CA3137030C - Procede et dispositif de traitement pour entrainer un reseau neuronal - Google Patents

Procede et dispositif de traitement pour entrainer un reseau neuronal

Info

Publication number
CA3137030C
CA3137030C CA3137030A CA3137030A CA3137030C CA 3137030 C CA3137030 C CA 3137030C CA 3137030 A CA3137030 A CA 3137030A CA 3137030 A CA3137030 A CA 3137030A CA 3137030 C CA3137030 C CA 3137030C
Authority
CA
Canada
Prior art keywords
neural network
segmented
uncertainty
map
training
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CA3137030A
Other languages
English (en)
Other versions
CA3137030A1 (fr
Inventor
Maryam Ziaeefard
Simon CORBEIL-LETOURNEAU
David Beach
Freddy LECUE
Florian Martet
Original Assignee
Thales Canada Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Thales Canada Inc filed Critical Thales Canada Inc
Priority to CA3242569A priority Critical patent/CA3242569A1/fr
Publication of CA3137030A1 publication Critical patent/CA3137030A1/fr
Application granted granted Critical
Publication of CA3137030C publication Critical patent/CA3137030C/fr
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/09—Supervised learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Image Analysis (AREA)

Abstract

L'invention concerne un procédé et un dispositif de traitement pour entraîner un réseau neuronal consistant à obtenir un réseau neuronal à entraîner, à générer un ensemble de données d'apprentissage, la génération consistant à obtenir un ensemble de données segmenté comprenant une pluralité de données multimodales, à fournir une carte d'incertitude pour chaque donnée multimodale segmentée de l'ensemble de données segmenté, la carte d'incertitude fournissant une indication d'une performance d'une segmentation correspondante, et à combiner chaque donnée multimodale segmentée de l'ensemble de données segmenté avec une carte d'incertitude correspondante de façon à fournir l'ensemble de données d'apprentissage, l'ensemble de données d'apprentissage comprenant une pluralité de données multimodales, chaque donnée multimodale étant segmentée à l'aide de la carte d'incertitude, à entraîner le réseau neuronal à l'aide de l'ensemble de données d'apprentissage et à fournir le réseau neuronal entraîné.
CA3137030A 2019-05-31 2020-05-28 Procede et dispositif de traitement pour entrainer un reseau neuronal Active CA3137030C (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CA3242569A CA3242569A1 (fr) 2019-05-31 2020-05-28 Procede et dispositif de traitement pour entrainer un reseau neuronal

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US201962855340P 2019-05-31 2019-05-31
US62/855,340 2019-05-31
PCT/IB2020/055088 WO2020240477A1 (fr) 2019-05-31 2020-05-28 Procédé et dispositif de traitement pour entraîner un réseau neuronal

Related Child Applications (1)

Application Number Title Priority Date Filing Date
CA3242569A Division CA3242569A1 (fr) 2019-05-31 2020-05-28 Procede et dispositif de traitement pour entrainer un reseau neuronal

Publications (2)

Publication Number Publication Date
CA3137030A1 CA3137030A1 (fr) 2020-12-03
CA3137030C true CA3137030C (fr) 2026-03-24

Family

ID=73553098

Family Applications (2)

Application Number Title Priority Date Filing Date
CA3137030A Active CA3137030C (fr) 2019-05-31 2020-05-28 Procede et dispositif de traitement pour entrainer un reseau neuronal
CA3242569A Pending CA3242569A1 (fr) 2019-05-31 2020-05-28 Procede et dispositif de traitement pour entrainer un reseau neuronal

Family Applications After (1)

Application Number Title Priority Date Filing Date
CA3242569A Pending CA3242569A1 (fr) 2019-05-31 2020-05-28 Procede et dispositif de traitement pour entrainer un reseau neuronal

Country Status (3)

Country Link
EP (1) EP3977364A4 (fr)
CA (2) CA3137030C (fr)
WO (1) WO2020240477A1 (fr)

Families Citing this family (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112528810A (zh) * 2020-12-04 2021-03-19 北京中科慧眼科技有限公司 适用于移动端部署的语义分割方法、系统和设备
CN112686817B (zh) * 2020-12-25 2023-04-07 天津中科智能识别产业技术研究院有限公司 一种基于不确定性估计的图像补全方法
CN112784818B (zh) * 2021-03-03 2023-03-14 电子科技大学 基于分组式主动学习在光学遥感图像上的识别方法
WO2022243337A2 (fr) * 2021-05-17 2022-11-24 Deep Safety Gmbh Système de détection et de gestion d'incertitude dans des systèmes de perception, pour la détection de nouveaux objets et pour l'anticipation de situation
WO2023280511A1 (fr) * 2021-07-06 2023-01-12 Asml Netherlands B.V. Détermination des erreurs localisées de la prédiction d'image pour améliorer un modèle d'apprentissage machine dans la prédiction d'une image
CN113569926A (zh) * 2021-07-13 2021-10-29 中国资源卫星应用中心 一种具备高迁移性的云分割模型的训练方法及装置
WO2023000159A1 (fr) * 2021-07-20 2023-01-26 海南长光卫星信息技术有限公司 Procédé, appareil et dispositif de classification semi-surveillée pour une image de télédétection à haute résolution, et support
CN113435411B (zh) * 2021-07-26 2022-06-17 中国矿业大学(北京) 一种基于改进DeepLabV3+的露天矿区土地利用识别方法
CN113850825B (zh) * 2021-09-27 2024-03-29 太原理工大学 基于上下文信息和多尺度特征融合的遥感图像道路分割方法
CN114282594A (zh) * 2021-11-19 2022-04-05 广东省人民医院 医学图像分类方法、系统和存储介质
CN114494959B (zh) * 2022-01-24 2025-03-25 中国矿业大学 一种针对视频目标分割的注意力引导的对抗性攻击方法
CN114595748B (zh) * 2022-02-21 2024-02-13 南昌大学 一种用于跌倒防护系统的数据分割方法
CN115471651B (zh) * 2022-08-31 2026-04-07 浙江大学 一种基于点云时空记忆网络的4d目标分割方法
CN115546225B (zh) * 2022-09-06 2025-07-08 华南理工大学 一种提升语义分割网络预测概率分布质量的模型训练方法
CN115482383B (zh) * 2022-09-22 2025-12-23 贵州大学 一种具有稀疏化效应的持续学习语义分割方法
CN116188845A (zh) * 2022-12-30 2023-05-30 支付宝(杭州)信息技术有限公司 对抗攻击的检测方法及系统
CN116434037B (zh) * 2023-04-21 2023-09-22 大连理工大学 基于双层优化学习的多模态遥感目标鲁棒识别方法
CN116844063B (zh) * 2023-05-31 2026-02-10 兰州大学 基于多模态神经网络的沙尘识别方法、设备及存储设备
CN117036941B (zh) * 2023-08-07 2025-09-19 电子科技大学长三角研究院(湖州) 一种基于孪生Unet模型的建筑物变化检测方法及系统
CN117598713B (zh) * 2023-11-30 2024-12-24 网易传媒科技（北京）有限公司 癫痫发作间期的异常放电检测方法、装置、介质及设备
CN118152564A (zh) * 2024-02-29 2024-06-07 北京轩宇信息技术有限公司 一种中文自然语言处理模型的对抗文本生成方法
CN119693811B (zh) * 2024-11-14 2025-12-02 南京航空航天大学 基于多模态遥感图像的小目标检测方法
CN119579460B (zh) * 2024-11-19 2025-11-18 东南大学 一种基于双级优化驱动高频增强的遥感影像泛锐化方法及系统
CN120430418B (zh) * 2025-07-01 2025-09-30 科大讯飞股份有限公司 大模型推理方法、装置、相关设备及计算机程序产品

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10192129B2 (en) * 2015-11-18 2019-01-29 Adobe Systems Incorporated Utilizing interactive deep learning to select objects in digital visual media
EP4145391A1 (fr) 2016-12-23 2023-03-08 HeartFlow, Inc. Systèmes et procédés de segmentation probabiliste dans un traitement d'image anatomique
GB201709672D0 (en) * 2017-06-16 2017-08-02 Ucl Business Plc A system and computer-implemented method for segmenting an image
CN109035263B (zh) * 2018-08-14 2021-10-15 电子科技大学 基于卷积神经网络的脑肿瘤图像自动分割方法

Also Published As

Publication number Publication date
CA3242569A1 (fr) 2020-12-03
EP3977364A4 (fr) 2023-06-14
CA3137030A1 (fr) 2020-12-03
EP3977364A1 (fr) 2022-04-06
WO2020240477A1 (fr) 2020-12-03

Similar Documents

Publication Publication Date Title
CA3137030C (fr) Procede et dispositif de traitement pour entrainer un reseau neuronal
Wang et al. Uncertainty estimation for stereo matching based on evidential deep learning
Mukhoti et al. Evaluating bayesian deep learning methods for semantic segmentation
CN111652321B (zh) 一种基于改进yolov3算法的海上船舶检测方法
US11790492B1 (en) Method of and system for customized image denoising with model interpretations
CN114586078B (zh) 手部姿态估计方法、装置、设备以及计算机存储介质
KR20210122855A (ko) 검출 모델 훈련 방법과 장치, 컴퓨터 장치, 및 저장 매체
CN112132012B (zh) 基于生成对抗网络的高分辨率sar船舶图像生成方法
CN116109812A (zh) 一种基于非极大值抑制阈值优化的目标检测方法
US20250355107A1 (en) Multi-target detection using convex sparsity prior
Upadhyay et al. Robustness via uncertainty-aware cycle consistency
Amjad et al. Kalman filter-based convolutional neural network for robust tracking of froth-middling interface in a primary separation vessel in presence of occlusions
CN118552924A (zh) 一种基于起始点引导的车道线检测方法
CN120510189B (zh) 遥感视频的多运动目标跟踪方法及存储介质
CN112862002B (zh) 多尺度目标检测模型的训练方法、目标检测方法和装置
CN114283463A (zh) 图像处理方法、装置、电子设备及存储介质
CN119625319A (zh) 证据不确定性渐进引导的可信医学影像分割方法及设备
Nazarkevych et al. Method of Dataset Filling and Recognition of Moving Objects in Video Sequences based on YOLO.
CN114663470A (zh) 基于软选择的自适应跨模态视觉跟踪方法
CN120122071B (zh) 基于自适应流形判别回归的多雷达协同目标识别方法
CN119920016B (zh) 一种基于4d毫米波雷达的人体姿势行为识别方法
US20230196752A1 (en) Information processing apparatus, information processing method, and non-transitory computer-readable storage medium
Xu et al. An edge-oriented framework for saliency detection
CN120765956A (zh) 混凝土中纤维提取方法、装置、设备及存储介质
Singh et al. Building Semantic Segmentation on Cartosat-2E Satellite Images Using a Three-Stage Cross-Resolution Pipeline

Legal Events

Date Code Title Description
EEER Examination request

Effective date: 20211105

D00 Search and/or examination requested or commenced

Free format text: ST27 STATUS EVENT CODE: A-2-2-D10-D00-D163 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: RESPONSE TO CONDITIONAL NOTICE OF ALLOWANCE

Effective date: 20240704

B12 Application deemed to be withdrawn, abandoned or lapsed

Free format text: ST27 STATUS EVENT CODE: N-2-6-B10-B12-B302 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: DEEMED ABANDONED - CONDITIONS FOR GRANT DETERMINED NOT COMPLIANT

Effective date: 20240903

MFA Maintenance fee for application paid

Free format text: FEE DESCRIPTION TEXT: MF (APPLICATION, 5TH ANNIV.) - STANDARD

Year of fee payment: 5

U00 Fee paid

Free format text: ST27 STATUS EVENT CODE: N-6-6-U10-U00-U101 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: MAINTENANCE REQUEST RECEIVED

Effective date: 20250304

U11 Full renewal or maintenance fee paid

Free format text: ST27 STATUS EVENT CODE: N-6-6-U10-U11-U102 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: MAINTENANCE FEE PAYMENT PAID IN FULL

Effective date: 20260106

D22 Grant of ip right intended

Free format text: ST27 STATUS EVENT CODE: A-2-4-D10-D22-D143 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: PRE-GRANT

Effective date: 20260214

W00 Other event occurred

Free format text: ST27 STATUS EVENT CODE: A-2-2-W10-W00-W111 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: CORRESPONDENT DETERMINED COMPLIANT

Effective date: 20260214

Q17 Modified document published

Free format text: ST27 STATUS EVENT CODE: A-4-4-Q10-Q17-Q103 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: DOCUMENT PUBLISHED

Effective date: 20260319

F11 Ip right granted following substantive examination

Free format text: ST27 STATUS EVENT CODE: A-4-4-F10-F11-X000 (AS PROVIDED BY THE NATIONAL OFFICE); EVENT TEXT: GRANT BY ISSUANCE

Effective date: 20260324