Evaluation Method for Dynamic Effect of Camouflage Based on Battlefield Environment Twinning

doi:10.12382/bgxb.2022.0378

Abstract

Abstract:

The existing evaluation methods for camouflage effectare mainly based on static single target images, which cannot truly respond to the dynamic multi-view characteristics in military reconnaissance, and the corresponding evaluated results are lack of comprehensiveness. A new evaluationmethod fordynamic effect of camouflage based on battlefield environment twinning is proposed. In the new method,a digital twin model of 3D battlefield environment is established, the dynamic multi-view effect map of camouflage targets in the battlefield environment is acquired with the help of 3D simulation and computing technology, the camouflage motion targets in the scene are identified and analyzedby using semantic segmentation and target tracking model, and a new camouflage evaluation index system and a multi-attribute comprehensive evaluation model with multiple reconnaissance views are constructed, finally realizing the dynamic multi-view effect evaluation of different camouflage schemes. The experiments were conducted with simulated military truck model as the experimental object, Unity 3D was used to model the real battlefield environment, and two contrasting camouflage design schemes were evaluated and analyzed. The experimental results show the reasonableness and practicality of the new method.

Key words: camouflage, dynamic effect evaluation, battlefieldenvironment twinning, multi-view reconnaissance, multi-attribute synthesis

CLC Number:

TP391.9

YANG Di, ZHOU Ying, XIE Zhenping. Evaluation Method for Dynamic Effect of Camouflage Based on Battlefield Environment Twinning[J]. Acta Armamentarii, 2024, 45(1): 349-362.

Figures/Tables 15

Fig.1 Technical framework of the proposed method

Fig.2 Construction process of 3D virtual battlefield environment and dataset

Fig.3 Moving object recognition model

Table 1 Experimental camouflage patterns and targets

Fig.4 Virtual battlefield scene used inthe experiments

Fig.5 Examples of image segmentation training dataset

Fig.6 Illustration of evaluation dataset construction perspective

Fig.7 Tracking dataset for evaluation

Table 2 Experimental contrast camouflage design schemes

Fig.8 Performance results on high and low altitude segmentation jamming rates

Fig.9 Performance results on high and lowaltitude camouflage deformations

Fig.10 Performance results on multi-view detection accuracy

Fig.11 Performance results on interference rate of multi-view detection

Table 3 Evaluation decision matrix of two compared camouflage strategies

迷彩伪装	A_i	S_i	E_i	L_i	G_i
迷彩伪装1	{ $0.54 (低)$ , $0.66 (高)$ }	{ $0.53 (低)$ , $0.61 (高)$ }	{ $0.61 (低)$ , $0.72 (高)$ , $0.43 (左)$ , $0.54 (右)$ }	{ $0.72 (低)$ , $0.66 (高)$ , $0.45 (左)$ , $0.46 (右)$ }	{ $0.78 (低)$ , $0.65 (高)$ }
迷彩伪装2	{ $0.44 (低)$ , $0.52 (高)$ }	{ $0.49 (低)$ , $0.42 (高)$ }	{ $0.41 (低)$ , $0.48 (高)$ , $0.41 (左)$ , $0.36 (右)$ }	{ $0.35 (低)$ , $0.14 (高)$ , $0.32 (左)$ , $0.23 (右)$ }	{ $0.54 (低)$ , $0.46 (高)$ }

Table 3 Evaluation decision matrix of two compared camouflage strategies

迷彩伪装	A_i	S_i	E_i	L_i	G_i
迷彩伪装1	{ $0.54 (低)$ , $0.66 (高)$ }	{ $0.53 (低)$ , $0.61 (高)$ }	{ $0.61 (低)$ , $0.72 (高)$ , $0.43 (左)$ , $0.54 (右)$ }	{ $0.72 (低)$ , $0.66 (高)$ , $0.45 (左)$ , $0.46 (右)$ }	{ $0.78 (低)$ , $0.65 (高)$ }
迷彩伪装2	{ $0.44 (低)$ , $0.52 (高)$ }	{ $0.49 (低)$ , $0.42 (高)$ }	{ $0.41 (低)$ , $0.48 (高)$ , $0.41 (左)$ , $0.36 (右)$ }	{ $0.35 (低)$ , $0.14 (高)$ , $0.32 (左)$ , $0.23 (右)$ }	{ $0.54 (低)$ , $0.46 (高)$ }

Table 4 Compared results of camouflage effects of two compared camouflage strategies

平衡系数、熵值法和最大偏差法	正理想点	属性权重	方案与理想点相似度	方案排序
α=β=λ= $13$	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.19,0.20,0.14,0.20,0.27]^T	S(X₁)=0.75, S(X₂)=0.48	X₁>X₂
α=β=0,λ=1	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.15,0.19,0.31,0.33,0.20]^T	S(X₁)=0.41, S(X₂)=0.18	X₁>X₂
熵值法	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.18,0.14,0.16,0.21,0.31]^T	S(X₁)=0.61, S(X₂)=0.52	X₁>X₂
最大偏差法	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.19,0.11,0.23,0.32,0.15]^T	S(X₁)=0.68, S(X₂)=0.46	X₁>X₂

Table 4 Compared results of camouflage effects of two compared camouflage strategies

平衡系数、熵值法和最大偏差法	正理想点	属性权重	方案与理想点相似度	方案排序
α=β=λ= $13$	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.19,0.20,0.14,0.20,0.27]^T	S(X₁)=0.75, S(X₂)=0.48	X₁>X₂
α=β=0,λ=1	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.15,0.19,0.31,0.33,0.20]^T	S(X₁)=0.41, S(X₂)=0.18	X₁>X₂
熵值法	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.18,0.14,0.16,0.21,0.31]^T	S(X₁)=0.61, S(X₂)=0.52	X₁>X₂
最大偏差法	X⁺={0.66,0.61,0.72,0.72,0.78}	[0.19,0.11,0.23,0.32,0.15]^T	S(X₁)=0.68, S(X₂)=0.46	X₁>X₂

References 24

[1]	颜云辉, 王展, 董德威. 军事伪装技术的发展现状与趋势[J]. 中国机械工程, 2012, 23(17):2136-2141.
	YAN Y H, WANG Z, DONG D W. The development status and trend of military camouflage technology[J]. China Mechanical Engineering, 2012, 23(17):2136-2141. (in Chinese)
[2]	HOULBROOK A W, GILMORE M A, MOORHEAD I R, et al. Scene simulation for camouflage assessment[J]. Targets and Backgrounds Ⅵ: Characterization, 2000, 4029:247-255.
[3]	王展, 颜云辉, 焦学勇. 基于灰色理论的迷彩伪装多指标综合评价[J]. 兵工学报, 2013, 34(10): 1250-1257. doi: 10. 3969/ j. issn. 1000-1093. 2013. 10. 009
	WANG Z, YAN Y H, JIAO X Y. Multi-index comprehensive evaluation of camouflage based on grey theory[J]. Acta Armamentarii, 2013, 34(10): 1250-1257. (in Chinese) doi: 10. 3969/ j. issn. 1000-1093. 2013. 10. 009
[4]	杨骏堂, 许卫东, 曲杨, 等. 一种改进的Surendra运动目标伪装效果检测方法[J]. 兵工学报, 2017, 38(1):190-194. doi: 10.3969/j.issn.1000-1093.2017.01.025
	YANG J T, XU W D, QU Y, et al. An improved detection method for camouflage effect of surendra moving target[J]. Acta Armamentarii, 2017, 38(1):190-194. (in Chinese)
[5]	杨鑫, 许卫东, 贾其. 一种基于特征统计的动态伪装效果评估方法[J]. 兵工学报, 2019, 40(8):1693-1699. doi: 10.3969/j.issn.1000-1093.2019.08.019
	YANG X, XU W D, JIA Q. A dynamic camouflage effect assessment method based on feature statistics[J]. Acta Armamentarii, 2019, 40(8): 1693-1699. (in Chinese) doi: 10.3969/j.issn.1000-1093.2019.08.019
[6]	GAN Y Y, LIU C T, LI H C, et al. An evaluation method of dynamic camouflage effect based on multifeature constraints[J]. IEEE Access, 2020, 8:193845-193855. doi: 10.1109/Access.6287639 URL
[7]	郑万里, 杨萍, 闫少强, 等. 军事伪装技术研究现状及发展趋势分析[J]. 现代防御技术, 2022, 50(1):81-86. doi: 10.3969/j.issn.1009-086x.2022.01.012
	ZHENG W L, YANG P, YAN S Q, et al. Analysis of the current situation and development trend of military camouflage technology research[J]. Modern Defense Technology, 2022, 50(1): 81-86. (in Chinese)
[8]	张勇, 吴文健, 刘志明. 仿生迷彩伪装设计[J]. 计算机工程, 2009, 35(6):35-37,40. doi: 10.3969/j.issn.1000-3428.2009.06.012
	ZHANG Y, WU W J, LIU Z M. Bionic camouflage camouflage design[J]. Computer Engineering, 2009, 35(6): 35-37,40. (in Chinese) doi: 10.3969/j.issn.1000-3428.2009.06.012
[9]	武国晶, 吕绪良, 许卫东, 等. 数码迷彩立体动态成像算法研究[J]. 兵工学报, 2016, 37(1): 103-108. doi: 10.3969/j.issn.1000-1093.2016.01.016
	WU G J, LÜ X L, XU W D, et al. Research on digital camouflage stereo dynamic imaging algorithm[J]. Acta Armamentarii, 2016, 37(1): 103-108. (in Chinese)
[10]	WANG S, MAO Z L, ZENG C H, et al. A new method of virtual reality based on Unity3D[C]// Proceedings of the 2010 18th International Conference on Geoinformatics. Beijing, China: IEEE, 2010: 1-5.
[11]	OLAVERRI-MONREAL C, ERREA-MORENO J, DÍAZ-ÁLVAREZ A, et al. Connection of the SUMO microscopic traffic simulator and the Unity 3D game engine to evaluate V2X communication-based systems[J]. Sensors, 2018, 18(12):4399. doi: 10.3390/s18124399 URL
[12]	HUO A Q, XU J R, LI H P, et al. Design and implementation of drilling engineering simulation system based on Unity3D[J]. Journal of Physics: Conference Series. Xi’an, China: IOP Publishing, 2021, 1894(1):012043.
[13]	MERILAITA S, SCOTT-SAMUEL N E, CUTHILL I C. How camouflage works[J]. Philosophical Transactions of the Royal Society of London, 2017, 372(1724):20160341.
[14]	CHEN L C, ZHU Y, PAPANDREOU G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation:arXiv:1802.02611[R]. Ithaca,NY,US:Cornell,University, 2018: 1802.02611.
[15]	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection: arXiv: 1506.02640[R]. Ithaca,NY,US:Cornell University, 2015:1506.02640.
[16]	WOJKE N, BEWLEY A, PAULUS D. Simple online and realtime tracking with a deep association metric: arXiv:1703.07402[R]. Ithaca,NY, US: Cornell University, 2017:1703.07402.
[17]	MONDAL A. Camouflage design, assessment and breaking techniques: a survey[J]. Multimedia Systems, 2022, 28(1): 141-160. doi: 10.1007/s00530-021-00813-6
[18]	陈琳, 刘允刚. 面向无人机的视觉目标跟踪算法:综述与展望[J]. 信息与控制, 2022, 51(1):23-40. doi: 10.13976/j.cnki.xk.2022.1144
	CHEN L, LIU Y G. Visual target tracking algorithms for UAVs:a review and outlook[J]. Information and Control, 2022, 51(1): 23-40. (in Chinese) doi: 10.13976/j.cnki.xk.2022.1144
[19]	麻永平, 张炜, 刘东旭. 高光谱侦察技术特点及其对地面军事目标威胁分析[J]. 上海航天, 2012, 29(1):37-40,59.
	MA Y P, ZHANG W, LIU D X. Analysis of the characteristics of hyperspectral reconnaissance technology and its threat to ground military targets[J]. Shanghai Space, 2012, 29(1): 37-40,59. (in Chinese)
[20]	TORRA V, NARUKAWA Y. On hesitant fuzzy sets and decision[C]// Proceedings of 2009 IEEE International Conference on Fuzzy Systems.Jeju, Korea: IEEE, 2009:1378-1382.
[21]	刘小弟, 朱建军, 张世涛, 等. 考虑属性权重优化的犹豫模糊多属性决策方法[J]. 控制与决策, 2016, 31(2):297-302.
	LIU X D, ZHU J J, ZHANG S T, et al. A hesitant fuzzy multi-attribute decision making method considering attribute weight optimization[J]. Control and Decision Making, 2016, 31(2): 297-302. (in Chinese)
[22]	喻钧, 双晓. 仿造数码迷彩的设计方法[J]. 应用科学学报, 2012, 30(4):331-334.
	YU J, SHUANG X. Design method of imitation digital camouflage[J]. Journal of Applied Science, 2012, 30(4):331-334. (in Chinese)
[23]	XU Z S, XIA M M. Hesitant fuzzy entropy and cross-entropy and their use in multiattribute decision-making[J]. International Journal of Intelligent Systems, 2012, 27(9):799-822. doi: 10.1002/int.v27.9 URL
[24]	XU Z S, ZHANG X L. Hesitant fuzzy multi-attribute decision making based on TOPSIS with incomplete weight information[J]. Knowledge-Based Systems, 2013, 52:53-64. doi: 10.1016/j.knosys.2013.05.011 URL