Research on Deep Learning Network Compression Method for Missile Borne Image

Aiming at the problem that the target recognition algorithm based on deep learning has the characteristics of complex network structure, large amount of parameters and high computation delay, which is difficult to directly apply to the missile-borne task, the network lightweight method are summarize...

Full description

Saved in:
Bibliographic Details
Published inHangkong Bingqi Vol. 30; no. 1; pp. 95 - 103
Main Author Gao Yibo, Yang Chuandong, Chen Dong, Ling Chong
Format Journal Article
LanguageChinese
Published Editorial Office of Aero Weaponry 01.02.2023
Subjects
network model compression|lightweight network|missile borne image|deep learning model|algorithm transplantation
Online AccessGet full text

Cover

More Information
Summary:Aiming at the problem that the target recognition algorithm based on deep learning has the characteristics of complex network structure, large amount of parameters and high computation delay, which is difficult to directly apply to the missile-borne task, the network lightweight method are summarized. The advantages and respective characteristics of existing compression methods and lightweight networks are introduced, and the excellent algorithms in va-rious aspects are selected for comparison. Finally, combined with the development of deep learning in the field of target detection, the lightweight missile borne image target recognition algorithm is prospected.
ISSN:1673-5048
DOI:10.12132/ISSN.1673-5048.2022.0079