由于偏振特性是材料自身所决定的物理特性, 其偏振图像含有丰富的目标信息, 利用偏振信息对目标进行识别一直是国内外目标探测领域的研究热点, 而主动偏振成像较之被动偏振成像更具有信噪比高以及可控性好等优势。 在详细分析了偏振菲涅尔反射比分布的理论基础上, 提出了一种利用探测物体表面的偏振菲涅尔反射比的主动偏振成像方法。 该方法在发射端将偏振方向正交的两种偏振态的光源交替照射到目标场景中, 在探测端用分别装有两个偏振方向垂直的偏振片的CCD采集偏振图像。 同时, 将探测端架构在不同的探测方向采集目标经主动光源照射后的偏振数据, 最后将这些数据传输到计算中心, 通过最优化拟合技术反演出不同目标的光学常数, 由于不同目标的表面材质不同, 其反映出的光学常数就不同, 从而达到辨识不同材质目标的目的。 实验分别采用了仿真数据和实测数据来验证该方法的有效性。 仿真实验显示, 所提出的方法利用材料的光学常数对不同材料进行区分不仅是科学的而且更方便有效。 实测数据进一步验证了该方法能够较好的恢复出目标的相关光学常数, 尤其在区分金属和非金属材料方面表现突出, 并且探测方法结构简单实用, 在目标探测、 伪装识别等领域中有较大应用前景。

As the polarization characteristics are the physical property determined by the material itself, its corresponding polarization image contains abundant target’s information. Using polarization information to identify the target is always a hot research topic in the field of the target detection. Active polarization imaging has more advantages compared with passive polarization imaging because of its high signal-to-noise ratio and good controllability. In this paper, based on the detailed analysis of the theory of the distribution of polarization Fresnel reflectance ratio, a kind of active polarization imaging method is proposed with detecting the polarization Fresnel ratio of the surface of the object. The proposed method adopts two kind of polarization light with orthogonal polarization direction at the light emission part to exposure to the target scenario alternately. Then two cameras side-by-side at the detecting part respectively equipped with two orthogonal polarization direction filters to capture the polarization images. Meanwhile, the detectors are placed in different detecting direction to acquire the polarization imaging with active polarization light source illuminating. Finally, with transmitting the data to the calculating center, optical constants can be recovered from the polarization data by the optimization fitting technique. Because the materials of target’s surface are different, the corresponding optical constants are different. Then the purpose of discriminating the targets with different materials is achieved. The simulated and actual measured experiments are explored to verify the effectiveness of the proposed method. Simulation experiment shows it is not only scientific but also more convenient and effective in that the proposed method can distinguish the different materials using the calculated optical constants. The actual measured data further shows that the method is able to do better in recover optical constants of targets, especially in the distinction between metal and dielectric materials. Furthermore, the system has great application prospect in the field of target detection and camouflage recognition with its simple structure and practicability.