版权说明 操作指南
首页 > 成果 > 成果详情

An enhanced strategy for query-based defect detector via adaptive spatial feature Reorganization And cross-stage query Injection

认领
导出
Link by DOI
反馈
分享
QQ微信 微博
成果类型:
期刊论文
作者:
Ma, Liangcheng;Shao, Haidong;Xu, Xiaoru;Wu, Xizhi
通讯作者:
Wu, XZ
作者机构:
[Xu, Xiaoru; Wu, Xizhi; Ma, Liangcheng] Cent South Univ Forestry & Technol, Coll Mat Sci & Engn, Changsha 410082, Peoples R China.
[Shao, Haidong] Hunan Univ, Coll Mech & Vehicle Engn, Changsha 410082, Peoples R China.
通讯机构:
[Wu, XZ ] C
Cent South Univ Forestry & Technol, Coll Mat Sci & Engn, Changsha 410082, Peoples R China.
语种:
英文
关键词:
Surface defect detection;Deep learning;Adaptive spatial feature reorganization;Cross-stage query injection
期刊:
Advanced Engineering Informatics
ISSN:
1474-0346
年:
2026
卷:
69
页码:
103910
基金类别:
CRediT authorship contribution statement Liangcheng Ma: Writing – review & editing, Writing – original draft, Software, Methodology, Data curation. Haidong Shao: Supervision, Project administration, Methodology. Xiaoru Xu: Visualization, Software, Data curation. Xizhi Wu: Writing – review & editing, Supervision, Resources, Project administration, Investigation, acquisition. The project is supported by Serving Rural Revitalization of Hunan Provincial Science and Technology Commissioner (No 2023NK4285 ), China Postdoctoral Science Foundation (No 2021M690768 ).
机构署名:
本校为第一且通讯机构
院系归属:
材料科学与工程学院
摘要:
Detection of surface defects from images is crucial to ensure high quality products in manufacturing applications, where surface detection of small defects plays a vital role and has received much attention in the manufacturing industry. However, existing detection solutions perform unevenly in different small defect scenarios. Therefore, this paper proposes an efficient enhancement strategy (RAI) to enhance the model’s ability to detect small surface defects. It consists of two major parts: (i) the feature information enhancement part (ASFR),...

反馈

验证码:
看不清楚,换一个
确定
取消

成果认领

标题:
用户 作者 通讯作者
请选择
请选择
确定
取消

提示

该栏目需要登录且有访问权限才可以访问

如果您有访问权限,请直接 登录访问

如果您没有访问权限,请联系管理员申请开通

管理员联系邮箱:yun@hnwdkj.com