基于战场环境孪生的迷彩伪装动态效果评估方法

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

现有的迷彩伪装效果评估方法主要针对静态的单幅目标图像,不能真实地响应军事侦察中的动态多视角特性,相应的评估结果也缺乏全面性。为此提出一种基于战场环境孪生的迷彩伪装动态效果评估新方法。新方法以构建三维战场环境数字孪生模型为基础,借助三维仿真计算技术获取战场环境中迷彩伪装目标的动态多视角效果图,采用语义分割模型和目标跟踪模型识别分析场景中的迷彩运动目标;构建新的迷彩伪装评价指标体系及多侦察视角的多属性综合评估模型,实现不同迷彩方案在给定孪生战场环境中的伪装动态效果评价。以仿真军用卡车模型为实验对象,使用Unity 3D建模真实战场环境,以两种对比性迷彩伪装设计方案为例进行评估分析,实验结果表明新方法的合理性和实用性。

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.