Infrared and Laser Engineering, Volume. 51, Issue 4, 20210309(2022)
Zernike’s feature selection based on LGBM and identification methods of infrared image target
Figures & Tables(6)
Fig. 1. Flow chart of Zernike’s feature selection based on LGBM and identification methods of infrared image target
Table 1. Recognition results of the proposed method on the original test samples
View tableView in ArticleTable 1. Recognition results of the proposed method on the original test samples
Class Pickup SUV BTR70 BRDM2 BMP2 T72 ZSU23/4 2S3 MTLB D20 Recognition rate Pickup 97 0 2 0 1 0 0 0 0 0 97% SUV 0 98 0 1 0 0 0 0 0 1 98% BTR70 0 0 100 0 0 0 0 0 0 0 100% BRDM2 1 1 0 98 0 0 0 0 0 0 98% BMP2 0 0 1 0 97 0 0 1 0 1 97% T72 0 0 0 2 1 96 0 1 0 0 96% ZSU23/4 2 0 1 0 0 0 97 0 0 0 97% 2S3 0 0 0 0 0 0 0 100 0 0 100% MTLB 0 0 0 0 0 0 0 0 99 1 99% D20 0 1 0 1 0 1 0 0 0 97 97% Average 97.9%
Table 2. Performance of different methods on original samples
View tableView in ArticleTable 2. Performance of different methods on original samples
Method Average recognition rate Proposed method 97.9% SVM 94.5% SRC 95.1% Zernike 96.9% CNN 97.2%
Tools
Get Citation
Copy Citation Text
Mianrong Yang, Liping Niu. Zernike’s feature selection based on LGBM and identification methods of infrared image target[J]. Infrared and Laser Engineering, 2022, 51(4): 20210309
Paper Information
Category: Image processing
Received: Dec. 25, 2021
Accepted: --
Published Online: May. 18, 2022
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20