Multi-dimensional camouflage against VIS-NIR hyperspectral, MIR intensity and polarization imaging

Under the impact of big data and artificial intelligence, optical detection techniques based on single physical quantities can no longer meet the needs of acquiring multi-dimensional information in complex environments. Therefore, multi-dimensional detection methods, such as hyperspectral imaging and polarization imaging are developing rapidly, placing higher demands on modern stealth technology. However, traditional stealth technologies are mainly designed for a single dimension, such as achieving visible camouflage by simulating background color, or evading infrared thermal imaging detection by controlling surface temperature and emissivity. And these methods can be easily recognized under synergistic detection of multi-dimensional imaging due to their spectral or polarization characteristics.

 

“The ‘eyes’ on the battlefield are becoming increasingly sensitive and diverse,” the author of the paper introduced. “Traditional camouflage are often designed for only one type of ‘eye.’ Therefore, developing multi-dimensional camouflage technology that can deceive multiple detection methods at the same time has become an urgent need to improve survivability.”

 

To tackle this critical issue, a team led by Professor Qiang Li from the State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, China, proposed a novel camouflage strategy for multi-dimensional detection across spectrum, intensity and polarization in a new paper published in Light: Science & Applications. Based on rough surface, silver nanowires (AgNWs) and biometric coatings, they have developed a hierarchical structure consisting of three functional layers from top to bottom: a top polarization camouflage layer, a middle intensity camouflage layer, and a bottom hyperspectral camouflage layer. The top structure is composed of a rough-surfaced polyethylene (PE) film. By introducing a large number of micron-scale disordered structures, the diffuse scattering effect of infrared light is enhanced, and the polarization effect caused by specular reflection is reduced, thereby achieving low degree of linear polarization (DoLP) across a wide angle range. Furthermore, the spectral transparency of the PE structure from the visible to mid-infrared ensures functional decoupling of the top layer without disturbing the functions below. For the middle camouflage layer, a transparent conductive layer composed of silver nanowires is employed. By controlling the diameter and density of the silver nanowires, the constructed metal wire network can act as a metallic reflective layer in the infrared region, effectively reducing the device's thermal emissivity while maintaining high transparency in the visible-near-infrared band, allowing the transmission of spectral camouflage signals from the bottom. For the bottom layer, a biomimetic coating material based on a composite system of Cr₂O₃, MgCl₂·6H₂O and waterborne polyurethane is used. This system utilizes the simulated green reflection peak and steep slope characteristics of vegetation in the near-infrared band created by Cr₂O₃, combined with the water-retention properties of MgCl₂·6H₂O to simulate the absorption characteristics of near-infrared water in vegetation, thus achieving accurate simulation of the vegetation's spectral characteristics.

 

To verify the optical properties of this structure, the research team fabricated multi-dimensional camouflage devices utilizing hot pressing and blade-coating processes, and characterized their visible-near-infrared (VIS-NIR) and mid-infrared (MIR) optical properties (Figures 2c and 2d). On the one hand, VIS-NIR diffuse spectroscopy revealed that the PE-AgNWs structure exhibits a high transmittance of approximately 0.8, enabling compatibility with any VIS-NIR camouflage technique beneath. And the whole device effectively simulated the spectral characteristics of vegetation ("green peaks," "red edges," "near-infrared plateaus," and "water absorption valleys"). On the other hand, MIR emissivity measurements showed that the device possesses an emissivity of approximately 0.7 in the long-wave infrared (LWIR, 8-14 μm) band, effectively camouflaging ground targets with a temperature around 60°C. Furthermore, the structure can further reduce infrared emissivity to 0.3 at the cost of near-infrared camouflage performance using indium tin oxide (ITO). Moreover, the research team characterized the angular polarization characteristics of the device and some common materials using infrared polarization imaging (Figure 2e). The device exhibited low linear polarization (below 1.5%) across a wide angular range of 0°–85°, indicating excellent infrared polarization camouflage performance.

 

To further verify its camouflage performance in practical applications, the research team placed the device and a vehicle model on the grass and used a hyperspectral camera and an infrared polarization camera to simulate multi-dimensional detection of ground targets. It can be seen that under the visible-near-infrared hyperspectral camera, the spectral similarity between the device and vegetation reached over 96% (Figure 3b); and in the spectral angle matching classification algorithm (Figure 3c), the device effectively deceived the classification algorithm throughout the 400-2500nm band, disguising itself as a vegetation background similar to leaves and grass. In the observation with the infrared polarization camera (Figure 3d), for ground targets heated by an internal heat source at 80℃, the area covered by the device not only blended well into the low-emissivity background in the infrared intensity image, but also achieved a good camouflage effect in the polarization image with a DoLP far lower than that of the ITO sample.

 

“This work provides feasibility verification for polarization camouflage.” the researchers added. “it also lays a foundation for countering multimodal imaging and multidimensional coordinated control of electromagnetic waves.”

---

---