Small object detection on highways via balance feature fusion and task-specific encoding network  

作　　者：YU Minming CHAN Sixian ZHOU Xiaolong LAI Zhounian 

机构地区：[1]College of Computer Science and Technology,Zhejiang University of Technology,Hangzhou 310023,China [2]Hangzhou Xsuan Technology Co.,Ltd.,Hangzhou 310051,China [3]Quzhou University,Quzhou 324000,China [4]Huzhou Institute of Zhejiang University,Huzhou 313002,China2

出　　处：《Optoelectronics Letters》2024年第7期424-429,共6页光电子快报（英文版）

基　　金：supported by the National Natural Science Foundation of China(Nos.61906168 and 62202337);the Zhejiang Provincial Natural Science Foundation of China(Nos.LY23F020023 and LZ23F020001);the Construction of Hubei Provincial Key Laboratory for Intelligent Visual Monitoring of Hydropower Projects(No.2022SDSJ01);the Hangzhou AI Major Scientific and Technological Innovation Project(No.2022AIZD0061)。

摘　　要：Detecting small objects on highways is a novel research topic.Due to the small pixel of objects on highways,traditional detectors have difficulty in capturing discriminative features.Additionally,the imbalance of feature fusion methods and the inconsistency between classification and regression tasks lead to poor detection performance on highways.In this paper,we propose a balance feature fusion and task-specific encoding network to address these issues.Specifically,we design a balance feature pyramid network(FPN)to integrate the importance of each layer of feature maps and construct long-range dependencies among them,thereby making the features more discriminative.In addition,we present task-specific decoupled head,which utilizes task-specific encoding to moderate the imbalance between the classification and regression tasks.As demonstrated by extensive experiments and visualizations,our method obtains outstanding detection performance on small object detection on highways(HSOD)dataset and AI-TOD dataset.

关 键 词：NETWORK utilize HIGHWAY 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] U495[自动化与计算机技术—计算机科学与技术]