摘要：【中文】本发明提供一种混合资产的分析处理方法、装置、电子设备及存储介质，涉及资产管理技术领域。该方法包括：进行资产探测，获得混合资产信息；对混合资产信息进行分级分域匹配，获得匹配数据；对匹配数据进行数据格式统一处理，获得格式数据；对格式数据进行关系映射，获得关联数据；对关联数据进行聚合分析，获得资产可视化信息，并将资产可视化信息进行展示。解决了云上资产类型增多、资产分布在多个云环境和本地数据中心难以统一管理的问题，让资产真正成为可视可感可知可查的有效资产，同时提高了运维效率，减少运维难度。 【EN】The invention provides an analysis processing method and device of a mixed asset, electronic equipment and a storage medium, and relates to the technical field of asset management. The method comprises the following steps: detecting assets to obtain mixed asset information; carrying out grading and domain-division matching on the mixed asset information to obtain matching data; performing data format unified processing on the matched data to obtain format data; carrying out relation mapping on the format data to obtain associated data; and performing aggregation analysis on the associated data to obtain asset visualization information, and displaying the asset visualization information. The method solves the problems that the types of assets on the cloud are increased, the assets are distributed in a plurality of cloud environments and a local data center and are difficult to manage in a unified mode, the assets are enabled to really become visible and searchable effective assets, meanwhile, the operation and maintenance efficiency is improved, and the operation and maintenance difficulty is reduced.

摘要：【中文】本发明公开了一种汽车的选配方法，包括：获取用户对目标汽车配置的选配请求；判断所述选配请求中是否包含用户意图；若包含，依据所述选配请求，在预设汽车选配知识图谱中选取与所述选配请求对应的推荐配置；检测用户对所述推荐配置是否满意；若否，通过交互的方式对所述推荐配置进行调整，得到目标汽车配置。上述的方法，当用户对依据用户意图生成的推荐配置不满意时，可以通过交互的方式对推荐配置进行调整，避免了现有技术中，用户只可以在固定的几种选配器内选择预设的几种配置，选择受限，不可以进行交互的问题。 【EN】The invention discloses an automobile matching method, which comprises the following steps: acquiring a matching request of a user for the configuration of a target automobile; judging whether the matching request contains user intention; if yes, selecting recommended configuration corresponding to the matching request from a preset automobile matching knowledge graph according to the matching request; detecting whether the user is satisfied with the recommended configuration; and if not, adjusting the recommended configuration in an interactive mode to obtain the target automobile configuration. According to the method, when the user is unsatisfied with the recommended configuration generated according to the user intention, the recommended configuration can be adjusted in an interactive mode, and the problems that in the prior art, the user can only select preset configurations in fixed distributors, the selection is limited, and interaction cannot be performed are solved.

摘要：【中文】本发明涉及轨道车辆司机控制器，特别是一种司机控制器面板防护结构，其包括面板（1）和操控手柄，所述操控手柄经所述面板上设置的操控孔伸出所述面板，所述操控孔的下面活动设置防护板，所述防护板的中部套设在所述操控手柄上，且所述防护板随所述操控手柄移动，所述操纵孔在所述操控手柄移动过程中被所述防护板遮盖。所述面板的背面设置安装座，所述安装座上设置防护板移动轨道，所述防护板的两端穿过所述防护板移动轨道。本发明操作简单，可靠性高，可用性高，能有效防止异物掉落进入司机控制器。 【EN】The invention relates to a railway vehicle driver controller, in particular to a driver controller panel protection structure which comprises a panel (1) and a control handle, wherein the control handle extends out of the panel through a control hole formed in the panel, a protection plate is movably arranged below the control hole, the middle part of the protection plate is sleeved on the control handle, the protection plate moves along with the control handle, and the control hole is covered by the protection plate in the moving process of the control handle. The back of panel sets up the mount pad, set up guard plate removal track on the mount pad, the both ends of guard plate are passed guard plate removal track. The invention has simple operation, high reliability and high usability, and can effectively prevent foreign matters from falling into a driver controller.

摘要：【中文】本发明公开了一种基于深度学习的动态场景HDR重建方法，克服了现有技术中图像处理效果有待改进的问题。该发明含有以下步骤，在同一静态场景内用固定相机获取欠曝光、正常曝光和过曝光三幅图像；在动态场景中，用手持相机获取上述三幅图像，记为D1、D2和D3；用LK光流法对D1、S2和D3进行配准，图像序列记为R1、R2和R3和步骤1中得到的Ground Truth组成配对的训练集；利用相机相应曲线将R1、R2和R3变换到线性域，记为H1、H2和H3；利用对比度算子提取H1、H2和H3图像中的亮度信息，记为M1、M2和M3；利用梯度算子提取R1、R2和R3图像中细节信息，记为L1、L2和L3；设计基于Resnet的Attention模块。该技术生成的HDR图像细节丰富，对比度高，具有广色域高动态范围。 【EN】The invention discloses a dynamic scene HDR reconstruction method based on deep learning, which solves the problem that the image processing effect needs to be improved in the prior art. The method comprises the following steps of acquiring three images of underexposure, normal exposure and overexposure in the same static scene by using a fixed camera; in a dynamic scene, the three images are acquired by a handheld camera and are marked as D1, D2 and D3; registering D1, S2 and D3 by using an LK optical flow method, and recording image sequences as a training set of pairs consisting of R1, R2 and R3 and the group Truth obtained in the step 1; transforming R1, R2 and R3 into linear domains, denoted as H1, H2 and H3, using the corresponding curves of the camera; extracting brightness information in H1, H2 and H3 images by using a contrast operator, and recording the brightness information as M1, M2 and M3; extracting detail information in the images of R1, R2 and R3 by using a gradient operator, and recording the detail information as L1, L2 and L3; the orientation module based on Resnet is designed. The HDR image generated by the technology has rich details, high contrast and wide color gamut and high dynamic range.

摘要：【中文】本发明属于展示平台技术领域，尤其为一种新型物联网大数据智慧科技展示平台装置，包括底座，所述底座的顶端开设有第二凹槽，所述第二凹槽内壁固定连接有双轴电机，所述双轴电机的两个输出轴上分别固定连接有第一丝杆和第二丝杆；本发明，通过双轴电机工作，能够带动第一丝杆和第二丝杆转动，通过第一丝杆和第二丝杆表面螺纹方向相反，进而能够使两个丝杆套筒带动连接块相互靠近或远离，借助升降架的活动作用，能够带动升降板上下移动，该装置采用驱动组件、第二转轴、双轴电机、第一丝杆、第二丝杆和升降架，实现了对展示平台使用高度和使用方位的调节，使用灵活度高，整个装置结构合理，使用方便，实用性强。 【EN】The invention belongs to the technical field of display platforms, and particularly relates to a novel Internet of things big data intelligent science and technology display platform device which comprises a base, wherein a second groove is formed in the top end of the base, a double-shaft motor is fixedly connected to the inner wall of the second groove, and a first screw rod and a second screw rod are fixedly connected to two output shafts of the double-shaft motor respectively; according to the invention, the first lead screw and the second lead screw can be driven to rotate by the operation of the double-shaft motor, the screw thread directions of the surfaces of the first lead screw and the second lead screw are opposite, so that the two lead screw sleeves can drive the connecting blocks to mutually approach or separate, and the lifting plate can be driven to move up and down by virtue of the action of the lifting frame.

摘要：【中文】本发明公开了一种厚板剪板机自动上料装置，属于剪板机技术领域，其特征在于，包括：用于承载工件的底座平台；用于对工件位置进行调整的伺服对中部和气缸夹紧部；用于检测工件厚度的厚度检测部；用于夹持工件的气缸夹钳部；用于将工件推送至剪板工位的伺服进给送料部和推出部；其中：所述伺服进给送料部包括滑台、带动所述滑台动作的直线模组，所述滑台通过安装竖板分别与推出部的底座、气缸夹钳部的底座固定连接。通过采用上述技术方案，本发明能够实现对板料的自动精准上料，在节约人力的同时，通过厚度检测、对中、自动进给，进而实现工件的高精度上料。 【EN】The invention discloses an automatic feeding device of a thick plate shearing machine, belonging to the technical field of plate shearing machines, and being characterized by comprising the following components: a base platform for carrying a workpiece; the servo centering part and the cylinder clamping part are used for adjusting the position of a workpiece; a thickness detection section for detecting a thickness of the workpiece; a cylinder clamp part for clamping a workpiece; the servo feeding part and the pushing part are used for pushing the workpiece to the plate shearing station; wherein: the servo feeding and feeding part comprises a sliding table and a linear module for driving the sliding table to move, and the sliding table is fixedly connected with the base of the pushing part and the base of the cylinder clamp part through mounting vertical plates. By adopting the technical scheme, the automatic and accurate feeding device can realize automatic and accurate feeding of the plate, and can realize high-precision feeding of the workpiece through thickness detection, centering and automatic feeding while saving labor.

摘要：【中文】本发明属于高光谱图像超分辨率处理技术领域，公开了一种基于频带间注意力机制对抗学习高光谱单图超分辨率方法，搭建三维生成对抗网络模型，设置其隐含层层数、激活函数、训练函数、损失函数、学习率；将频带注意力机制嵌入生成网络的残差块之间；使用训练数据集对三维生成对抗网络模型进行训练，调整其内部参数；使用测试数据集对三维生成对抗网络模型进行检核，根据模型输出求平均峰值信噪比MPSNR、平均结构相似性指数MSSIM、平均均方根误差MRMSE和光谱角制图SAM，评价三维生成对抗网络模型的超分辨率处理性能。本发明是端对端可训练的高光谱单图超分辨率模型，能够有效实现高光谱图像空间分辨率的增强和光谱保真。 【EN】The invention belongs to the technical field of hyperspectral image super-resolution processing, and discloses a hyperspectral single image super-resolution method for countercheck learning based on an inter-band attention mechanism, wherein a three-dimensional generation countercheck network model is built, and the number of hidden layers, an activation function, a training function, a loss function and a learning rate of the model are set; embedding a frequency band attention mechanism between residual blocks of a generated network; training the three-dimensional generation confrontation network model by using a training data set, and adjusting internal parameters of the three-dimensional generation confrontation network model; and checking the three-dimensional generation countermeasure network model by using the test data set, solving the average peak signal-to-noise ratio (MPSNR), the average structure similarity index (MSSIM), the average root mean square error (MRMSE) and the Spectral Angle Mapping (SAM) according to the output of the model, and evaluating the super-resolution processing performance of the three-dimensional generation countermeasure network model. The hyperspectral single-image super-resolution model can be trained end to end, and the enhancement of the spatial resolution of a hyperspectral image and the spectral fidelity can be effectively realized.

摘要：【中文】本发明属于高光谱图像超分辨率处理技术领域，公开了一种基于1维‑2维卷积神经网络高光谱单图超分辨方法，搭建1维卷积神经网络模型获取光谱信息；搭建基于注意力机制的2维卷积神经网络模型获取空间信息；将空间和光谱信息以渐进的方式进行融合；使用训练数据集对并行结构的神经网络模型进行训练，调整其内部参数；使用测试数据集对神经网络模型进行检核，根据模型输出求平均峰值信噪比MPSNR、平均结构相似性指数MSSIM、平均均方根误差MRMSE和光谱角制图SAM，评价神经网络模型的超分辨率处理性能。本发明神经网络模型参数设置简单，易于实现，计算量小，能较快获得超分辨率处理结果；能够更有效地实现光谱保真。 【EN】The invention belongs to the technical field of hyperspectral image super-resolution processing, and discloses a hyperspectral single-image super-resolution method based on a 1-dimensional-2-dimensional convolutional neural network, wherein a 1-dimensional convolutional neural network model is built to obtain spectral information; constructing a 2-dimensional convolution neural network model based on an attention mechanism to obtain spatial information; fusing spatial and spectral information in a progressive manner; training a neural network model of a parallel structure by using a training data set, and adjusting internal parameters of the neural network model; and checking the neural network model by using the test data set, solving the average peak signal-to-noise ratio (MPSNR), the average structure similarity index (MSSIM), the average root mean square error (MRMSE) and the Spectral Angle Mapping (SAM) according to the output of the model, and evaluating the super-resolution processing performance of the neural network model. The neural network model parameter setting is simple, the implementation is easy, the calculation amount is small, and the super-resolution processing result can be obtained quickly; spectral fidelity can be more effectively achieved.

摘要：【中文】本发明提出了一种基于决策级融合的彩色图像感兴趣区域提取方法，主要解决现有技术中感兴趣区域提取方法的适用范围有限，感兴趣区域提取的精度不足的问题。其实现步骤为：获取训练数据集；获取RGB图像I的目标分割图I₁；获取RGB图像I的景深分割图I₂；获取RGB图像I的热度分割图I₃；获取RGB图像I的显著性分割图I₄；对I₁、I₂、I₃和I₄进行决策级融合。本发明从像素级分类角度、摄像机与感兴趣区域的距离角度、视觉关注角度、频率角度综合分析，融合四个方面的优势，拓宽感兴趣区域提取方法的适用范围，提高感兴趣区域提取的准确性，可用于目标检测、障碍物识别、图像视觉监控、行人检测等领域。 【EN】The invention provides a color image region-of-interest extraction method based on decision-level fusion, which mainly solves the problems that the region-of-interest extraction method in the prior art is limited in application range and insufficient in accuracy of region-of-interest extraction. The method comprises the following implementation steps: acquiring a training data set; target segmentation chart I for acquiring RGB image I₁(ii) a Obtaining depth-of-field segmentation map I of RGB image I₂(ii) a Obtaining a heat segmentation chart I of an RGB image I₃(ii) a Obtaining a saliency map I of an RGB image I₄(ii) a To I₁、I₂、I₃And I₄And performing decision-level fusion. The method comprehensively analyzes the four aspects of the classification angle of the pixel level, the distance angle between the camera and the interested region, the visual attention angle and the frequency angle, integrates the advantages of the four aspects, and widens the extracting method of the interested regionThe method has the advantages of improving the accuracy of extracting the region of interest, and being applicable to the fields of target detection, obstacle identification, image vision monitoring, pedestrian detection and the like.

摘要：【中文】本发明涉及一种嵌入式平台功能配置方法、配置工具及系统。其中配置方法包括：配置工具加载嵌入式平台的目标文件；根据变量的名称进行查找，获取变量的内存地址，从内存地址中提取变量的值，输出操作页面，操作页面显示变量的值和变量的值的修改窗口；等待操作人员完成对变量的值的修改；将修改结果保存为二进制文件，传输到嵌入式平台；嵌入式平台接收二进制文件，进行加载以完成功能配置。该方法直接在目标文件中查找待修改的变量，并对变量的值进行修改，将修改结果保存为二进制文件，以在加载二进制文件时，直接填充变量的值，根据该值使能或者关闭某些功能，减少了解析文件的过程，减少了文件操作指令，提高功能配置的效率。 【EN】The invention relates to a function configuration method, a configuration tool and a system of an embedded platform. The configuration method comprises the following steps: loading a target file of the embedded platform by the configuration tool; searching according to the name of the variable, acquiring a memory address of the variable, extracting a value of the variable from the memory address, and outputting an operation page which displays the value of the variable and a modification window of the value of the variable; waiting for the operator to complete modification of the value of the variable; saving the modification result as a binary file, and transmitting the binary file to the embedded platform; and the embedded platform receives the binary file and loads the binary file to complete the function configuration. The method directly searches the variable to be modified in the target file, modifies the value of the variable, saves the modification result as a binary file, directly fills the value of the variable when loading the binary file, enables or closes some functions according to the value, reduces the process of analyzing the file, reduces the file operation instruction, and improves the efficiency of function configuration.