摘要：【中文】本发明涉及压力机技术领域内一种压力机组合机身用液压预紧螺母，包括拉紧螺栓和液压螺母，液压螺母的螺孔外周的周向设有若干活塞腔，活塞腔的上部配合设有定活塞，定活塞上侧与活塞腔的内壁之间设有固定部件，定活塞的下部设有与活塞腔配合的动活塞，动活塞与定活塞的相对端之间形成油腔，动活塞相对定活塞设有预定的滑动行程，油腔侧向的液压螺母外壁上设有进出油口，动活塞的下部设有活塞杆，活塞杆对应的液压螺母底侧设有向下贯通的导向孔，活塞杆与导向孔导向配合并且底部与压力机组合机身抵靠，油腔内打压至预定油压后动活塞相对定活塞下移以顶起液压螺母并预紧拉紧螺栓，液压螺母底侧设有用于锁固液压螺母预紧状态的U型垫片。 【EN】The invention relates to a hydraulic pre-tightening nut for a combined machine body of a press machine in the technical field of press machines, which comprises a tightening bolt and a hydraulic nut, wherein a plurality of piston cavities are arranged on the periphery of a screw hole of the hydraulic nut in the circumferential direction, a fixed piston is arranged on the upper part of each piston cavity in a matching way, a fixed part is arranged between the upper side of each fixed piston and the inner wall of each piston cavity, a movable piston matched with each piston cavity is arranged on the lower part of each fixed piston, an oil cavity is formed between the movable piston and the opposite end of the fixed piston, the movable piston is provided with a preset sliding stroke relative to the fixed piston, an oil inlet and outlet is arranged on the outer wall of the hydraulic nut on the lateral side of each oil cavity, a piston rod is arranged on the lower part of the movable piston, a guide hole penetrating downwards is arranged on the bottom side of the hydraulic nut corresponding to the piston rod, the, and a U-shaped gasket for locking the pre-tightening state of the hydraulic nut is arranged at the bottom side of the hydraulic nut.

摘要：【中文】本发明属于医学图像识别技术领域，具体涉及应用在临床实践上的一种基于神经网络的脑梗死MRI图像识别方法。本发明包括以下步骤：采集明确具有脑梗死疾病的脑部MRI图像；数据预处理，对标注好的原始图像进行缩放和裁剪处理；构建神经网络，卷积神经网络总深度为11层,卷积核大小为3*3,第2、4、7层步长为2,其余层步长为1；训练神经网络；测试模型；诊断测试；输出模型和整理结果。本发明的优点在于：使用卷积神经网络进行头部MRI图像识别,相比于传统方法,准确率更高,诊断过程更加智能化，且可以应用于除头部MRI识别外的其他应用场景。本发明使用全卷积网络作为基础网络结构,保留局部信息,使得学习到的特征更易被可视化和理解。 【EN】The invention belongs to the technical field of medical image identification, and particularly relates to a neural network-based cerebral infarction MRI image identification method applied to clinical practice. The invention comprises the following steps: acquiring a brain MRI image which definitely has cerebral infarction diseases; data preprocessing, namely zooming and cutting the marked original image; constructing a neural network, wherein the total depth of the convolutional neural network is 11 layers, the size of a convolutional kernel is 3 x 3, the step sizes of the 2 nd, 4 th and 7 th layers are 2, and the step sizes of the rest layers are 1; training a neural network; testing the model; a diagnostic test; and outputting the model and sorting the result. The invention has the advantages that: compared with the traditional method, the method has higher accuracy and more intelligent diagnosis process, and can be applied to other application scenes except head MRI identification. The invention uses the full convolution network as the basic network structure, and keeps the local information, so that the learned characteristics are easier to visualize and understand.

摘要：【中文】本发明属于运动目标跟踪检测技术领域，具体涉及一种基于特征点网格化匹配的运动目标跟踪检测方法。本发明通过先提取运动目标和待检测目标的特征点进行匹配，然后划分网格并统计特征点落于网格的位置，再根据网格中的特征点找到运动目标在待检测图像中的位置。本发明将运动的平滑性限制转换成去除错误匹配的数据测量，利用特征点正确匹配对周围比错误对周围有更多的支持点对，筛选剔除错误点对，大大提高了特征点匹配的正确率，从而提升了目标检测的准确度。本发明可用于解决由于运动的平滑性限制而造成的运动目标检测困难问题，本发明能够迅速剔除错误的匹配，提高匹配的稳定性，从而迅速定位运动目标的位置。 【EN】The invention belongs to the technical field of moving target tracking detection, and particularly relates to a moving target tracking detection method based on feature point gridding matching. The method comprises the steps of extracting characteristic points of a moving target and a target to be detected for matching, then dividing grids, counting the positions of the characteristic points falling on the grids, and finding the position of the moving target in the image to be detected according to the characteristic points in the grids. The method converts the smoothness limit of the movement into the data measurement for removing the error matching, screens and rejects error point pairs by utilizing the fact that the correct matching of the characteristic points has more supporting point pairs around than the error pairs, and greatly improves the correct rate of the matching of the characteristic points, thereby improving the accuracy of target detection. The invention can be used for solving the problem of difficulty in detecting the moving target caused by the limitation of the moving smoothness, can quickly eliminate wrong matching and improve the matching stability, thereby quickly positioning the position of the moving target.

摘要：【中文】本发明属于人脸识别技术领域，具体涉及一种基于卷积神经网络的多重风格人脸特征点检测方法。本发明通过输入多重风格的人脸图像对残差网络(ResNet‑152)进行微调训练，以此来获得风格辨别的特征；利用风格辨别特征，使用K‑means对原始图像进行风格聚类；通过生成性对抗网络生成风格聚合的人脸图像集合；最后，将原始人脸图像与风格聚合人脸图像共同作为输入，以级联策略生成人脸特征点预测。本发明降低了多重风格对人脸检测造成的误差，提高了人脸检测的精确度，可应用于人脸识别、头部姿势估计、面部重建和3D面部重建等。 【EN】The invention belongs to the technical field of face recognition, and particularly relates to a multiple style face characteristic point detection method based on a convolutional neural network. The method comprises the steps of performing fine tuning training on a residual error network (ResNet-152) by inputting multiple styles of human face images so as to obtain style distinguishing characteristics; carrying out style clustering on the original image by using K-means by utilizing style distinguishing characteristics; generating a style aggregated facial image set through a generative confrontation network; and finally, the original face image and the style aggregated face image are used as input together, and the face characteristic point prediction is generated by a cascading strategy. The method reduces the error of the face detection caused by multiple styles, improves the accuracy of the face detection, and can be applied to face recognition, head posture estimation, face reconstruction, 3D face reconstruction and the like.

摘要：【中文】本发明属于图像超分辨率重建技术领域，具体涉及一种基于地形匹配的声呐图像的超分辨率重建方法。本发明通过使用中值滤波器、拉普拉斯滤波器对声呐图像进行预处理，在声呐图像中识别并选定地形作为参照物，对同一地形参照物进行配准，解决声呐图像在产生的过程中同一地形在图像中的位置不同的问题。本发明利用声呐图像中的同一地形，将多张声呐图像依据该地形进行配准，使用多张声呐图像重建一张新的图像，直到最终高分辨率图像的生成，解决了声呐图像分辨率低、信噪比低、对比度低、斑点噪声突出的问题，提高了声呐图像的分辨率从而获得更多信息。 【EN】The invention belongs to the technical field of super-resolution reconstruction of images, and particularly relates to a super-resolution reconstruction method of a sonar image based on terrain matching. The method comprises the steps of preprocessing a sonar image by using a median filter and a Laplace filter, identifying and selecting a terrain in the sonar image as a reference object, and registering the same terrain reference object, so that the problem that the positions of the same terrain in the image are different in the generation process of the sonar image is solved. The method and the device have the advantages that the same terrain in the sonar images is utilized, the sonar images are registered according to the terrain, the sonar images are used for reconstructing a new image until the final high-resolution image is generated, the problems of low resolution, low signal-to-noise ratio, low contrast and prominent speckle noise of the sonar images are solved, and the resolution of the sonar images is improved so as to obtain more information.

摘要：【中文】本发明涉及一种打车平台中补偿激励的任务分配方法领域。所述方法包括如下步骤：任务发布者向打车平台发布打车任务并提供打车任务酬金；打车平台预测以打车任务位置为圆心的半径内接受任务的司机人数期望；在任务发布范围内的司机收到任务后选择是否接受任务，接受任务的司机提交接受任务申请给平台；打车平台收到接受任务司机的申请后，将这些司机作为任务候选者发送给打车任务发布者；任务发布者可以根据自身的需求从平台提供的候选者中选择一个合适的司机接受任务；任务完成后，平台支付任务报酬。本发明能够将任务指定到接受任务意愿较高的司机人群中，打车任务会发布到离打车任务更近范围内的司机，使得任务能够在更近的范围内被接受。 【EN】The invention relates to the field of a compensation excitation task allocation method in a taxi taking platform. The method comprises the following steps: the task publisher publishes the taxi taking task to the taxi taking platform and provides toll for the taxi taking task; the taxi taking platform predicts the expectation of the number of drivers who receive the taxi within a radius taking the taxi taking task position as a circle center; after receiving the task, a driver in the task issuing range selects whether to accept the task, and the driver accepting the task submits a task accepting application to the platform; after receiving the application of accepting the task drivers, the taxi taking platform sends the drivers as task candidates to taxi taking task issuers; the task publisher can select a suitable driver from candidates provided by the platform according to the requirement of the task publisher to accept the task; and after the task is completed, the platform pays the task reward. The invention can assign the task to the driver group with higher task accepting willingness, and the taxi taking task can be issued to the driver within a closer range to the taxi taking task, so that the task can be accepted within the closer range.

摘要：【中文】本发明公开了一种数值水池虚拟试验数据中间件的设计方法。数值水池虚拟试验应用系统(200)通过API模块(112)和通信模块(113)与主中间件(120)的通信模块(122)双向传输信号，通信模块(122)通过任务处理模块(124)与数值水池虚拟试验底层分布式NoSQL数据库(300)双向传输信号。由于数值水池虚拟试验的数据包括数值型数据、字符串、数组和图像等各种异构数据，而且属于大规模数据，用户对数据的安全和并发任务处理要求比较高；因此需要在底层配置分布式NoSQL数据库集群模式，满足对多用户、多任务的高并发需求。 【EN】The invention discloses a design method of a virtual test data middleware of a numerical water pool. The numerical value pool virtual test application system (200) transmits signals to and from a communication module (122) of the main middleware (120) in a bidirectional mode through an API module (112) and a communication module (113), and the communication module (122) transmits signals to and from a numerical value pool virtual test underlying distributed NoSQL database (300) in a bidirectional mode through a task processing module (124). The data of the numerical value pool virtual test comprises various heterogeneous data such as numerical value type data, character strings, arrays, images and the like, and belongs to large-scale data, so that the data safety and concurrent task processing requirements of users are higher; therefore, a distributed NoSQL database cluster mode needs to be configured on the bottom layer, and high concurrency requirements on multiple users and multiple tasks are met.

摘要：【中文】本发明属于通信技术领域，具体涉及一种基于空间波形两维加权的雷达通信一体化设计方法，构建收发阵型，并确定基带信号；根据所述收发阵型得到最优发射权组；利用所述最优发射权组对所述基带信号进行调制得到调制基带信号；利用所述收发阵型发送所述调制基带信号并接收回波信号和检测通信信息。具有通信速率快的有益效果。 【EN】The invention belongs to the technical field of communication, and particularly relates to a radar communication integrated design method based on space waveform two-dimensional weighting, which is used for constructing a receiving and transmitting array type and determining a baseband signal; obtaining an optimal transmitting weight group according to the transmitting-receiving array type; modulating the baseband signal by using the optimal transmitting weight group to obtain a modulated baseband signal; and transmitting the modulation baseband signal and receiving an echo signal and detecting communication information by using the transmitting and receiving array. The method has the advantage of high communication speed.

摘要：【中文】本发明提供一种制备2‑羟基‑4‑甲硫基丁酸及其中间体的方法和装置，所述制备2‑羟基‑4‑甲硫基丁酸的中间体包括3‑甲硫基丙醛、2‑羟基‑4‑甲硫基丁腈。本发明提供的制备2‑羟基‑4‑甲硫基丁酸的方法，包括：步骤(1)，使丙烯醛与甲硫醇反应制备3‑甲硫基丙醛的步骤；步骤(2)，使3‑甲硫基丙醛与氢氰酸反应制备2‑羟基‑4‑甲硫基丁腈的步骤；步骤(3)，2‑羟基‑4‑甲硫基丁腈用硫酸进行水合、再水解制备2‑羟基‑4‑甲硫基丁酸的步骤；其中，在步骤(1)、(2)和(3)中在线检测物料的反应情况，并根据检测结果控制物料配比以使得反应进行完全。 【EN】The invention provides a method and a device for preparing 2-hydroxy-4-methylthio butyric acid and an intermediate thereof, wherein the intermediate for preparing the 2-hydroxy-4-methylthio butyric acid comprises 3-methylthio propionaldehyde and 2-hydroxy-4-methylthio butyronitrile. The invention provides a method for preparing 2-hydroxy-4-methylthio butyric acid, which comprises the following steps: a step (1) of reacting acrolein with methyl mercaptan to produce 3-methylthiopropanal; a step (2) of reacting 3-methylthiopropanal with hydrocyanic acid to prepare 2-hydroxy-4-methylthiobutyronitrile; step (3), hydrating and hydrolyzing the 2-hydroxy-4-methylthiobutyronitrile with sulfuric acid to prepare 2-hydroxy-4-methylthiobutyric acid; wherein, the reaction condition of the materials is detected on line in the steps (1), (2) and (3), and the material proportion is controlled according to the detection result so as to ensure that the reaction is completely carried out.