典型文献
Pre-locate net for object detection in high-resolution images
文献摘要：
Small-object detection has long been a challenge.High-megapixel cameras are used to solve this problem in industries.However,current detectors are inefficient for high-resolution images.In this work,we propose a new module called Pre-Locate Net,which is a plug-and-play structure that can be combined with most popular detectors.We inspire the use of classification ideas to obtain candidate regions in images,greatly reducing the amount of calculation,and thus achieving rapid detection in high-resolution images.Pre-Locate Net mainly includes two parts,can-didate region classification and behavior classification.Candidate region classification is used to obtain a candidate region,and behavior classification is used to estimate the scale of an object.Dif-ferent follow-up processing is adopted according to different scales to balance the variance of the network input.Different from the popular candidate region generation method,we abandon the idea of regression of a bounding box and adopt the concept of classification,so as to realize the prediction of a candidate region in the shallow network.We build a high-resolution dataset of air-craft and landing gears covering complex scenes to verify the effectiveness of our method.Com-pared to state-of-the-art detectors(e.g.,Guided Anchoring,Libra-RCNN,and FASF),our method achieves the best mAP of 94.5 on 1920×1080 images at 16.7 FPS.
文献关键词：
中图分类号：
作者姓名：
Yunhao ZHANG;Tingbing XU;Zhenzhong WEI
作者机构：
Key Laboratory of Precision Opto-mechatronics Technology,Ministry of Education,and the School of Instrumentation and Optoelectronic Engineering,Beihang University,Beijing 100083,China
文献出处：
引用格式：
[1]Yunhao ZHANG;Tingbing XU;Zhenzhong WEI-.Pre-locate net for object detection in high-resolution images)[J].中国航空学报（英文版）,2022(10):313-325
A类：
megapixel,FASF
B类：
Pre,locate,object,detection,high,resolution,images,Small,has,long,been,challenge,High,cameras,used,solve,this,problem,industries,However,current,detectors,inefficient,In,propose,new,module,called,Locate,Net,which,plug,play,structure,that,combined,most,popular,We,inspire,classification,ideas,obtain,candidate,regions,greatly,reducing,amount,calculation,thus,achieving,rapid,mainly,includes,parts,behavior,Candidate,estimate,follow,up,processing,adopted,according,different,scales,balance,variance,network,input,Different,from,generation,method,abandon,regression,bounding,box,concept,realize,prediction,shallow,build,dataset,air,craft,landing,gears,covering,complex,scenes,verify,effectiveness,our,Com,pared,state,Guided,Anchoring,Libra,RCNN,achieves,best,mAP,FPS
AB值：
0.541007
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
