计算机工程与应用2024，Vol.60Issue(15)：211-220,10.DOI:10.3778/j.issn.1002-8331.2305-0254
采用动态样本分配的特征融合目标检测算法
Feature Fusion Target Detection Algorithm Using Dynamic Sample Assignment
摘要
Abstract
A multi-scale feature fusion target detection algorithm with dynamic sample allocation strategy is proposed to address the problems of low detection accuracy and poor prediction ability of small targets in the lightweight target detec-tion algorithm SSD-Lite.Firstly,the feature pyramid network(FPN)is introduced in the neck network of the lightweight target detection algorithm SSD-Lite and designed to be lightweight,while the residual feature augmentation(RFA)mod-ule is introduced,which uses residual branches to inject different.Then,this paper inserts a lightweight attention mecha-nism ECA module into the feature pyramid structure to improve the ability of network to focus on important features.Finally,to address the problems of poor adaptability of positive and negative sample assignment and difficulty in selecting high-quality positive samples caused by the fixed Intersection-over-Union(IOU)threshold sample assignment strategy used in the network training process,this paper designs a dynamic sample assignment strategy,which eliminates the pre-setting of anchor frames and adopts the centroid sampling method,while combining the sample mean and standard devia-tion as screening thresholds to reduce the influence of artificial a priori and improve the algorithm performance without changing the network structure.The algorithm is tested on Pascal VOC dataset,and the experimental results show that the overall prediction accuracy of the algorithm is improved by 1.9 percentage points compared with the benchmark algo-rithm,the detection ability of small targets is improved by 3.3 percentage points,and the inference delay of the algo-rithm is increased by only 2.32%.The experiments demonstrate that the algorithm can significantly improve the predic-tion accuracy of the algorithm with a small performance cost.关键词
特征金字塔结构/残差特征增强模块/轻量级注意力机制/动态样本分配策略Key words
feature pyramid structure/residual feature augmentatione module/lightweight attention mechanism/dynamic sample assignment strategy分类
信息技术与安全科学引用本文复制引用
牛文涛,王鹏,陈遵田,李晓艳,郜辉,孙梦宇..采用动态样本分配的特征融合目标检测算法[J].计算机工程与应用,2024,60(15):211-220,10.基金项目
国家自然科学基金(62171360) （62171360）
陕西省科技厅重点研发计划(2022GY-110) （2022GY-110）
西安工业大学校长基金面上培育项目(XGPY200217) （XGPY200217）
西安市智能兵器重点实验室项目(2019220514SYS020CG042) （2019220514SYS020CG042）
国家重点研发计划(2022YFF0604900) （2022YFF0604900）
2022年度陕西高校青年创新团队项目 （）
山东省智慧交通重点实验室项目(筹). （筹）
