Research on Small Size Object Detection in Complex Background

Peng Du, Xiujie Qu, Tianbo Wei, Cheng Peng, Xinru Zhong, Chen Chen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Citations (Scopus)

Abstract

In object detection tasks, the detection of small size objects is very difficult since these small targets are always tightly grouped and interfered by background information. In order to solve this problem, we propose a novel network architecture based on YOLOv3 and a new feature fusion mechanism. We added multi-scale convolution kernels and differential receptive fields into YOLOv3 to extract the semantic features of the objects by using an Inception-like architecture. We also optimize the weights of feature fusion by selecting appropriate channel number ratios. Our model outperforms YOLOv3 when detecting small and easy clustering objects, such as airplane, bird, and person, and the detection speed is comparable with YOLOv3.

Original languageEnglish
Title of host publicationProceedings 2018 Chinese Automation Congress, CAC 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4216-4220
Number of pages5
ISBN (Electronic)9781728113128
DOIs
Publication statusPublished - 2 Jul 2018
Event2018 Chinese Automation Congress, CAC 2018 - Xi'an, China
Duration: 30 Nov 20182 Dec 2018

Publication series

NameProceedings 2018 Chinese Automation Congress, CAC 2018

Conference

Conference2018 Chinese Automation Congress, CAC 2018
Country/TerritoryChina
CityXi'an
Period30/11/182/12/18

Keywords

  • Inception
  • YOLO
  • information fusion
  • object detection
  • small size objects

Fingerprint

Dive into the research topics of 'Research on Small Size Object Detection in Complex Background'. Together they form a unique fingerprint.

Cite this