基于DT-CWT和SVM的踏面旋转不变纹理提取算法

针对列车车轮踏面旋转纹理信息无法准确、有效提取的问题,提出一种基于Radon变换和双树复小波变换（DT-CWT）的列车车轮踏面特征提取方法。首先,对车轮踏面图像进行Radon变换;然后,对变换后的图像进行DT-CWT分解,使用分解后的各层低频子带系数和高频子带系数模的均值和标准方差构造特征向量,将其作为区分列车车轮踏面是否发生损伤的依据;最后,由支持向量机（SVM）进行分类决策。使用动车所采集的图像及人为加噪声后的图像进行分类实验,结果表明,本文使用的Radon和DT-CWT算法能有效地进行旋转不变纹理的提取,SVM分类正确率可以达到95%,可为列车车轮踏面状况检测提供更为准确便捷的方法支撑。     

改进U-Net型网络的遥感图像道路提取

目的 遥感图像道路提取在城市规划、交通管理、车辆导航和地图更新等领域中发挥了重要作用,但遥感图像受光照、噪声和遮挡等因素以及识别过程中大量相似的非道路目标干扰,导致提取高质量的遥感图像道路有很大难度。为此,提出一种结合上下文信息和注意力机制的U-Net型道路分割网络。方法 使用Resnet-34预训练网络作为编码器实现特征提取,通过上下文信息提取模块对图像的上下文信息进行整合,确保对道路的几何拓扑结构特征的提取;使用注意力机制对跳跃连接传递的特征进行权重调整,提升网络对于道路边缘区域的分割效果。结果 在公共数据集Deep Globe道路提取数据集上对模型进行测试,召回率和交并比指标分别达到0.847 2和0.691 5。与主流方法U-Net和CE-Net（context encoder network）等进行比较,实验结果表明本文方法在性能上表现良好,能有效提高道路分割的精确度。结论 本文针对遥感图像道路提取中道路结构不完整和道路边缘区域不清晰问题,提出一种结合上下文信息和注意力机制的遥感道路提取模型。实验结果表明该网络在遥感图像道路提取上达到良好效果,具有较高的研究和应用价值。     

弱监督深度语义分割网络的多源遥感影像水体检测

目的 深度语义分割网络的优良性能高度依赖于大规模和高质量的像素级标签数据。在现实任务中,收集大规模、高质量的像素级水体标签数据将耗费大量人力物力。为了减少标注工作量,本文提出使用已有的公开水体覆盖产品来创建遥感影像对应的水体标签,然而已有的公开水体覆盖产品的空间分辨率低且存在一定错误。对此,提出采用弱监督深度学习方法训练深度语义分割网络。方法 在训练阶段,将原始数据集划分为多个互不重叠的子数据集,分别训练深度语义分割网络,并将训练得到的多个深度语义分割网络协同更新标签,然后利用更新后的标签重复前述过程,重新训练深度语义分割网络,多次迭代后可以获得好的深度语义分割网络。在测试阶段,多源遥感影像经多个代表不同视角的深度语义分割网络分别预测,然后投票产生最后的水体检测结果。结果 为了验证本文方法的有效性,基于原始多源遥感影像数据创建了一个面向水体检测的多源遥感影像数据集,并与基于传统的水体指数阈值分割法和基于低质量水体标签直接学习的深度语义分割网络进行比较,交并比（intersection-over-union,IoU）分别提升了5.5%和7.2%。结论 实验结果表明,本文方法具有收敛性,并且光学影像和合成孔径雷达（synthetic aperture radar,SAR）影像的融合有助于提高水体检测性能。在使用分辨率低、噪声多的水体标签进行训练的情况下,训练所得多视角模型的水体检测精度明显优于基于传统的水体指数阈值分割法和基于低质量水体标签直接学习的深度语义分割网络。     

Intelligent fault diagnosis for rolling bearings based on graph shift regularization with directed graphs

Graph shift regularization is a new and effective graph-based semi-supervised classification method, but its performance is closely related to the representation graphs. Since directed graphs can convey more information about the relationship between vertices than undirected graphs, an intelligent method called graph shift regularization with directed graphs (GSR-D) is presented for fault diagnosis of rolling bearings. For greatly improving the diagnosis performance of GSR-D, a directed and weighted k-nearest neighbor graph is first constructed by treating each sample (i.e., each vibration signal segment) as a vertex, in which the similarity between samples is measured by cosine distance instead of the commonly used Euclidean distance, and the edge weights are also defined by cosine distance instead of the commonly used heat kernel. Then, the labels of samples are considered as the graph signals indexed by the vertices of the representation graph. Finally, the states of unlabeled samples are predicted by finding a graph signal that has minimal total variation and satisfies the constraint given by labeled samples as much as possible. Experimental results indicate that GSR-D is better and more stable than the standard convolutional neural network and support vector machine in rolling bearing fault diagnosis, and GSR-D only has two tuning parameters with certain robustness.     

A novel deep learning prediction model for concrete dam displacements using interpretable mixed attention mechanism

Dam displacements can effectively reflect its operational status, and thus establishing a reliable displacement prediction model is important for dam health monitoring. The majority of the existing data-driven models, however, focus on static regression relationships, which cannot capture the long-term temporal dependencies and adaptively select the most relevant influencing factors to perform predictions. Moreover, the emerging modeling tools such as machine learning (ML) and deep learning (DL) are mostly black-box models, which makes their physical interpretation challenging and greatly limits their practical engineering applications. To address these issues, this paper proposes an interpretable mixed attention mechanism long short-term memory (MAM-LSTM) model based on an encoder-decoder architecture, which is formulated in two stages. In the encoder stage, a factor attention mechanism is developed to adaptively select the highly influential factors at each time step by referring to the previous hidden state. In the decoder stage, a temporal attention mechanism is introduced to properly extract the key time segments by identifying the relevant hidden states across all the time steps. For interpretation purpose, our emphasis is placed on the quantification and visualization of factor and temporal attention weights. Finally, the effectiveness of the proposed model is verified using monitoring data collected from a real-world dam, where its accuracy is compared to a classical statistical model, conventional ML models, and homogeneous DL models. The comparison demonstrates that the MAM-LSTM model outperforms the other models in most cases. Furthermore, the interpretation of global attention weights confirms the physical rationality of our attention-based model. This work addresses the research gap in interpretable artificial intelligence for dam displacement prediction and delivers a model with both high-accuracy and interpretability.     

Fixed‐time synchronization for complex‐valued BAM neural networks with time delays

In this paper, the fixed‐time synchronization for complex‐valued bidirectional associative memory (BAM) neural networks with time delays is studied. Based on the fixed‐time stability, the Lyapunov functional method and some inequality techniques, a new criterion is presented to guarantee that the addressed systems achieve synchronization in fixed time and a more accurate estimation independent of the initial conditions is given for the settling time. Meanwhile, a new nonlinear delayed controller different from the existing ones is designed. In the end, two numerical examples are provided to illustrate the effectiveness of the obtained result.     

基于RBF神经网络优化的装备保障系统效能评估

针对现有装备保障系统效能评估方法存在依赖专家经验、主观性强等问题,提出一种基于径向基函数（RBF）神经网络优化的装备保障系统效能评估方法。通过完备交叉与预变异策略对遗传算法（GA）进行改进,利用改进的GA对RBF神经网络中心、宽度及输出权值进行全局寻优,并采用优化的RBF神经网络实现装备保障系统效能评估。仿真结果表明,该方法相比AGA-RBF和GA-RBF评估方法能更准确地评估装备保障系统效能值,并且可在一定程度上降低评估过程中主观因素的影响,确保评估结果更客观真实。     

改进的YOLOv3安检包裹中危险品检测算法

针对目前的安检工作中检测效率较低的问题,提出一种改进的YOLOv3安检包裹中危险品检测算法,用于辅助安检人员完成安检工作,提升安检机的自动化和智能化。将YOLOv3中原来每个网格预测3个边界框减少到预测2个边界框,检测速度提升了约6%。利用K-means聚类根据数据集计算出先验框,平均精度均值提高了约1.13个百分点。为了解决样本量少的问题,采用数据增强方法,平均精度均值提升了约7.8个百分点。采用多尺度输入训练策略,不仅增强了模型检测不同尺度图像时的鲁棒性,而且平均精度均值提升了约1.22个百分点。     

基于多种信息组合模式的非负矩阵分解链路预测模型

针对基于拓扑结构相似性的链路预测算法中网络稀疏性和噪声问题,提出一种基于非负矩阵分解的链路预测模型。该模型从微观与宏观两个层面出发,融合网络内部和外部的辅助信息,减轻了网络稀疏性造成的影响,提升了算法预测的整体性能。提出的三种信息组合模式体现出宏观与微观角度下的信息融合策略。在13个真实网络数据集上的实验结果展示了算法预测性能上的优越性。