南方科技大学知识苑(SUSTech KC): 图像分类模型自学习的方法、装置、电子设备及存储介质

Title	图像分类模型自学习的方法、装置、电子设备及存储介质
Alternative Title	Image classification model self-learning method and device, electronic equipment and storage medium
Author	宋丽妍; 姚新; 武晓宇; 胡崝
First Inventor	宋丽妍
Original applicant	南方科技大学 ; 华为技术有限公司
First applicant	南方科技大学
Address of First applicant	518055 广东省深圳市南山区西丽学苑大道1088号
Current applicant	南方科技大学 ; 华为技术有限公司
Address of Current applicant	518055 广东省深圳市南山区西丽学苑大道1088号 (广东,深圳,南山区)
First Current Applicant	南方科技大学
Address of First Current Applicant	518055 广东省深圳市南山区西丽学苑大道1088号 (广东,深圳,南山区)
Application Number	CN202110241737.2
Application Date	2021-03-04
Open (Notice) Number	CN112966739A
Date Available	2021-06-15
Status of Patent	实质审查
Legal Date	2021-07-02
Subtype	发明申请
SUSTech Authorship	First
Abstract	本发明实施例公开了一种图像分类模型自学习的方法、装置、电子设备及存储介质，该方法包括：获取训练好的图像分类模型；对所述图像分类模型的输出层增加输出节点；获取测试图像并判断是否属于所述输出节点后按预设规则输入所述图像分类模型中；通过所述图像分类模型分类识别所述测试图像并进行更新。本发明实施例提供的一种图像分类模型自学习的方法，通过对图像分类模型进行自动更新，解决了现有技术中图像分类模型在部署后很可能出现的性能退化问题，实现了基于规则和模型判断的策略大大降低了人工标注图像数据的成本，从而获得更具有代表性的新的、有标注的图像数据的效果。
Other Abstract	The embodiment of the invention discloses an image classification model self-learning method and device, electronic equipment and a storage medium. The method comprises the steps: acquiring a trained image classification model; adding output nodes to an output layer of the image classification model; obtaining a test image, judging whether the test image belongs to the output node or not, and inputting the test image into the image classification model according to a preset rule; and classifying and identifying the test image through the image classification model and updating the test image. According to the image classification model self-learning method provided by the embodiment of the invention, the image classification model is automatically updated, so that the problem of performance degradation possibly occurring after deployment of the image classification model in the prior art is solved, the strategy based on rule and model judgment is realized, the cost of manually annotating image data is greatly reduced, and the effect of new and marked image data with more representativeness is obtained.
CPC Classification Number	G06F18/217 ; G06F18/24 ; G06F18/214
IPC Classification Number	G06K9/62
INPADOC Legal Status	（ENTRY INTO FORCE OF REQUEST FOR SUBSTANTIVE EXAMINATION）[2021-07-02][CN]
INPADOC Patent Family Count	1
Extended Patent Family Count	1
Priority date	2021-03-04
Patent Agent	潘登
Agency	北京品源专利代理有限公司
URL	[Source Record]
Data Source	PatSnap
Document Type	Patent
Identifier	http://kc.sustech.edu.cn/handle/2SGJ60CL/532699
Department	Department of Computer Science and Engineering
Recommended Citation GB/T 7714	宋丽妍,姚新,武晓宇,等. 图像分类模型自学习的方法、装置、电子设备及存储介质.

Files in This Item:
There are no files associated with this item.

If you have any objections to this item, please fill out the form below and the administrator will contact you as soon as possible.
Content:
Email：	*
Affiliation Sequential No:
Verification Code:	Refresh

Any comments and suggestions are welcomed.
Title:	*
Content:
Email：	*
Verification Code:	Refresh