Simulation Learning Method for Discovery of Camouflage Targets Based on Deep Neural Networks

Liu Zhuo; Xiaoqi Chen; Zhenping Xie; Xiaojun Jiang; Daokun Bi

doi:10.3788/LOP56.071102

Laser & Optoelectronics Progress, Volume. 56, Issue 7, 071102(2019)

Simulation Learning Method for Discovery of Camouflage Targets Based on Deep Neural Networks

Liu Zhuo^1,2, Xiaoqi Chen^1,2, Zhenping Xie^1,2、*, Xiaojun Jiang³, and Daokun Bi³

Author Affiliations

¹ School of Digital Media, Jiangnan University, Wuxi, Jiangsu 214122, China

² Jiangsu Key Laboratory of Media Design and Software Technology, Wuxi, Jiangsu 214122, China

³ Science and Technology on Near-Surface Detection Laboratory, Wuxi, Jiangsu 214035, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(10)

Fig. 1. Simulation learning framework for discovery of camouflage targets

Download full size

Fig. 2. Simulation model of camouflage scenes

Download full size

Fig. 3. Simulation results of camouflage scenes

Download full size

Fig. 4. Preprocessing of scene area in which target can be placed. (a) Original scene map; (b) preprocessed binary map

Download full size

Fig. 5. Parallel dilated convolution

Download full size

Fig. 6. Multi-scale discovery model of camouflage targets

Download full size

Fig. 7. Experimental discovery output results of representative camouflage targets. (a)(c) Test input images; (b) (d) corresponding model test outputs

Download full size

Table 1. Performance results obtained by proposed method at different scale parameters%
View table
Table 1. Performance results obtained by proposed method at different scale parameters%
Method P_A R M_IOU
Scale1(0.5) 93.17 90.35 88.77
Scale2(0.75) 94.68 91.05 88.84
Scale3(1.0) 96.15 91.93 89.14
Scale1+2 96.42 92.14 89.42
Out(Scale1+2+3) 98.78 93.56 90.96

Table 2. Discovery performance results of camouflage targets obtained by different methods%
View table
Table 2. Discovery performance results of camouflage targets obtained by different methods%
Method P_A R M_IOU
FCN 86.17 85.35 76.71
Pix2Pix 93.68 89.35 86.88
Out 98.78 93.56 90.96

Table 3. Model performance results obtained on a group of real scene images%
View table
Table 3. Model performance results obtained on a group of real scene images%
Expert interpretation level P_A R M_IOU
Great 78.04 85.01 71.60
Good 91.68 90.74 84.78
Bad 93.99 91.54 87.12

Tools

Get Citation

Copy Citation Text

Liu Zhuo, Xiaoqi Chen, Zhenping Xie, Xiaojun Jiang, Daokun Bi. Simulation Learning Method for Discovery of Camouflage Targets Based on Deep Neural Networks[J]. Laser & Optoelectronics Progress, 2019, 56(7): 071102

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Imaging Systems

Received: Sep. 29, 2018

Accepted: Oct. 22, 2018

Published Online: Jul. 30, 2019

The Author Email: Zhenping Xie (xiezhenping@hotmail.com)

DOI:10.3788/LOP56.071102

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Performance results obtained by proposed method at different scale parameters%

Table 1. Performance results obtained by proposed method at different scale parameters%

Table 2. Discovery performance results of camouflage targets obtained by different methods%

Table 2. Discovery performance results of camouflage targets obtained by different methods%

Table 3. Model performance results obtained on a group of real scene images%

Table 3. Model performance results obtained on a group of real scene images%