摘要
为提升复杂环境装甲目标的检测精度,提出了一种显著目标检测算法,该算法通过视觉注意机制与联合金字塔上采样模块分别获取视觉注意机制约束的低层次特征与多尺度池化语义特征,利用聚合策略进行融合,提升低对比度或遮挡情况下的目标表征能力。测试结果表明,文中算法对复杂场景下多尺度目标均取得了良好的检测效果,其精度、召回率与平均精度分别为72.2%、71.4%与77.1%,能够满足实际应用需求。
To improve the detection accuracy for armored targets in complex environment,a salient target detection algorithm was proposed.The low-level features and multi-scale pooling semantic features constrained by visual attention mechanism were respectively obtained by visual attention mechanism and joint pyramid upsampling module.Then the aggregation strategy was used to fuse,so as to improve the ability of target representation in low contrast or occlusion conditions.The experimental results show that the proposed algorithm obtains good detection results for multi-scale targets in complex scenes,the precision,recall rate and mean average precision(mAP)are 72.2%,71.4%and 77.1%,respectively,which can meet the practical application requirements.
作者
马少斌
张成文
MA Shaobin;ZHANG Chengwen(School of Digital Media,Lanzhou University of Arts and Science,Lanzhou 730010,China;VR Technology R&D and Promotion Center,Lanzhou University of Arts and Science,Lanzhou 730010,China)
出处
《应用光学》
CAS
CSCD
北大核心
2021年第6期1056-1061,共6页
Journal of Applied Optics
基金
甘肃省教育厅产业支撑引导项目(2019c-09)
甘肃省高等学校科研项目(2018A-138)
兰州文理学院科技成果培育转化项目(G2H001)。
关键词
目标检测
池化语义
注意机制
部分遮挡
聚合策略
target detection
pooling semantics
attention mechanism
partial occlusion
aggregation strategy
