Skip to main navigation Skip to search Skip to main content

A fast recognition algorithm for dimension decreased extraction of space non-cooperative target

  • Yue Liu*
  • , Jinyu Fu
  • , Yifan Liu
  • , Guanghui Sun
  • *Corresponding author for this work

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

Abstract

A rapid space non-cooperative target recognition method based on dimension reduction feature extraction was proposed to identify the target satellite containing interference information quickly and efficiently in the three-dimensional space. First, K-means clustering was performed on the source point cloud to obtain n clusters of point cloud. Next, valid points that met the requirements were retained according to the target characteristics. Then kd-tree algorithm was used to get the point density of each point, and further, the main features of various clusters were obtained after projecting the n clusters of point cloud based on the point density. Finally, all kinds of point cloud after dimension reduction were matched with a given reference point cloud, which aimed to get rough and precise two-dimensional rotation matrices respectively and corresponding rotation errors. By searching for the minimum rotation error, the cluster of the target could be searched. The source point cloud clustered into 4 and 6 clusters respectively was matched with the given 6 sets of reference point cloud. The result of the experiment shows that the proposed algorithm is not affected by the number of clusters for the point cloud data with interference information, and it can find the target accurately with a 100% recognition rate within 2 seconds.

Original languageEnglish
Title of host publicationSeventh Symposium on Novel Photoelectronic Detection Technology and Applications
EditorsJunhong Su, Junhao Chu, Qifeng Yu, Huilin Jiang
PublisherSPIE
ISBN (Electronic)9781510643611
DOIs
StatePublished - 2021
Externally publishedYes
Event7th Symposium on Novel Photoelectronic Detection Technology and Applications - Kunming, China
Duration: 5 Nov 20207 Nov 2020

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11763
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference7th Symposium on Novel Photoelectronic Detection Technology and Applications
Country/TerritoryChina
CityKunming
Period5/11/207/11/20

Keywords

  • 3D point cloud
  • Feature extraction
  • K-means clustering
  • Kd-tree search
  • Non-cooperative target recognition

Fingerprint

Dive into the research topics of 'A fast recognition algorithm for dimension decreased extraction of space non-cooperative target'. Together they form a unique fingerprint.

Cite this