一种基于目标梯度特征和轨迹预测的复杂云背景目标检测算法

对于复杂云层背景下的红外弱小目标检测跟踪一直是图像处理的研究热点。受复杂云层和强光等因素的干扰,目标很容易淹没在背景中。针对传统方法无法兼顾较高的目标检测概率和较低的目标检测虚警率的问题,提出了一种目标检测跟踪方法。通过梯度预滤波的方法获取初始目标,二次比对的方法获取真实目标;通过目标航迹建立与删除的方法降低虚警率;通过抗云层遮挡卡尔曼滤波跟踪的方法提升目标检测能力。实验结果表明,该技术不仅很好地抑制了复杂云层背景和强光干扰背景,而且通过目标建航和抗遮挡跟踪等方法极大提升了目标检测概率,降低了虚警率。     

光谱成像技术在海上目标探测识别中的应用探讨

光谱成像可同时获得光谱和空间信息,这使得光谱成像技术在解决复杂海洋环境背景下目标探测与识别时表现出独特的优势,已成为光电目标检测领域的一个前沿研究方向。对光谱成像技术的原理、特点进行了讨论并重点列举了该技术在探测领域的应用;介绍了光谱成像技术在海上目标探测识别中的研究进展,并对光谱成像技术在该领域的应用价值进行了探讨。     

Dielectric layer-dependent surface plasmon effect of metallic nanoparticles on silicon substrate

The electromagnetic interaction between Ag nanoparticles on the top of the Si substrate and the incident light has been studied by numerical simulations. It is found that the presence of dielectric layers with different thicknesses leads to the varied resonance wavelength and scattering cross section and consequently the shifted photocurrent response for all wavelengths. These different behaviours are determined by whether the dielectric layer is beyond the domain where the elcetric field of metallic plasmons takes effect, combined with the effect of geometrical optics. It is revealed that for particles of a certain size, an appropriate dielectric thickness is desirable to achieve the best absorption. For a certain thickness of spacer, an appropriate granular size is also desirable. These observations have substantial applications for the optimization of surface plasmon enhanced silicon solar cells.