WO2021259392A3 - 网络训练方法及装置、图像识别方法和电子设备 - Google Patents

网络训练方法及装置、图像识别方法和电子设备 Download PDF

Info

Publication number
WO2021259392A3
WO2021259392A3 PCT/CN2021/122144 CN2021122144W WO2021259392A3 WO 2021259392 A3 WO2021259392 A3 WO 2021259392A3 CN 2021122144 W CN2021122144 W CN 2021122144W WO 2021259392 A3 WO2021259392 A3 WO 2021259392A3
Authority
WO
WIPO (PCT)
Prior art keywords
image
image recognition
recognition network
training
state
Prior art date
Application number
PCT/CN2021/122144
Other languages
English (en)
French (fr)
Other versions
WO2021259392A2 (zh
Inventor
吴振洲
Original Assignee
北京安德医智科技有限公司
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 北京安德医智科技有限公司 filed Critical 北京安德医智科技有限公司
Publication of WO2021259392A2 publication Critical patent/WO2021259392A2/zh
Publication of WO2021259392A3 publication Critical patent/WO2021259392A3/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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; CALCULATING OR 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Medical Informatics (AREA)
  • Image Analysis (AREA)

Abstract

一种网络训练方法及装置、图像识别方法和电子设备。该方法包括：根据训练集中的第一图像组，对初始状态的图像识别网络进行训练，得到第一状态的图像识别网络（S1），第一图像组至少包括已标注的第一样本图像；根据训练集中的第二图像组，对第一状态的图像识别网络进行训练，得到第二状态的图像识别网络，第二图像组包括已标注的第一、第二样本图像及未标注的第三样本图像（S2）；根据第一图像组，对第二状态的图像识别网络进行训练，得到目标状态的图像识别网络（S3）。能够充分利用粗标注与无标注的图像数据，减少对精细标注的数据需求量，提高识别网络的训练效果，从而提高图像识别网络对图像数据的分割识别精度。
PCT/CN2021/122144 2021-01-21 2021-09-30 网络训练方法及装置、图像识别方法和电子设备 WO2021259392A2 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110078577.4 2021-01-21
CN202110078577.4A CN112396605B (zh) 2021-01-21 2021-01-21 网络训练方法及装置、图像识别方法和电子设备

Publications (2)

Publication Number Publication Date
WO2021259392A2 WO2021259392A2 (zh) 2021-12-30
WO2021259392A3 true WO2021259392A3 (zh) 2022-02-17

Family

ID=74624969

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/122144 WO2021259392A2 (zh) 2021-01-21 2021-09-30 网络训练方法及装置、图像识别方法和电子设备

Country Status (2)

Country Link
CN (1) CN112396605B (zh)
WO (1) WO2021259392A2 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112396605B (zh) * 2021-01-21 2021-04-23 北京安德医智科技有限公司 网络训练方法及装置、图像识别方法和电子设备
CN115346076B (zh) * 2022-10-18 2023-01-17 安翰科技(武汉)股份有限公司 病理图像识别方法及其模型训练方法、系统和存储介质
CN116258717B (zh) * 2023-05-15 2023-09-08 广州思德医疗科技有限公司 病灶识别方法、装置、设备和存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108681743A (zh) * 2018-04-16 2018-10-19 腾讯科技（深圳）有限公司 图像对象识别方法和装置、存储介质
US10430692B1 (en) * 2019-01-17 2019-10-01 Capital One Services, Llc Generating synthetic models or virtual objects for training a deep learning network
CN110472737A (zh) * 2019-08-15 2019-11-19 腾讯医疗健康(深圳)有限公司 神经网络模型的训练方法、装置和医学图像处理系统
CN111582371A (zh) * 2020-05-07 2020-08-25 广州视源电子科技股份有限公司 一种图像分类网络的训练方法、装置、设备及存储介质
CN112396605A (zh) * 2021-01-21 2021-02-23 北京安德医智科技有限公司 网络训练方法及装置、图像识别方法和电子设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110245721B (zh) * 2019-06-25 2023-09-05 深圳市腾讯计算机系统有限公司 神经网络模型的训练方法、装置和电子设备
CN111126481A (zh) * 2019-12-20 2020-05-08 湖南千视通信息科技有限公司 一种神经网络模型的训练方法及装置
CN111489366A (zh) * 2020-04-15 2020-08-04 上海商汤临港智能科技有限公司 神经网络的训练、图像语义分割方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108681743A (zh) * 2018-04-16 2018-10-19 腾讯科技（深圳）有限公司 图像对象识别方法和装置、存储介质
US10430692B1 (en) * 2019-01-17 2019-10-01 Capital One Services, Llc Generating synthetic models or virtual objects for training a deep learning network
CN110472737A (zh) * 2019-08-15 2019-11-19 腾讯医疗健康(深圳)有限公司 神经网络模型的训练方法、装置和医学图像处理系统
CN111582371A (zh) * 2020-05-07 2020-08-25 广州视源电子科技股份有限公司 一种图像分类网络的训练方法、装置、设备及存储介质
CN112396605A (zh) * 2021-01-21 2021-02-23 北京安德医智科技有限公司 网络训练方法及装置、图像识别方法和电子设备

Also Published As

Publication number Publication date
CN112396605B (zh) 2021-04-23
WO2021259392A2 (zh) 2021-12-30
CN112396605A (zh) 2021-02-23

Similar Documents

Publication Publication Date Title
WO2021259392A3 (zh) 网络训练方法及装置、图像识别方法和电子设备
CN109993160B (zh) 一种图像矫正及文本与位置识别方法及系统
TWI808386B (zh) 圖像識別方法、裝置、終端和儲存媒體
WO2021259393A3 (zh) 图像处理方法及装置、电子设备
EP3876150A3 (en) Vehicle tracking method and apparatus
EP3869397A3 (en) Method, apparatus, device and storage medium for processing image
CN112949562A (zh) 智适应学习方法及系统
EP3872760A3 (en) Method and apparatus of training depth estimation network, and method and apparatus of estimating depth of image
EP4116861A3 (en) Method and apparatus for pre-training semantic representation model and electronic device
EP3955216A3 (en) Method and apparatus for recognizing image, electronic device and storage medium
EP3819824A3 (en) Image processing method, apparatus, electronic device and storage medium
CN110971457A (zh) 一种基于elm的时间同步方法
EP3910590A3 (en) Method and apparatus of processing image, electronic device, and storage medium
WO2023143640A8 (zh) 搜索意图的查询理解方法和装置、存储介质和电子设备
CN111062221A (zh) 数据处理方法、装置、电子设备以及存储介质
WO2004029833A3 (fr) Procede et dispositif de mesure de similarite entre images
WO2023116137A1 (zh) 一种文本识别方法及其相关设备
Rao et al. Sign Language Recognition using LSTM and Media Pipe
EP4134920A3 (en) Entity recognition method and apparatus, and computer program product
CN109409359A (zh) 一种基于深度学习的视频字幕提取方法
CN113177478A (zh) 一种基于迁移学习的短视频语义标注方法
CN113420839A (zh) 用于堆叠平面目标物体的半自动标注方法及分割定位系统
CN113436198A (zh) 一种协同图像超分辨率重建的遥感图像语义分割方法
CN113821595A (zh) 一种定位引擎的实现方法
CN113591761A (zh) 一种视频镜头语言识别方法

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21830119

Country of ref document: EP

Kind code of ref document: A2