公开(公告)号：US20240362911A1

公开(公告)日：2024-10-31

申请号：US18566567

申请日：2022-06-01

申请人： HUMMINGBIRD TECHNOLOGIES LIMITED

发明人： Panagiotis SIDIROPOULOS

IPC分类号： G06V20/10 ,  G06N3/0464 ,  G06Q50/02 ,  G06T5/20 ,  G06V10/26 ,  G06V10/75 ,  G06V20/17

CPC分类号： G06V20/188 ,  G06N3/0464 ,  G06Q50/02 ,  G06T5/20 ,  G06V10/26 ,  G06V10/751 ,  G06V20/17 ,  G06T2207/20021 ,  G06T2207/20081 ,  G06T2207/30188

摘要： Aspects include methods and apparatuses generally relating to agricultural technology and artificial intelligence and, more particularly, to counting and sizing plants in a field. One aspect relates to a method for analysing plants in an area of interest that generally includes providing at least one aerial image of the area of interest; performing object detection, segmentation and instantiation using an object-mask-predicting region convolutional neural network, Mask R-CNN, wherein the Mask R-CNN is trained to detect a selected vegetable; determining numbers and sizes of objects detected; dividing the area of interest into multiple cells; calculating the average object size per cell; and displaying results in the form of a map of the area of interest with colour or scale for each cell corresponding to the average size of object in the cell

公开(公告)号：US20240362277A1

公开(公告)日：2024-10-31

申请号：US18309246

申请日：2023-04-28

申请人： Palo Alto Networks, Inc.

发明人： Nandini Ramanan ,  William Redington Hewlett, II ,  Mrunmayi Bharat Nandgaonkar ,  Anurag Mukund Phadke ,  Sreejith Rajkumar

IPC分类号： G06F16/951 ,  G06N3/0464

CPC分类号： G06F16/951 ,  G06N3/0464

摘要： An automated software-as-a-service (SaaS) security posture management (SSPM) system disclosed herein detects and maintains security posture for SaaS applications according to correct implementation of configuration settings. Based on detecting a previously unseen SaaS application with unknown implementation of configuration settings, the SSPM system scrapes the Internet for web content for the SaaS application and preprocesses/inputs the web content into a machine learning model to obtain predictions of correct/incorrect implementation of configuration settings as output. Based on the predictions not having sufficiently high confidence, the SSPM system obtains additional application content by logging into the SaaS application and scraping locally rendered pages therein. The application content is preprocessed/input to the machine learning model to obtain additional high confidence predictions.

公开(公告)号：US20240362101A1

公开(公告)日：2024-10-31

申请号：US18308474

申请日：2023-04-27

申请人： SAP SE

发明人： Tim Breitenbach ,  Patrick Jahnke

IPC分类号： G06F11/07 ,  G06N3/0464 ,  G06N3/047 ,  G06N3/08

CPC分类号： G06F11/076 ,  G06F11/073 ,  G06F11/079 ,  G06N3/0464 ,  G06N3/047 ,  G06N3/08

摘要： Embodiments of the present disclosure include techniques for predictive memory maintenance. In one embodiment, locations of correctable errors in a memory are observed. A machine learning (ML) system may be trained with patterns of correctable errors that result in uncorrectable errors. A trained ML monitors correctable errors to predict when memory requires maintenance. In another embodiment, error rates from multiple memories are monitored to predict memory channel and other upstream device failures.

公开(公告)号：US20240345680A1

公开(公告)日：2024-10-17

申请号：US18406567

申请日：2024-01-08

申请人： Samsung Electronics Co., Ltd.

发明人： Bongjun KO ,  Sanghun KWAK ,  Dongnam BYUN ,  Jongmin WI ,  Hoondo HEO ,  Jinyoung HWANG

IPC分类号： G06F3/041 ,  G06N3/0464 ,  G06N3/08

CPC分类号： G06F3/0418 ,  G06N3/0464 ,  G06N3/08 ,  G06F3/0443 ,  G06F3/0446 ,  G06F2203/04102

摘要： An electronic device is provided. The electronic device includes a display, a touch sensor disposed in the display and including a plurality of lines and a plurality of nodes formed by the plurality of lines, a memory storing instructions, and a processor. The instructions, when being executed by the processor, cause the electronic device to obtain a plurality of node values from each of the plurality of nodes, change the touch sensitivity of the touch sensor from a first touch sensitivity to a second touch sensitivity lower than the first touch sensitivity based on identifying a pattern by the plurality of node values, identify whether to recognize contact on the touch sensor as a touch input through the plurality of line values obtained from each of the plurality of lines based on identifying changes of the pattern by the plurality of node values.

公开(公告)号：US12118191B2

公开(公告)日：2024-10-15

申请号：US18401775

申请日：2024-01-02

申请人： Truist Bank

发明人： Alexis Pastore

IPC分类号： G06F3/0484 ,  G06F3/0482 ,  G06F18/21 ,  G06F18/2321 ,  G06F18/2413 ,  G06F18/243 ,  G06N3/0442 ,  G06N3/045 ,  G06N3/0464 ,  G06N3/08 ,  G06N3/084 ,  G06N3/088 ,  G06N20/00 ,  G06N20/10 ,  G06Q20/22 ,  G06Q20/24 ,  G06Q20/32 ,  G06Q20/38 ,  G06Q20/40 ,  G06Q40/03

CPC分类号： G06F3/0484 ,  G06F3/0482 ,  G06Q20/227 ,  G06Q20/24 ,  G06Q40/03 ,  G06F18/217 ,  G06F18/2321 ,  G06F18/24143 ,  G06F18/24323 ,  G06N3/0442 ,  G06N3/045 ,  G06N3/0464 ,  G06N3/08 ,  G06N3/084 ,  G06N3/088 ,  G06N20/00 ,  G06N20/10 ,  G06Q20/3221 ,  G06Q20/389 ,  G06Q20/405

摘要： A system and method for allowing a user to manage transactions in an online credit card application. The system includes a back-end server operating the online application and including a processor for processing data and information, a communications interface communicatively coupled to the processor, and a memory device storing data and executable code. When the code is executed, the processor can link one or more external bank accounts to the online application, provide a list of transactions that were made using the credit card, enable a user to select one or more of the transactions in the list to be paid independent of the other transactions, and enable the user to pay the selected transactions using the one or more external bank accounts.

公开(公告)号：US12112253B2

公开(公告)日：2024-10-08

申请号：US17795693

申请日：2022-04-02

申请人： ZHEJIANG UNIVERSITY CITY COLLEGE

发明人： Anping Wan ,  Jie Yang ,  Jiantao Yuan ,  Jinglin Wang ,  Tianmin Shan

IPC分类号： G06F30/27 ,  G05B23/02 ,  G06N3/0464 ,  G01M15/14 ,  G06N3/08

CPC分类号： G06N3/0464 ,  G05B23/0283 ,  G01M15/14 ,  G06N3/08

摘要： An aero-engine fault diagnosis method based on 5G edge computing and deep learning is provided. The method includes following steps: performing data acquisition, preprocessing, and storage based on a new 5th Generation Mobile Communication Technology (5G) cloud-edge-terminal network architecture; constructing a machine learning module in an edge cloud, where historical data stored in the aero-engine fault database management system is used as training samples of the machine learning module; and performing intelligent self-management of the aero-engine gear fault simulation platform and the aero-engine fault database management system.

公开(公告)号：US12106214B2

公开(公告)日：2024-10-01

申请号：US17968085

申请日：2022-10-18

申请人： SAMSUNG ELECTRONICS CO., LTD. ,  UNIVERSITAET ZUERICH

发明人： Stefan Braun ,  Daniel Neil ,  Enea Ceolini ,  Jithendar Anumula ,  Shih-Chii Liu

IPC分类号： G06N3/08 ,  G06F18/2413 ,  G06F18/25 ,  G06N3/04 ,  G06N3/0442 ,  G06N3/0455 ,  G06N3/0464 ,  G06N3/084 ,  G06V10/44 ,  G06V10/46 ,  G06V10/764 ,  G06V10/80 ,  G06V10/82 ,  G06V20/10 ,  G10L15/16 ,  G10L15/20 ,  G10L15/24

CPC分类号： G06N3/08 ,  G06F18/2413 ,  G06F18/256 ,  G06N3/04 ,  G06N3/0442 ,  G06N3/0455 ,  G06N3/0464 ,  G06N3/084 ,  G06V10/454 ,  G06V10/462 ,  G06V10/764 ,  G06V10/806 ,  G06V10/811 ,  G06V20/10 ,  G10L15/16 ,  G06V10/82 ,  G10L15/20 ,  G10L15/24

摘要： A sensor transformation attention network (STAN) model including sensors configured to collect input signals, attention modules configured to calculate attention scores of feature vectors corresponding to the input signals, a merge module configured to calculate attention values of the attention scores, and generate a merged transformation vector based on the attention values and the feature vectors, and a task-specific module configured to classify the merged transformation vector is provided.

公开(公告)号：US20240311621A1

公开(公告)日：2024-09-19

申请号：US18272714

申请日：2022-02-03

申请人： InterDigital CE Patent Holdings, SAS

发明人： Suresh Kirthi KUMARASWAMY ,  Quang Khanh Ngoc DUONG ,  Alexey OZEROV ,  Patrick FONTAINE ,  Francois SCHNITZLER ,  Anne LAMBERT ,  Ghyslain PELLETIER

IPC分类号： G06N3/0464 ,  G06N3/096

CPC分类号： G06N3/0464 ,  G06N3/096

摘要： The proposed approach deals with efficient transmission for distributed AI with a provision to switch among multiple bandwidths. During the distributed inference at edge devices, each device needs to load part of the AI model only once, but the input/output features communicated between them can be flexibly configured depending on the available transmission bandwidth by enabling/disabling connection between nodes in the Dynamic feature size Switch (DySw). When some nodes are connected or disconnected in order to achieve the desired compression factor, other parameters of the DNN remain the same. That is, the same DNN model is used for different compression factors, and no new DNN model needs to be downloaded to adapt to the compression factor or the network bandwidth.

公开(公告)号：US12093835B2

公开(公告)日：2024-09-17

申请号：US17971601

申请日：2022-10-23

申请人： BLINK TECHNOLOGIES INC.

发明人： Oren Haimovitch-Yogev ,  Tsahi Mizrahi ,  Andrey Zhitnikov ,  Almog David ,  Artyom Borzin ,  Gilad Drozdov

IPC分类号： G06N3/088 ,  G06F18/214 ,  G06N3/04 ,  G06N3/0455 ,  G06N3/0464 ,  G06N3/08 ,  G06N20/10 ,  G06T7/11 ,  G06V10/75 ,  G06V10/764 ,  G06V10/82 ,  G06V40/16 ,  G06V40/18 ,  G06V40/19

CPC分类号： G06N3/088 ,  G06F18/2155 ,  G06N3/04 ,  G06N3/08 ,  G06T7/11 ,  G06V10/755 ,  G06V10/82 ,  G06V40/165 ,  G06V40/171 ,  G06V40/19 ,  G06V40/193 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20132

摘要： Unsupervised, deep learning of eye-landmarks in a user-specific eyes' image data by capturing an unlabeled image comprising an eye region of a user, using an initial geometrically regularized loss function, training a plurality of convolutional autoencoders on the unlabeled image comprising the eye region of the user to recover a plurality of user-specific eye landmarks, training a convolutional neural network for autoencoded landmarks-based recovery from the unlabeled image, and where the initial geometrically regularized loss function is represented by the formula LAE=λreconLrecon+λconcLconc+λsepLsep+λeqvLeqv where LAE is total AutoEncoder Loss, λreconLrecon is λ-weighted reconstruction loss, λconcLconce is λ-weighted concentration loss, λsepLsep is λ-weighted separation loss, and λeqvLeqv is λ-weighted equivalence loss.

公开(公告)号：US20240303783A1

公开(公告)日：2024-09-12

申请号：US18595002

申请日：2024-03-04

申请人： MILESTONE SYSTEMS A/S

发明人： Andreas AAKERBERG ,  Kamal NASROLLAHI ,  Thomas B. MOESLUND

IPC分类号： G06T5/70 ,  G06N3/0464

CPC分类号： G06T5/70 ,  G06N3/0464 ,  G06T2207/20081

摘要： Training a neural network to extract a degradation map from a degraded image comprises generating training data comprising pairs of images, each pair of images comprising a clean source image and a degraded source image by, for each clean source image, generating a corresponding noisy image by adding spatially invariant noise to the clean source image, and blending the noisy image with the clean source image according to varying intensity levels defined by a spatially variant mask to obtain the degraded image. The training data is used to train the neural network by inputting each degraded source image to the neural network and extracting a degradation map from the degraded source image such that when the degradation map is applied to its corresponding clean source image the loss between the degraded source image and its corresponding clean source image after the degradation map is applied is minimised.