Single image dehazing using a new color channel

Images with hazy scene suffer from low-contrast, which reduces the visible quality of the scene, thus making object detection a more challenging task. Low-contrast can result from foggy weather conditions during image acquisition. Dehazing is a process of removal of haze from the photography of a hazy scene. Single-image dehazing based on dark channel priors are well-known techniques in this field. However, the performance of such techniques is limited to priors or constraints. Moreover, this type of method fails when images have sky-region. So, a method is proposed, which can restore the visibility of hazy images. First, a hazy image is divided into blocks of size 32 × 32, then the score of each block is calculated to select a block having the highest score. Atmospheric light is calculated from the selected block. A new color channel is considered to remove atmospheric scattering, obtained channel value and atmospheric light are then used to calculate the transmission map in the second step. Third, radiance is computed using a transmission map and atmospheric light. The illumination scaling factor is adopted to enhance the quality of a dehazed image in the final step. Experiments are performed on six datasets namely, I-HAZE, O-HAZE, BSDS500, FRIDA, RESIDE dataset and natural images from Google. The proposed method is compared against 11 state-of-the-art methods. The performance is analyzed using fourteen quantitative evaluation metrics. All the results demonstrate that the proposed method outperforms 11 state-of-the-art methods in most of the cases.     

基于Word2Vec模型特征扩展的Web搜索结果聚类性能的改进

对于用户泛化和模糊的查询，将Web搜索引擎返回的列表内容聚类处理，便于用户有效查找感兴趣的内容。由于返回的列表由称为片段（snippet）的短文本组成，而传统的单词频率-逆文档频率（TF-IDF）特征选择模型不能适用于稀疏的短文本，使得聚类性能下降。一个有效的方法就是通过一个外部的知识库对短文本进行扩展。受到基于神经网络词表示方法的启发，提出了通过词嵌入技术的Word2Vec模型对短文本扩展，即采用Word2Vec模型的TopN个最相似的单词用于对片段（snippet）的扩展，扩展文档使得TF-IDF模型特征选择得到聚类性能的提高。同时考虑到通用性单词造成的噪声引入，对扩展文档的TF-IDF矩阵进行了词频权重修正。实验在两个公开数据集ODP239和SearchSnippets上完成，将所提方法和纯snippet无扩展的方法、基于Wordnet的特征扩展方法和基于Wikipedia的特征扩展方法进行了对比。实验结果表明，所提方法在聚类性能方面优于对比方法。     

基于时空的弱点状运动目标检测技术

如何在复杂背景下持续有效地检测目标位置,一直是研究者们需要面对的主要挑战。本文在研究红外点状移动目标特征的基础上,根据目标无纹理,无形状的特性,提出一种改进的形态学目标增强算法,并利用目标连续时空不变性检测目标。首先,建立多尺度的图像金字塔,在每层上采用改进的形态学算法快速、粗糙定位小目标。然后进一步的根据目标在时空上的位置相关性,提出基于目标运动特征分析的精确检测方法。得到精确稳定的检测结果。最后实验结果表明与经典的形态学检测算法及其他算法相比,该技术能更有效地检测弱小目标,具有更高的鲁棒性。     

Torsional statics of two-dimensionally functionally graded microtubes

A size-dependent governing equation is derived to investigate the torsional static behaviors of two-dimensionally functionally graded microtubes based on the modified couple stress theory. The shear modulus is assumed to vary along the tube’s length direction according to an exponential distribute function, and varies along the tube’s radius direction according to a power-law function. A generalized differential quadrature method is developed to determine the rotational angle and shear stresses. Some illustrative examples are given to investigate the effects of applied torques, the length scale parameter and various material compositions on the torsional angle and shear stresses.     

Analysis of Cerebrospinal Fluid Extracellular Vesicles by Proximity Extension Assay: A Comparative Study of Four Isolation Kits

There is a lack of reliable biomarkers for disorders of the central nervous system (CNS), and diagnostics still heavily rely on symptoms that are both subjective and difficult to quantify. The cerebrospinal fluid (CSF) is a promising source of biomarkers due to its close connection to the CNS. Extracellular vesicles are actively secreted by cells, and proteomic analysis of CSF extracellular vesicles (EVs) and their molecular composition likely reflects changes in the CNS to a higher extent compared with total CSF, especially in the case of neuroinflammation, which could increase blood–brain barrier permeability and cause an influx of plasma proteins into the CSF. We used proximity extension assay for proteomic analysis due to its high sensitivity. We believe that this methodology could be useful for de novo biomarker discovery for several CNS diseases. We compared four commercially available kits for EV isolation: MagCapture and ExoIntact (based on magnetic beads), EVSecond L70 (size-exclusion chromatography), and exoEasy (membrane affinity). The isolated EVs were characterized by nanoparticle tracking analysis, ELISA (CD63, CD81 and albumin), and proximity extension assay (PEA) using two different panels, each consisting of 92 markers. The exoEasy samples did not pass the built-in quality controls and were excluded from downstream analysis. The number of detectable proteins in the ExoIntact samples was considerably higher (~150% for the cardiovascular III panel and ~320% for the cell regulation panel) compared with other groups. ExoIntact also showed the highest intersample correlation with an average Pearson’s correlation coefficient of 0.991 compared with 0.985 and 0.927 for MagCapture and EVSecond, respectively. The median coefficient of variation was 5%, 8%, and 22% for ExoIntact, MagCapture, and EVSecond, respectively. Comparing total CSF and ExoIntact samples revealed 70 differentially expressed proteins in the cardiovascular III panel and 17 in the cell regulation panel. To our knowledge, this is the first time that CSF EVs were analyzed by PEA. In conclusion, analysis of CSF EVs by PEA is feasible, and different isolation kits give distinct results, with ExoIntact showing the highest number of identified proteins with the lowest variability.     

Excitonic Properties of Chemically Synthesized 2D Organic–Inorganic Hybrid Perovskite Nanosheets

2D organic–inorganic hybrid perovskites (OIHPs) represent a unique class of materials with a natural quantum‐well structure and quasi‐2D electronic properties. Here, a versatile direct solution‐based synthesis of mono‐ and few‐layer OIHP nanosheets and a systematic study of their electronic structure as a function of the number of monolayers by photoluminescence and absorption spectroscopy are reported. The monolayers of various OIHPs are found to exhibit high electronic quality as evidenced by high quantum yield and negligible Stokes shift. It is shown that the ground exciton peak blueshifts by ≈40 meV when the layer thickness reduces from bulk to monolayer. It is also shown that the exciton binding energy remains effectively unchanged for (C₆H₅(CH₂)₂NH₃)₂PbI₄ with the number of layers. Similar trends are observed for (C₄H₉NH₃)₂PbI₄ in contrast to the previous report. Further, the photoluminescence lifetime is found to decrease with the number of monolayers, indicating the dominant role of surface trap states in nonradiative recombination of the electron–hole pairs.     

亚磷酸酯功能化聚硅氧烷扩链聚乳酸的阻燃性能研究

将聚氨基环氧基硅氧烷（PSQ）以及亚磷酸酯功能化的聚氨基环氧基硅氧烷（PPSQ）与聚乳酸（PLA）通过熔融共混制备了扩链PLA。研究了亚磷酸酯基团的引入对PSQ在PLA中的扩链和阻燃性能的影响，探究PPSQ在PLA中的阻燃机理。结果表明，相比于PSQ，PPSQ可以明显提高PLA的分子量和复合黏度，具有更好的扩链作用；PPSQ的引入可以明显改善PLA的阻燃性能，使得PLA?PPSQ的极限氧指数提高至25.4 %，比纯PLA提升了31.6 %，且使得PLA的热释放速率峰值和总热释放量分别下降了18.1 %和16.6 %。分析可知，由于亚磷酸酯基团的引入，促使PLA?PPSQ中的硅元素在燃烧过程中向炭层表面迁移形成更多富含碳化硅和氧化硅的致密炭层，起到隔热隔氧的作用，进而发挥优异的阻燃性能。     

考虑变幂数的畸变动力学相似试验模型设计方法及试验研究

动力学相似理论广泛应用于大型结构的振动试验，其中畸变相似用于解决结构的设计参数不满足等比例缩放时的相似问题，针对畸变相似关系中幂数为定值导致预测精度低的问题，提出考虑变幂数的畸变动力学相似试验模型设计方法。首先，在敏感性分析与中心差分理论推导的幂数表达式的基础上，将中心差分的范围不断扩大得到一组随相似比变化的幂数。采用最小二乘法将这组幂数拟合为关于相似比的函数并建立相似关系。然后，在多盘转子系统的数值算例中将变幂数法与文献中的两种方法进行对比，结果表明，变幂数法的预测精度较文献中的方法有很大提高。最后，通过试验研究验证了数值算例中动力学模型及临界转速求解的正确有效性。相似关系的建立是以模型的临界转速为基础，因此也验证了数值算例中的相似预测结果。     

煤矿井下小断层延伸长度定量化预测研究

以山西A矿开采的3号煤层采掘活动已揭露的小断层资料为基础,选取断层落差、倾角、延伸长度3个参数,通过多线性回归方法构建NW、NE 2个方向的断层延伸长度预本构模型,并进行现场应用。结果表明,构建的小断层预测模型预测的小断层延伸长度的准确率在90%以上,可以为后续揭露小断层延伸长度的确定提供可靠的参数指导。     

Output feedback tracking control for rigid body attitude via immersion and invariance angular velocity observers

A novel immersion and invariance (I&I) angular velocity observer is presented for the attitude tracking control of a rigid body with the lack of angular rate. Global exponential convergence of angular velocity estimate errors are guaranteed by an innovative filter design for the estimates' Euclidean norm. The proposed method requires fewer filter states compared with existing I&I angular velocity observer designs, which achieves a simpler closed-loop structure (dynamic reduction). The observer synthesis and convergence are independent of the control torque, which leads to much convenience in establishing “separation property” when combining a proportional-derivative attitude tracking controller driven by angular velocity estimates. A rigorous stability analysis is provided to ensure the (almost) global asymptotic convergence of the overall closed-loop tracking errors, and several numerical simulations are carried out to demonstrate the effectiveness of the combined implementation of proposed angular velocity observer and full-state feedback attitude tracking controller.