电工实践教学改革与实践

电工实践教学是培养大学生电类工程能力的重要环节，本文介绍我们以培养人才为中心在电工实习基地建设；实践教学内容，方法和手段等方面进行的改革与实践。     

唐山高技能人才队伍现状分析与思考

随着唐山市经济的快速发展,高技能人才缺乏的现象已日渐明显,对唐山市的人才现状进行了调研,并分析了造成人才短缺现象的原因,提出了加快高技能人才队伍建设的建议。     

基于视频的人体行走状态分析

提出了一种基于多视角的人体行走状态分析方法.首先,利用基于自适应背景更新的背景差分法检测出运动人体;其次,利用4向链码提取人体上半身质心运动轨迹的周期;最后,通过训练不同运动状态的周期得到对应的模糊集,根据隶属度函数判别慢走、走和跑3种运动状态.实验结果表明,该方法能够准确分析人体运动速度并判断其运动状态,并且对背景噪声、视角变化有良好的鲁棒性.     

基于多尺度特征融合的行人重识别方法

针对行人重识别任务在特征提取时缺乏对行人特征尺度变化的考虑,导致其易受环境影响而具有低行人重识别准确率的问题,提出了一种基于多尺度特征融合的行人重识别方法。首先,在网络浅层通过混合池化操作来提取多尺度的行人特征,从而帮助网络提升特征提取能力;然后,在残差块内添加条形池化操作以分别提取水平和竖直方向的远程上下文信息,从而避免无关区域的干扰;最后,在残差网络之后利用不同尺度的空洞卷积进一步保留多尺度的特征,从而帮助模型灵活有效地解析场景结构。实验结果表明,在Market-1501数据集上,所提方法的Rank1达到95.9%,平均精度均值（mAP）为88.5%;在DukeMTMC-reID数据集上,该方法的Rank1达到90.1%,mAP为80.3%。可见所提方法能够较好地保留行人特征信息,从而提高行人重识别任务准确率。     

结合域间与域内变化的跨域行人重识别算法

针对单域训练的行人重识别模型迁移到另一个域内测试时性能巨大下降的问题,提出一种结合域间与域内变化的跨域行人重识别算法。采用PR策略将行人特征图进行分区处理,提高模型泛化能力。针对域间变化,引入域间姿势不变性,缩小源域和目标域行人的姿势差距。针对域内变化,引入域内姿势不变性、样本不变性、邻域不变性和相机风格不变性,扩大不同行人之间的距离,缩小相同行人之间的距离。实验表明,PR策略和5个不变性能有效地增强跨域行人重识别中的域自适应性,与只注重域间变化或域内变化的算法相比,其识别精度得到有效提升。     

以课程设计为中心的排水工程课教学改革初探

排水工程是给排水科学与工程专业的一门专业核心课程,剖析了以排水工程课程设计为核心的课程教学模式改革,阐述了培养学生实践应用能力和创新能力的一些具体教学方法。     

基于聋人案例的空间隐喻语义认知计算

我们从心理语言学的角度,对空间隐喻进行相似性分类和计算,使用多维量表和聚类方法,以聋人为被试分别进行了两个实验。实验结果表明,聋人为实现空间隐喻理解的计算,使用了地形空间和语法空间的特征信息,同时受手语语言特点的影响,其空间隐喻的认知主题包括手势者自身参照系、参照物的相对坐标系、手势空间的饱和度、以手部或胸部为边界。同时表明由于两种空间的存在,聋人大脑对空间隐喻的理解存在着层次,并且在长期使用手语交流的过程中,其地形空间和语法空间相互作用,影响了聋人大脑空间隐喻的结构和表征,从而导致了独特的高效快速空间隐喻计算。
     

Studies on a New Concept of 3D Data Integration about Reaches and Forces of a Disabled Person on a Wheelchair (CAD Methods in Car and Market Ergonomics)

The article presents a new concept of combining the three dimensions (3D) of a person's manipulation space. The data concern information about the reach of the arms and biomechanical data about limiting the load of a disabled person sitting in a wheelchair. Measurement data were acquired empirically, on original measuring station. The data included, respectively, arms' reach (static and dynamic) or, alternatively, measurements of limiting forces. The obtained data were processed into virtual 3D surfaces of arms' reach and forces. These surfaces provide the required graphic model of anthropotechnical and biomechanical data. Developed model was utilized to perform a virtual analysis of the accessibility of a disabled person to technical means: in a market sale space and in the ergonomic analysis into the space of a personal car. The presented method of 3D graphic modeling of anthropometrical and biomechanical data can be universally applied in ergonomic designing of work stations not only for disabled persons. © 2011 Wiley Periodicals, Inc.     

Feature generation based on relation learning and image partition for occluded person re-identification

In order to solve the challenging tasks of person re-identification(Re-ID) in occluded scenarios, we propose a novel approach which divides local units by forming high-level semantic information of pedestrians and generates features of occluded parts. The approach uses CNN and pose estimation to extract the feature map and key points, and a graph convolutional network to learn the relation of key points. Specifically, we design a Generating Local Part (GLP) module to divide the feature map into different units. Based on different occluded conditions, the partition mode of GLP has high flexibility and variability. The features of the non-occluded parts are clustered into an intermediate node, and then the spatially correlated features of the occluded parts are generated according to the de-clustering operation. We conduct experiments on both the occluded and the holistic datasets to demonstrate its effectiveness.     

Transformer-based global–local feature learning model for occluded person re-identification

Most recent occluded person re-identification (re-ID) methods usually learn global features directly from pedestrian images, or use additional pose estimation and semantic analysis model to learn local features, while ignoring the relationship between global and local features, thus incorrectly retrieving different pedestrians with similar attributes as the same pedestrian. Moreover, learning local features using auxiliary models brings additional computational cost. In this work, we propose a Transformer-based dual-branch feature learning model for occluded person re-ID. Firstly, we propose a global–local feature interaction module to learn the relationship between global and local features, thus enhancing the richness of information in pedestrian features. Secondly, we randomly erase local areas in the input image to simulate the real occlusion situation, thereby improving the model’s adaptability to the occlusion scene. Finally, a spilt group module is introduced to explore the local distinguishing features of pedestrian. Numerous experiments validate the effectiveness of our proposed method.