edge corresponding to the pop-up window button.

公开(公告)号：US12094124B2

公开(公告)日：2024-09-17

申请号：US17062092

申请日：2020-10-02

申请人： Robert Bosch GmbH

发明人： Ze Guo

IPC分类号： G06T7/12 ,  B25J9/16 ,  G06F18/21 ,  G06F18/2431 ,  G06N3/08 ,  G06V10/28 ,  G06V10/34 ,  G06V10/44 ,  G06V10/776 ,  G06V20/10 ,  G06V20/56

CPC分类号： G06T7/12 ,  B25J9/163 ,  B25J9/1697 ,  G06F18/2163 ,  G06F18/217 ,  G06F18/2431 ,  G06N3/08 ,  G06V10/28 ,  G06V10/34 ,  G06V10/44 ,  G06V10/776 ,  G06V20/10 ,  G06V20/56

摘要： A method for evaluating the boundary performance of a semantic segmentation network in a convolutional neural network. A method for measuring the boundary performance of a semantic segmentation network includes the steps: determining an original boundary mask defining a boundary between different classes of an image; determining a thickened boundary mask dependent on the original boundary mask using morphological operations; determining a true positive rate, a false positive rate and a false negative rate dependent on the original boundary mask and/or the thickened boundary mask; and determining a final boundary metric, indicating the boundary performance, dependent on the determined true positive rate, the determined false positive rate and the determined false negative rate.

公开(公告)号：US20240273927A1

公开(公告)日：2024-08-15

申请号：US18642080

申请日：2024-04-22

申请人： PAIGE.AI, Inc.

发明人： Brandon ROTHROCK ,  Christopher KANAN ,  Julian VIRET ,  Thomas FUCHS ,  Leo GRADY

IPC分类号： G06V20/69 ,  G06F18/214 ,  G06N20/00 ,  G06T7/00 ,  G06T7/136 ,  G06T7/194 ,  G06V10/26 ,  G06V10/28

CPC分类号： G06V20/695 ,  G06F18/2155 ,  G06N20/00 ,  G06T7/0012 ,  G06T7/136 ,  G06T7/194 ,  G06V10/26 ,  G06V10/28 ,  G06V20/698 ,  G06T2207/20081 ,  G06T2207/30024

摘要： Systems and methods are disclosed for receiving one or more electronic slide images associated with a tissue specimen, the tissue specimen being associated with a patient and/or medical case, partitioning a first slide image of the one or more electronic slide images into a plurality of tiles, detecting a plurality of tissue regions of the first slide image and/or plurality of tiles to generate a tissue mask, determining whether any of the plurality of tiles corresponds to non-tissue, removing any of the plurality of tiles that are determined to be non-tissue, determining a prediction, using a machine learning prediction model, for at least one label for the one or more electronic slide images, the machine learning prediction model having been generated by processing a plurality of training images, and outputting the prediction of the trained machine learning prediction model.

公开(公告)号：US20240273869A1

公开(公告)日：2024-08-15

申请号：US18569997

申请日：2022-06-30

申请人： Agency for Science, Technology and Research

发明人： Jian Wei Jayce CHENG ,  Pawan KUMAR ,  Anas ABUTAHA ,  Daniil BASH ,  Tonio BUONASSISI ,  Kedar HIPPALGAONKAR

IPC分类号： G06V10/764 ,  G01N35/00 ,  G06V10/12 ,  G06V10/24 ,  G06V10/28 ,  G06V10/44 ,  G06V20/70

CPC分类号： G06V10/764 ,  G01N35/00732 ,  G06V10/12 ,  G06V10/24 ,  G06V10/28 ,  G06V10/44 ,  G06V20/70

摘要： A data collection apparatus (10) and a computer-implemented data collection method (50) using the same are provided. The data collection apparatus (10) includes a first linear stage (12) and a second linear stage (14). A sample carriage (16) is attached to the first linear stage (12), the first linear stage (12) being operable to move the sample carriage (16) along a first axis. A probe carriage (18) is attached to the second linear stage (14), the second linear stage (14) being operable to move the probe carriage (18) along a second axis. A third linear stage (20) is attached to the probe carriage (18). In use, the third linear stage (20) is operable to receive a detachable characterisation probe (22) and to move the characterisation probe (22) along a third axis. A camera (24) is attached to the probe carriage (18). In use, the camera (24) is configured to capture an image of one or more samples on the sample carriage (16).

公开(公告)号：US20240265662A1

公开(公告)日：2024-08-08

申请号：US18569438

申请日：2021-06-25

申请人： FUJI CORPORATION

发明人： Minoru KINUTA ,  Toshinori SHIMIZU

IPC分类号： G06V10/14 ,  G06V10/28 ,  G06V10/98 ,  H04N23/73 ,  H05K13/08

CPC分类号： G06V10/14 ,  G06V10/28 ,  G06V10/993 ,  H04N23/73 ,  H05K13/081 ,  G06V2201/06

摘要： An image analysis section executes instruction processing to image electronic component while changing a shutter speed of measurement camera between a predetermined lower limit speed and a predetermined upper limit speed calculation processing of comparing an image of lower surface of each lead included in each of multiple imaging data acquired from the measurement camera with an ideal image of lower surface of each lead included in ideal imaging data to calculate a missing rate at which the image of lower surface of each lead is missing with respect to the ideal image for each of the multiple imaging data, detection processing of detecting imaging data having a missing rate equal to or less than a predetermined first threshold among multiple missing rates calculated by the calculation processing, and selection processing of selecting a shutter speed used for capturing the imaging data detected by the detection processing.

公开(公告)号：US12051229B2

公开(公告)日：2024-07-30

申请号：US17574856

申请日：2022-01-13

申请人： Tata Consultancy Services Limited

发明人： Prakhar Pradhan ,  Hrishikesh Sharma ,  Balamuralidhar Purushothaman

IPC分类号： G06V10/28 ,  G06F18/2413 ,  G06T7/00 ,  G06T7/13 ,  G06V10/26 ,  G06V10/44 ,  G06V10/77 ,  G06V10/80 ,  G06V10/82

CPC分类号： G06V10/26 ,  G06V10/28 ,  G06V10/44 ,  G06V10/7715 ,  G06V10/806 ,  G06V10/82

摘要： This disclosure relates to a system and method for attention-based surface crack segmentation. Existing methods do not efficiently handle the sub-problem of data imbalance and inaccurate predicted pixels are ignored. The present disclosure obtains a binary edge map by passing a m-channel image through an edge detection algorithm and concatenate the obtained binary edge map with a channel dimension to obtain a (m+1)-channel image. Feature maps are extracted from an encoder and a decoder by feeding the obtained (m+1)-channel image into a network, wherein the feature maps are convolved with an attention mask and merged in a fused network. The merged feature maps are up sampled and concatenated to obtain a final fused feature map. The final fused feature map is passed through a sigmoid activation function to obtain a probability map which is iteratively thresholded to obtain a binary predicted image. The binary image is indicative of crack pixels.

公开(公告)号：US20240221108A1

公开(公告)日：2024-07-04

申请号：US18608091

申请日：2024-03-18

申请人： eBay Inc.

发明人： Robinson Piramuthu ,  Wen Zheng

IPC分类号： G06T1/00 ,  G06F18/00 ,  G06Q30/0601 ,  G06T11/60 ,  G06V10/22 ,  G06V10/28 ,  G06V20/20 ,  G06V20/64 ,  H04N5/222 ,  H04N23/60 ,  H04N23/63 ,  H04N23/90

CPC分类号： G06T1/0007 ,  G06F18/00 ,  G06Q30/0643 ,  G06T11/60 ,  G06V10/235 ,  G06V10/28 ,  G06V20/20 ,  G06V20/64 ,  H04N5/2226 ,  H04N23/631 ,  H04N23/64 ,  H04N23/90

摘要： Disclosed are embodiments for facilitating automatic-guided image capturing and presentation. In some embodiments, the method includes capturing an image of an item, removing a background of the image frame, performing manual mask editing, generating an item listing based on the manually edited mask, inferring item information from the image frame and applying the inferred item information to an item listing.

公开(公告)号：US20240203134A1

公开(公告)日：2024-06-20

申请号：US18125680

申请日：2023-03-23

申请人： HON HAI PRECISION INDUSTRY CO., LTD.

发明人： CHIEH LEE ,  CHIN-PIN KUO

IPC分类号： G06V20/56 ,  G06T7/70 ,  G06V10/28

CPC分类号： G06V20/588 ,  G06T7/70 ,  G06V10/28 ,  G06T2207/20084 ,  G06T2207/30256

摘要： A lane line detection method applied to a vehicle-mounted device is provided. In the method, the vehicle-mounted device converts a foreground image into a bird's-eye view (BEV) image, sets a peak of a horizontal histogram of the BEV image as a starting point for moving the sliding window, and generate a main lane line by fitting multiple non-zero pixels of the horizontal sliding window. The vehicle-mounted device calculates a confidence coefficient of each horizontal sliding window; and sets a previous one before the preset number of continuous horizontal sliding windows as a previous horizontal sliding window and determines an end point of the main lane line according to the previous horizontal sliding window when the confidence coefficients of preset number of continuous horizontal sliding windows are less than a predetermined threshold. The method improves accuracy of identifying lane lines.

公开(公告)号：US11983944B2

公开(公告)日：2024-05-14

申请号：US18114876

申请日：2023-02-27

申请人： Kofax, Inc.

发明人： Jiyong Ma ,  Stephen M. Thompson ,  Jan W. Amtrup

IPC分类号： G06V30/146 ,  G06F18/20 ,  G06F18/21 ,  G06N7/00 ,  G06N20/00 ,  G06T7/00 ,  G06T7/11 ,  G06T7/12 ,  G06T7/13 ,  G06T7/194 ,  G06T7/70 ,  G06T7/90 ,  G06V10/44 ,  G06V10/762 ,  G06V10/80 ,  G06V20/40 ,  G06V30/162 ,  G06V30/18 ,  G06V30/19 ,  G06V30/413 ,  G06V10/28 ,  G06V30/10 ,  G06V30/224

CPC分类号： G06V30/147 ,  G06F18/217 ,  G06F18/285 ,  G06N7/00 ,  G06N20/00 ,  G06T7/0002 ,  G06T7/11 ,  G06T7/12 ,  G06T7/13 ,  G06T7/194 ,  G06T7/70 ,  G06T7/90 ,  G06V10/44 ,  G06V10/762 ,  G06V10/809 ,  G06V20/40 ,  G06V30/162 ,  G06V30/18105 ,  G06V30/19147 ,  G06V30/19173 ,  G06V30/413 ,  G06T2207/10016 ,  G06T2207/10024 ,  G06T2207/20081 ,  G06T2207/30176 ,  G06V10/28 ,  G06V30/10 ,  G06V30/2253

摘要： Computer-implemented methods for detecting objects within digital image data based on color transitions include: receiving or capturing a digital image depicting an object; sampling color information from a first plurality of pixels of the digital image, wherein each of the first plurality of pixels is located in a background region of the digital image; assigning each pixel a label of either foreground or background using an adaptive label learning process; binarizing the digital image based on the labels assigned to each pixel; detecting contour(s) within the binarized digital image; and defining edge(s) of the object based on the detected contour(s). Corresponding systems and computer program products configured to perform the inventive methods are also described.