Institutional Repository of Xi'an Institute of Optics and Precision Mechanics, CAS
Not a member yet
30565 research outputs found
Sort by
Feature Extraction Based on Linear Embedding and Tensor Manifold for Hyperspectral Image
No full text为了使降维结果更好地体现高光谱数据的空间结构信息,并进一步提高分类精度,提出了一种基于线性嵌入和张量流形的高光谱特征提取算法。不同于其他流形结构的表达方法,所提算法采用协同表示理论求解全局线性嵌入的权重矩阵,更有利于保持高维数据的全局信息,提高了流形结构表达的准确性。同时,建立了基于多特征描述的张量流形降维框架,得到的显式映射具有较强的可靠性和全局适应性。实验结果表明:与主成分分析、局部线性嵌入、拉普拉斯特征映射和线性保留投影等算法相比,所提算法表现出了更优越的分类性能。 </p
High-Power Single-Longitudinal-Mode DFB Semiconductor Laser Based on Sampled Moiré Grating
No full textIn this mymargin letter, we experimentally demonstrated a 1550-nm high-power single-longitudinal-mode (SLM) distributed feedback (DFB) semiconductor laser based on sampled moiré grating (SMG). By designing the sampling structure with micrometer scale, moiré grating (MG) can be equivalently realized along the laser cavity. Then, we can reduce the coupling coefficient near laser facet so as to increase the output power. The cavity length and ridge width of the fabricated laser are 1.0 mm and 3.0 μm, respectively. The measured threshold current and the slope efficiency are 30.0 mA and 0.36 mW/mA at the heat-sink temperature of 25 °C, respectively. When the injection current is 800.0 mA, the maximum output power is about 183.0 mW. The saturation power is significantly improved compared with conventional DFB laser with uniform sampled grating (USG), which was fabricated on the same wafer. In addition, a four-channel DFB laser array based on SMG was also fabricated on the same wafer, showing good wavelength uniformity. ? 2019 IEEE.</p
The Near-Space Wind and Temperature Sensing Interferometer: Forward Model and Measurement Simulation
No full textWind and temperature observation in near space has been playing an increasingly important role in atmospheric physics and space science. This paper reports on the near-space wind and temperature sensing interferometer (NWTSI), which employs a wide-angle Michelson interferometer to observe O-2(a(g)(1)) dayglow near 1.27 m from a limb-viewing satellite, and presents the instrument modeling and observation simulations from the stratosphere to the mesosphere and lower thermosphere. The characteristics of atmospheric limb-radiance spectra and line selection rules are described. The observational strategy of using two sets of three emission lines with a line-strength difference of one order of magnitude is proved to be suitable for extending altitude coverage. The forward modeling and measurement simulation of the expected NWTSI observations are provided, and the measurement uncertainty of the wind and temperature is discussed. The signal-to-noise ratio (SNR) and the limb-view weight work together to affect the precision of the wind and temperature measurements. The simulated results indicate a wind measurement precision of 1 to 3 m/s and a temperature precision of 1 to 3 K over an altitude range from 40 to 80 km, which meets the observing requirement in measurement precision for near-space detection.</p
Research on image denoising algorithm based on improved anisotropic diffusion synthetic aperture radar
No full textThe presence of speckle noise seriously affects the application of synthetic aperture radar (SAR) images in image fusion, so it is especially important to suppress speckle noise. According to the formation mechanism of speckle noise, this paper proposes a SAR image speckle noise removal algorithm based on improved anisotropic diffusion. The algorithm improves the diffusion coefficient c(x) based on the P-M equation diffusion filter algorithm, and adds the iterative termination condition, and obtains the filtering algorithm suitable for SAR images. This method can not only solve the problem that there are many isolated noise points in the traditional P-M model filtering, but also has a good effect on image edge preservation. The simulation results show that the improved P-M model can eliminate noise well and maintain the edge information of the image well. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.</p
Detection of ships in inland river using high-resolution optical satellite imagery based on mixture of deformable part models
No full textShip detection using optical satellite imagery is of great significance in many applications such as traffic surveillance, pollution monitoring, etc. So far, a lot of ship detection methods have been developed for images covering open sea, offshore area and harbors. Compared to the ship detection in sea and offshore area, it is more difficult to detect ships in inland river due to several challenges. First of all, many ships in inland river are clustered together and hard to be separated from each other. Secondly, ships lying alongside the pier are very likely to be recognized as part of the pier. Thirdly, ships in inland river is usually smaller than those in the sea. A hierarchical method is proposed to detect the ships in inland river in this paper. The Regions of Interest (ROIs) are firstly extracted based on water–land segmentation using multi-spectral information. Then two kinds of ship candidates are extracted based on the panchromatic band. The isolated ships are detected by analyzing the shape of connected components and the clustered ships are detected by using mixtures multi-scale Deformable Part Models (DPM) and Histogram of Oriented Gradient (HOG). At last, a Back Propagation Neural Network (BPNN) is trained to classify the ship candidates using the multi-spectral bands. The experiments using Quickbird satellite images show that our approach is effective in ship detection and performs particularly well in separating the ships clustered together and staying alongside the pier. © 2019 Elsevier Inc.</p
X-ray transmission characteristics and potential communication application in plasma region
No full textX射线具有波长短、光子能量高等特点,有望在等离子体环境中实现信息的有效传输.本文首先采用基于连续介质中的WKB分层法,研究了黑障条件下, X射线在非均匀等离子体鞘套中的透过率特性,仿真了不同等离子体电子密度和碰撞频率下X射线信号的透过率,理论上证明了X射线可用于黑障区信息传输的可行性.其次通过搭建环形扩散辉光放电等离子体发生器及实验验证系统,进行了国内外首次X射线穿过等离子体鞘套的验证实验.实验结果表明,等离子体对X射线信号的透过率存在一定程度的衰减,透过等离子体前后的X射线信号能谱轮廓相似度优于95.5%,能谱峰值点的偏移量小于1.3%.此外,在原有理论模型的基础上,考虑等离子中的粒子与X射线的碰撞、吸收效应,优化了X射线在等离子体中的透过率模型,与传统的理论方法相比,该模型可对实验现象进行更好的解释.同时计算了X射线在临近空间的透过率,并分析了X射线通信所能达到的潜在指标.这些结果有望为解决黑障区信号传输提供一定的理论与实验依据. </p
Excessively tilted fiber grating-based vector magnetometer
No full textA compact optic-fiber vector magnetometer is proposed and experimentally demonstrated, which is based on an excessively tilted fiber grating (Ex-TFG) assistant with the magnetic fluid (MF). Without any complicated processing, the cladding mode resonances of the bare Ex-TFG packaged by the MF show high sensitivity to slight perturbations by the magnetic field. Due to the excellent magneto-optical properties of the MF and the azimuth-dependent refractive index sensitivity of the Ex-TFG, such a magnetometer can achieve the magnetic field intensity sensitivity of 2.45 nm/mT and the orientation sensitivity of 0.41 nm/deg. In addition, based on the spectral interrogation, the detection limit of the magnetic field intensity could reach around 8.1 mu T at the minimum wavelength measurement accuracy of 0.02 nm. (C) 2019 Optical Society of America</p
Convolutional Edge Constraint-Based U-Net for Salient Object Detection
No full textThe salient object detection is receiving more and more attention from researchers. An accurate saliency map will be useful for subsequent tasks. However, in most saliency maps predicted by existing models, the objects regions are very blurred and the edges of objects are irregular. The reason is that the hand-crafted features are the main basis for existing traditional methods to predict salient objects, which results in different pixels belonging to the same object often being predicted different saliency scores. Besides, the convolutional neural network (CNN)-based models predict saliency maps at patch scale, which causes the objects edges of the output to be fuzzy. In this paper, we attempt to add an edge convolution constraint to a modified U-Net to predict the saliency map of the image. The network structure we adopt can fuse the features of different layers to reduce the loss of information. Our SalNet predicts the saliency map pixel-by-pixel, rather than at the patch scale as the CNN-based models do. Moreover, in order to better guide the network mining the information of objects edges, we design a new loss function based on image convolution, which adds an L1 constraint to the edge information of saliency map and ground-truth. Finally, experimental results reveal that our SalNet is effective in salient object detection task and is also competitive when compared with 11 state-of-the-art models.</p