基于全U网络的混凝土路面裂缝检测算法

针对现有的混凝土裂缝检测算法在各种复杂环境中检测精度不够、鲁棒性不强的问题,根据深度学习理论和U-net模型,提出一种全U型网络的裂缝检测算法。首先,依照原U-net模型路线构建网络;然后,在每个池化层后都进行一次上采样,恢复其在池化层之前的特征图规格,并将其与池化之前的卷积层进行融合,将融合之后的特征图作为新的融合层与原U-net网络上采样之后的网络层进行融合;最后,为了验证算法的有效性,在测试集中进行实验。结果表明,所提算法的平均精确率可达到83.48%,召回率为85.08%,F1为84.11%,相较于原U-net分别提升了1.48%,4.68%和3.29%,在复杂环境中也能提取完整裂缝,保证了裂缝检测的鲁棒性。     

Preparation,structural and photocatalytic activity of Sn/Fe co-doped TiO2 nanoparticles by sol-gel method

Pure and Sn/Fe co-doped (0.2 at.% Sn and 0.6 at.% Fe, 0.6 at.% Sn and 0.2 at.% Fe, 1.0 at.% Sn and 1.0 at.% Fe) TiO₂ nanoparticles were synthesized via a sol-gel method and subsequently calcined at different temperatures. Furthermore, the particles were analyzed by TG-DSC, XRD, TEM, HRTEM, EDS, SAED and UV–Vis for investigating the influences of dopant and calcination temperature on the thermal effect, composition, morphology, energy band gap (E_g) and the degradation efficiency of methyl orange (MO) under various light irradiations respectively. Results indicated that Sn/Fe co-doping inhibited the crystallization transformation from anatase to rutile phase of TiO₂ and decreased the E_g. The increased calcination temperature and Sn/Fe co-doped effect brought about the abnormal grain growth of TiO₂ nanoparticles. 0.6 at.% Sn/0.2 at.% Fe and 1.0 at.% Sn/1.0 at.% Fe co-doped TiO₂ nanoparticles presented better photocatalytic performance than pure and 0.2 at.% Sn/0.6 at.% Fe co-doped TiO₂ nanoparticles under visible light irradiation mainly due to the decreased E_g. On the contrary, 0.2 at.% Sn and 0.6 at.% Fe co-doped TiO₂ nanoparticles calcined at 650 °C showed the most excellent photocatalytic performance under UV light irradiation, which was about twice as large as that of pure TiO₂ possibly due to the formed hybrid structure of anatase and rutile phase as well as the h⁺-mediated decomposition pathway.     

Fabrication of alumina ceramics with functional gradient structures by digital light processing 3D printing technology

Alumina ceramics with different unit numbers and gradient modes were prepared by digital light processing (DLP) 3D printing technology. The side length of each functional gradient structure was 10 mm, the porosity ratio was controlled to 70%, and the number of units were (1 × 1 × 1 unit) and (2 × 2 × 2 unit) respectively. The different gradient modes were named FCC, GFCC-1, GFCC-2 and GFCC-3. SEM, XRD, and other characterization methods proved that these gradient structures of alumina ceramics had only α-Al2O3 phase and good surface morphology. The mechanical properties and energy absorption properties of alumina ceramics with different functional gradient structures were studied by compression test. The results show that the gradient structure with 1 × 1 × 1 unit has better mechanical properties and energy absorption properties when the number of units is different. When the number of units is the same, GFCC-2 and GFCC-3 gradient structures have better compressive performance and energy absorption potential than FCC structures. The GFCC-2 gradient structure with 1 × 1 × 1 unit has a maximum compressive strength of 19.62 MPa and a maximum energy absorption value of 2.72 × 105 J/m3. The good performance of such functional gradient structures can provide new ideas for the design of lightweight and compressive energy absorption structures in the future.     

基于推理注意力机制的二阶段网络图像修复

现有的图像修复方法在处理大面积缺失或高度纹理化的图像时，通常会产生扭曲的结构或与周围区域不一致的模糊纹理，无法重建合理的图像结构。为此，提出了一种基于推理注意力机制的二阶段网络图像修复方法。首先通过边缘生成网络生成合理的幻觉边缘信息，然后在图像补全网络完成图像的重建工作。为了进一步生成视觉效果更逼真的图像，提高图像修复的精确度，在图像补全网络采用推理注意力机制，有效控制了生成特征的不一致性，从而生成更有效的信息。所提方法在多个数据集上进行了实验验证，结果表明该图像修复方法的结构相似性指数达到了88.9%，峰值信噪比达到了25.56 dB，与现有的图像修复方法相比，该方法具有更高的图像修复精确度，生成的图像更逼真。     

Learn decision trees with deep visual primitives

In this paper, we strive to propose a self-interpretable framework, termed PrimitiveTree, that incorporates deep visual primitives condensed from deep features with a conventional decision tree, bridging the gap between deep features extracted from deep neural networks (DNNs) and trees’ transparent decision-making processes. Specifically, we utilize a codebook, which embeds the continuous deep features into a finite discrete space (deep visual primitives) to distill the most common semantic information. The decision tree adopts the spatial location information and the mapped primitives to present the decision-making process of the deep features in a tree hierarchy. Moreover, the trained interpretable PrimitiveTree can inversely explain the constituents of the deep features, highlighting the most critical and semantic-rich image patches attributing to the final predictions of the given DNN. Extensive experiments and visualization results validate the effectiveness and interpretability of our method.     

计算机网络办公自动化及安全策略探究

互联网通讯采取标准化模式主要以TCP/IP协议为载体,通讯的优越特性体现在同时具备便捷性与开放性,为办公提供很大的便利,但基于网络系统也会入侵病毒、也会给信息数据与办公体系安全性造成威胁,直接影响企业综合稳定发展。据此,为保障办公工作的顺利开展,本文对计算机网络办公自动化及安全策略进行详细分析。     

Dense and core-rim structured B4C-TiB2 ceramics with Mo-Co-WC additive

B₄C-TiB₂ ceramics (TiB₂ ranging 5~70 vol%) with Mo-Co-WC as the sintering additive were prepared by spark plasma sintering. In comparison with B₄C-TiB₂ without additive, the enhanced densification was evident in the sintered specimen with Mo-Co-WC additive. Core-rim structured grain was observed around TiB₂ grains. The interface of the rim between TiB₂ and B₄C phases demonstrated different feature: the inner borderline of the rim exhibited a smooth feature, whereas a sharp curved grain boundary was observed between the rim and the B₄C grain. The formation mechanism is discussed: the epitaxial growth of (Ti,Mo,W)B₂ rim around the TiB₂ core may occur as a result of the solid solution and dissolution-precipitation between TiB₂ phase and the sintering additive. It was revealed that the fracture toughness increased as the content of TiB₂ content increased, alongside the decreased hardness. B₄C-30 vol% TiB₂ specimen demonstrated the optimal combination of mechanical properties, reaching Vickers hardness of 24.3 GPa and fracture toughness of 3.33 MPa·m^1/2.     

Enhanced wide-activated residual network for efficient and accurate image deblocking

To save bandwidth and storage space as well as speed up data transmission, people usually perform lossy compression on images. Although the JPEG standard is a simple and effective compression method, it usually introduces various visually unpleasing artifacts, especially the notorious blocking artifacts. In recent years, deep convolutional neural networks (CNNs) have seen remarkable development in compression artifacts reduction. Despite the excellent performance, most deep CNNs suffer from heavy computation due to very deep and wide architectures. In this paper, we propose an enhanced wide-activated residual network (EWARN) for efficient and accurate image deblocking. Specifically, we propose an enhanced wide-activated residual block (EWARB) as basic construction module. Our EWARB gives rise to larger activation width, better use of interdependencies among channels, and more informative and discriminative non-linearity activation features without more parameters than residual block (RB) and wide-activated residual block (WARB). Furthermore, we introduce an overlapping patches extraction and combination (OPEC) strategy into our network in a full convolution way, leading to large receptive field, enforced compatibility among adjacent blocks, and efficient deblocking. Extensive experiments demonstrate that our EWARN outperforms several state-of-the-art methods quantitatively and qualitatively with relatively small model size and less running time, achieving a good trade-off between performance and complexity.     

Influence of α′‐/α‐crystal polymorphism on properties of poly(l‐lactic acid)

Poly(l ‐lactic acid) (PLLA) is a biodegradable and biocompatible thermoplastic polyester produced from renewable sources, widely used for biomedical devices, in food packaging and in agriculture. It is a semicrystalline polymer, and as such its properties are strongly affected by the developed semicrystalline morphology. As a function of the crystallization temperature, PLLA can form different crystal modifications, namely α′‐crystals below about 120 °C and α‐crystals at higher temperatures. The α′ modification is therefore of special importance as it may be the preferred polymorph developing at processing‐relevant conditions. It is a metastable modification which typically transforms into the more stable α‐crystals on annealing at elevated temperature. The structure, kinetics of formation and thermodynamics of α′‐ and α‐crystals of PLLA are reviewed in this contribution, together with the effect of α′‐/α‐crystal polymorphism on the properties of PLLA. © 2018 Society of Chemical Industry     

基于K-means和SOM的水下传感器数据采集方法

随着海洋资源勘探和海洋污染物监控工作的开展，水文数据的监测和采集等已经成为重要的研究方向。其中，水下无线传感器网络在水文数据采集过程中起着举足轻重的作用。本文研究的是水下无线传感器二维监测网络模型中，传感器节点数据采集的问题，其设计方法是通过自组织映射（Self-organizing mapping，SOM）对传感器节点进行路径最优化处理，结合优化的路径图形和K-means算法找到路径内部聚合点，利用聚合点和传感器的节点得到传感器通信半径内的数据采集点，最后通过SOM得到水下机器人（Autonomous underwater vehicle，AUV）到各个数据采集点采集数据的最优路径。经过实验验证，在水下1 200 m×1 750 m范围内布置52个传感器节点的情景下，数据采集点相比于传感器节点路径规划采用相同的采集顺序得到的路径优化了6.7%；对数据采集点重新进行自组织路径规划得到的路径比传感器结点路径的最优解提高了12.2%。增加传感器节点的数量，其结果也大致相同，因此采用该方法可以提高水下机器人采集数据的效率。