中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 脑图谱与类脑智能实验室
Contour Primitive of Interest Extraction Network Based on Dual-Metric One-Shot Learning for Vision Measurement

文献类型：期刊论文

作者Qin, Fangbo2,3; Lin, Shan1; Xu, De2,3
刊名IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
出版日期2023-04-01
卷号19期号:4页码:5839-5848
ISSN号1551-3203
关键词Feature extraction Measurement Task analysis Imaging Image segmentation Prototypes Training Contour extraction deep learning metric learning one-shot learning vision measurement
DOI10.1109/TII.2022.3201008
通讯作者Xu, De(de.xu@ia.ac.cn)
英文摘要Although many existing vision measurement systems have achieved high performances, they are object-specific and have limitations in flexibility. Toward intelligent vision measurement that can be conveniently reused for novel objects, this article focuses on the image geometric feature extraction with one-shot learning ability. We propose a contour primitive of interest (CPI) extraction network with dual metric (CPieNet-DM), which can obtain a designated CPI in a query image of a novel object under the guidance of only one annotated support image. First, the dual-metric learning mechanism is proposed, which not only utilizes inter-image similarity as guidance but also leverages the intra-image coherency of CPI pixels to facilitate the inference. Second, a neural network is designed to infer the CPI map based on the dual metric, which also predicts the CPI's geometric parameters. Moreover, the dual context aggregator is plugged in to provide the awareness of both images' contexts. Third, the network training is jointly supervised by the multiple tasks of dual-metric learning, geometric parameters regression, and CPI extraction. The online hard example mining is utilized to improve the training outcome. The effectiveness of the proposed methods is validated with a series of experiments.
WOS关键词IMAGES
资助项目National Natural Science Foundation of China[62103413] ; National Natural Science Foundation of China[61873266] ; National Key Research and Development Program of China[2021ZD0200402]
WOS研究方向Automation & Control Systems ; Computer Science ; Engineering
语种英语
出版者IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号WOS:000965767100001
资助机构National Natural Science Foundation of China ; National Key Research and Development Program of China
源URL[http://ir.ia.ac.cn/handle/173211/53852]  
专题脑图谱与类脑智能实验室
中科院工业视觉智能装备工程实验室
通讯作者Xu, De
作者单位1.Univ Calif San Diego, Dept Elect & Comp Engn, La Jolla, CA 92093 USA
2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
推荐引用方式
GB/T 7714		 Qin, Fangbo,Lin, Shan,Xu, De. Contour Primitive of Interest Extraction Network Based on Dual-Metric One-Shot Learning for Vision Measurement[J]. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS,2023,19(4):5839-5848.
APA Qin, Fangbo,Lin, Shan,&Xu, De.(2023).Contour Primitive of Interest Extraction Network Based on Dual-Metric One-Shot Learning for Vision Measurement.IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS,19(4),5839-5848.
MLA Qin, Fangbo,et al."Contour Primitive of Interest Extraction Network Based on Dual-Metric One-Shot Learning for Vision Measurement".IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 19.4(2023):5839-5848.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。