Infrared and Laser Engineering, Volume. 51, Issue 6, 20220249(2022)
Polarized multispectral image classification of typical ground objects based on neural network (Invited)
Figures & Tables(8)
Fig. 1. (a) Structure diagram of polarized multispectral imaging system; (b) Physical system image
Fig. 4. (a) Samples selected from a set of images; (b)
Fig. 5. (a) Training accuracy iterative image of network; (b) Training loss iterative image of network
Fig. 6. (a) Experimental scene; (b) One original image; (c) Classification results of this set of images by neural network
Table 1. Main device parameters of polarized multispectral imaging system
View tableView in ArticleTable 1. Main device parameters of polarized multispectral imaging system
Device Performance Parameter Specifications of CCD sensor Type MER2-502-79 U3 M/C Port USB3.0 Resolution 2 448(H) × 2 048(V) Frame rate 79.1 fps Sensor 2/3", Sony IMX250 Global shutter CMOS Pixel size 3.45 μm×3.45 μm Specifications of lens Lens mount Port C Specifications of polarizer Type GSP-25 Extinction ratio >1000:1 Specifications of filter Half bandwidth 10 nm Peak transmittance 50% Mechanical specifications Weight 1.5 kg Size 18(L) cm×10(W) cm×13(H) cm
Table 2. Comparison of different classification methods
View tableView in ArticleTable 2. Comparison of different classification methods
Classification method Training set accuracy Validation set accuracy Accuracy ranking Kappa coefficient Kappa coefficient ranking Network 97.3% 94.2% 1 0.898 1 MLC 95.3% 91.7% 3 0.851 3 MDC 76.5% 76.5% 5 0.584 5 SVM-RBF 71.9% 73.6% 6 0.456 6 SVM-linear 88.9% 85.5% 4 0.725 4
Tools
Get Citation
Copy Citation Text
Ying Zhang, Heshen Li, Hao Wang, Junhua Sun, Xi Zhang, Huilan Liu, Yanhong Lv. Polarized multispectral image classification of typical ground objects based on neural network (Invited)[J]. Infrared and Laser Engineering, 2022, 51(6): 20220249
Paper Information
Category: Invited paper
Received: Apr. 10, 2022
Accepted: --
Published Online: Dec. 20, 2022
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20