Institutional Repository of Xi'an Institute of Optics and Precision Mechanics, CAS
Not a member yet
30565 research outputs found
Sort by
Study on aerodynamic optimization design method analysis and control algorithm of the pod
No full textQuadruped Animal Moving Target Extraction and Background Reconstruction Algorithm
No full textWith the rapid development of computer technology, video surveillance systems are spread all over the streets, bringing convenience to people's lives. Video is also becoming more and more intelligent. Among them, the identification, tracking and extraction of moving targets are the challenges faced by intelligent monitoring systems. Based on the movement of the quadruped, this paper proposes an algorithm for the extraction and background reconstruction of the quadruped animal Firstly, the moving target of the video sequence is detected by combining the background subtraction method and the interframe difference method, and then the target tracking is performed based on the moving target feature. In the extraction of moving targets, an algorithm for animal moving target extraction based on spatiotemporal saliency is proposed, and the background reconstruction based on probability and statistics theory is used. Through the experiments of two similar quadruped animals, such as dogs and wolves, the proposed method has better anti-interference and better results in the recognition and extraction of moving targets.</p
3G structure for image caption generation
No full textIt is a big challenge of computer vision to make machine automatically describe the content of an image with a natural language sentence. Previous works have made great progress on this task, but they only use the global or local image feature, which may lose some important subtle or global information of an image. In this paper, we propose a model with 3-gated model which fuses the global and local image features together for the task of image caption generation. The model mainly has three gated structures. (1) Gate for the global image feature, which can adaptively decide when and how much the global image feature should be imported into the sentence generator. (2) The gated recurrent neural network (RNN) is used as the sentence generator. (3) The gated feedback method for stacking RNN is employed to increase the capability of nonlinearity fitting. More specially, the global and local image features are combined together in this paper, which makes full use of the image information. The global image feature is controlled by the first gate and the local image feature is selected by the attention mechanism. With the latter two gates, the relationship between image and text can be well explored, which improves the performance of the language part as well as the multi-modal embedding part. Experimental results show that our proposed method outperforms the state-of-the-art for image caption generation. ? 2018</p
Muti-stage learning for gender and age prediction
No full textAutomatic gender and age prediction has become relevant to an increasing amount of applications, particularly under the rise of social platforms and social media. However, the performances of existing methods on real-world images are still not satisfactory as we expected, especially when compared to that of face recognition. The reason is that, facial images for gender and age prediction have inherent small inter-class and big intra-class differences, i.e., two images with different skin colors and same age category label have big intra-class difference. However, most existing methods have not constructed discriminative representations for digging out these inherent characteristics very well. In this paper, a method based on muti-stage learning is proposed: The first stage is marking the object regions with an encoder-decoder based segmentation network. Specifically, the segmentation network can classify each pixel into two classes, "people" and others, and only the "people" regions are used for the subsequent processing. The second stage is precisely predicting the gender and age information with the proposed prediction network, which encodes global information, local region information and the interactions among different local regions into the final representation, and then finalizes the prediction. Additionally, we evaluate our method on three public and challenging datasets, and the experimental results verify the effectiveness of our proposed method. (C) 2019 Elsevier B.V. All rights reserved.</p
Spectral Tensor Synthesis Analysis for Hyperspectral Image Spectral-Spatial Feature Extraction
No full textFeature extraction is a preprocessing step for hyperspectral image classification. Principal component analysis only uses the spectral information, but it does not use spatial information of a hyperspectral image. Both spatial and spectral information are used when hyperspectral image is modelled as tensor, that is, decreasing the noise on spatial dimension and reducing the dimension on a spectral dimension at the same time. However, in this model, a hyperspectral image is modelled only as a data cube. The factors affecting the spectral features of ground objects is not considered and these factors are barely distinguished. This means that further improving classification is very difficult. Therefore, a new model on hyperspectral image is proposed by the authors. In the new model, many factors that impact the spectral features of ground objects are synthesized as the within-class factor. The within-class factor, the class factor and the pixel spectral are selected as a mode, respectively. The pixel spectrals in the training set are modelled as a third-order tensor. The experiment results indicate that the new method improves the classification compared with the previous methods.</p
A Fast Lossless Data Compression Method for the Wedge Filter Spectral Imager
No full text楔形滤光片型光谱成像仪具有无运动部件、低光机复杂度等优点,是低成本微型化光谱成像仪的一个重要发展方向。不同于传统色散型光谱成像仪,楔形滤光片型光谱成像仪获取的数据是光谱-空间混合调制的图像。针对直接应用CCSDS123进行楔形滤光片型光谱成像仪数据压缩时压缩比较低的问题,结合楔形滤光片型光谱成像仪"谱像混合"、"推扫成谱"的特点,通过定义新的局部差向量,构建了一种低运算复杂度适合硬件实现的快速无损压缩方法 WCCSDS123。新的局部差向量中参与计算的像元集合代表的是同一被观测点的光谱信息。WCCSDS123方法首先利用局部和与改进的局部差向量对采样点的值进行预测,再利用预测值与真实值计算预测残差并对其进行整数映射,最后采用采样自适应熵编码对映射预测残差进行编码完成压缩。在6组楔形滤光片型光谱成像仪数据上分别采用WCCSDS123和CCSDS123进行了压缩实验。实验结果表明,与CCSDS123相比,WCCSDS123的压缩比提高了约21.62%,压缩耗时没有明显差异。因此,该方法在提高压缩比同时,继承了CCSDS123复杂度低,易于硬件实现的优点。该方法 WCCSDS123具有较低的计算复杂度,能够更加有效地利用空间光谱冗余信息,获得更好的压缩效果,是针对楔形滤光片型光谱成像仪的一种良好的快速无损数据压缩方法。</p
Hybrid phase-amplitude superoscillation element for nonscanning optical superresolution imaging
No full textIn this paper, we report a nonscanning optical superresolution imaging method based on a hybrid phaseamplitude superoscillation element. Using the Chebyshev polynomials as a basis set on the superoscillation waveform, the optimal combination of these, representing the optimal focal-spot in the local field of view, is found by genetic algorithm. Our numerical calculations demonstrate that a subwavelength focal spot with a full width at half-maximum as small as 253 nm is realized, which has more than 30 times improvement in sidelobe suppression ratio, and crucially, a greatly extended needle with continuously shrunken focal spot is yielded, which allows a large imaging tolerance in the axial displacement of the object. We then present our simulated results of the superresolution imaging on sparse point objects and continuous objects, where the practicality and effectiveness of this method are analyzed and discussed in detail. (c) 2019 Optical Society of America.</p
On-chip polarization beam splitter with multimode operation in a SOI waveguide
No full textThe polarization beam splitter (PBS) based on the directional coupler is numerically investigated to operate transverse-electric (TE) and transverse-magnetic (TM) modes including high-order modes. Based on the phase-matching condition, all TM-polarization modes are switched from the bus waveguide to the adjacent waveguide without changing the spatial profile of the electric field and mode order while all TE-polarization modes go along the bus waveguide without significant coupling. The results show a low cross talk characteristic (>20 dB at 1.55 mu m) and big fabrication tolerance (delta w = +/- 20 nm). It becomes an essential supplement for polarization splitting with multimode operation in photonics integrated circuits. (C) 2018 The Japan Society of Applied Physics</p
Research on key technology of the optical measurement device used in liquid oxygen tank of carrier rocket
No full textFor some extreme environment, such as vibration and shock during rocket launching or ultralow temperature in tank of cryogenic propellant, some key technology of the optical measurement device used in liquid oxygen tank of carrier rocket have been researched. In order to obtain a light-weight, high-stiffness and large-heat resistance main board structure for optical measurement device, a thermo-mechanical topology optimization technique is introduced into the main board structure design of optical measurement device for improving the designing quality. Compared with that before the optimization the main board structure reduces weight loses 46.8 percent with the first-order natural frequency above 420 Hz. The numerical results indicate that after adopting the topologic optimization design method, not only the design course is shortened, but also the main board structure weight is effectively reduced, heat resistance is effectively increased and the capability of the main board structure is enhanced. The extreme temperature test results show that the method is effective. © 2019 SPIE.</p