RESUMEN
The stability of the moving mirror of a Michelson Fourier transform spectrometer (M-FTS) has a non-negligible influence on its spectral quality, which limits its application. We proposed a spectrometer scheme with a pair of rotating parallel mirrors (RPM-FTS), which has advantages of fast response and high stability. The influence of the parallelism error of parallel mirrors on interference was analyzed by establishing a rotation vector model between the parallelism error, rotation angle, and optical path. The modulation depth of the RPM-FTS is more insensitive with the same installation error of the M-FTS; thus, more spectral details can be displayed easily.
RESUMEN
Motion blur recovery is a common method in the field of remote sensing image processing that can effectively improve the accuracy of detection and recognition. Among the existing motion blur recovery methods, the algorithms based on deep learning do not rely on a priori knowledge and, thus, have better generalizability. However, the existing deep learning algorithms usually suffer from feature misalignment, resulting in a high probability of missing details or errors in the recovered images. This paper proposes an end-to-end generative adversarial network (SDD-GAN) for single-image motion deblurring to address this problem and to optimize the recovery of blurred remote sensing images. Firstly, this paper applies a feature alignment module (FAFM) in the generator to learn the offset between feature maps to adjust the position of each sample in the convolution kernel and to align the feature maps according to the context; secondly, a feature importance selection module is introduced in the generator to adaptively filter the feature maps in the spatial and channel domains, preserving reliable details in the feature maps and improving the performance of the algorithm. In addition, this paper constructs a self-constructed remote sensing dataset (RSDATA) based on the mechanism of image blurring caused by the high-speed orbital motion of satellites. Comparative experiments are conducted on self-built remote sensing datasets and public datasets as well as on real remote sensing blurred images taken by an in-orbit satellite (CX-6(02)). The results show that the algorithm in this paper outperforms the comparison algorithm in terms of both quantitative evaluation and visual effects.
Asunto(s)
Algoritmos , Procesamiento de Imagen Asistido por Computador , Procesamiento de Imagen Asistido por Computador/métodos , Movimiento (Física)RESUMEN
Two types of ultra-broadband long wavelength infrared (LWIR) absorbers with small period and super thin thickness are designed. The absorption with high absorptivity and large bandwidth is achieved through combined propagating and localized surfaced plasmon resonances. We first design a three-layer absorber with a Ti-Ge-Ti configuration, the period of the structure is only 1.4 µm (nearly 1/8 of the center wavelength), the thickness of its dielectric is only 0.5 µm (1/22 of the center wavelength), and the average absorption is 87.9% under normal incident from 8µm to 14µm. Furthermore, the four-layer absorber with a Ti-Ge-Si3N4-Ti configuration is designed to obtain more average absorption increasing to 94.5% from 8 µm to 14µm under normal incident, the period of the structure increases to 1.6 µm and the total thickness of dielectric increases to 0.6µm. The proposed absorber is polarization-independent and possesses a good tolerance of incident angle. We calculate that the average absorption of the four-layer absorber for both TE- and TM-modes still exceeds 90% up to an incident angle of θ = 40° (90.7% for TE-mode, 91.9% for TM-mode), and exceed 80% up to an incident angle of θ = 60° (80.2% for TE-mode, 82.1% for TM-mode).
RESUMEN
A dual-band metamaterial absorber based on local surface plasmon resonance is designed, which is composed of a periodic arrangement of stacked nanodisk structures. The structure unit consists of two dielectric layers and three metal layers. Based on the finite difference time domain method, under the condition of vertically incident plane light, two absorption peaks in the mid-wave infrared and long-wave infrared (MWIR/LWIR) are obtained, and the absorption is greater than 98%. The absorber has good incident state tolerance characteristics. We can modulate the MWIR/LWIR absorption peaks by changing the radius of the stacked disk structure, and MWIR and LWIR dual-band broadband absorption can be achieved by integrating different size elements in the plane. The average absorption is 71% for MWIR with 1.1 µm bandwidth from 3.2 to 4.3 µm and 88% for LWIR with 3 µm bandwidth from 8.5 to 11.5 µm. At the same time, the structure also has effective refractive index (RI) sensitivity characteristics. In the RI range of 1.8-2, the maximum RI sensitivity of the LWIR and the MWIR is 1085 nm/refractive index unit (RIU) and 1472 nm/RIU, respectively.