数码迷彩立体动态成像算法研究  

Research on 3D Dynamic Imaging Algorithm for Digital Camouflage

作　　者：武国晶 吕绪良[1] 许卫东[1] 李凌[1] 翁枭迪[1] 

机构地区：[1]解放军理工大学野战工程学院,江苏南京210007

出　　处：《兵工学报》2016年第1期103-108,共6页Acta Armamentarii

摘　　要：现有数码迷彩图案的平面性和固定不变性使之难以对抗未来高技术的侦察监测,针对这个问题,提出了一种基于柱透镜光栅的数码迷彩立体动态成像算法。该算法充分考虑人双目立体成像的生理特征和柱透镜光栅分光分像的光学特性,阐述数码迷彩立体动态化的可行性。在此基础上,推导出迷彩立体动态成像的视差设计原理并根据背景深度信息将数码迷彩图像分为多幅迷彩深度序列图。利用梳状函数对迷彩深度序列图进行采样合成,最终生成一幅光栅图像,用来裱贴于光栅板后即可实现立体动态数码迷彩。对该算法有效性进行了仿真实验并加以分析。The current digital camouflage is not able to withstand the future high-tech reconnaissance and monitoring due to its planarity and inalterability. To address the issue, a 3D dynamic imaging algorithm based on lenticular raster is proposed. The proposed algorithm fully takes the biological characteristics of human binocular stereo imaging and the optical characteristics of splitting light and imaging of lenticular raster into account to describe the feasibility of 3D dynamic digital camouflage. On this basis, the paral- lax design principle of 3D dynamic imaging for digital camouflage is deduced, and the multiple sequence diagrams are generated according to depth information of background. The comb function is used to sam- ple and synthesize the sequence diagrams. The resultant raster image could be mounted on the back of a grating plate to achieve 3D dynamic imaging. The simulation results show that the proposed algorithm can be used for the 3D dynamic imaging of digital camouflage.

关 键 词：兵器科学与技术 伪装 数码迷彩 立体动态成像 柱透镜光栅 视差 

分 类 号：E951.4[军事—军事工程]