摘要：【中文】本发明公开了一种全像影像片体、全像影像记录方法与重建方法，其中该全像影像记录方法包括预备步骤、照射步骤及记录步骤。其中预备步骤包括叠置全像底片于透明基板上。照射步骤包括发出物光与参考光，参考光照射进入透明基板内并于透明基板的厚度内进行多次全反射以形成内部全反射光。记录步骤包括内部全反射光与物光彼此干涉产生一全像影像干涉纹路并被感光记录于全像底片上。 【EN】The invention discloses a holographic image sheet, a holographic image recording method and a reconstruction method, wherein the holographic image recording method comprises a preparation step, an irradiation step and a recording step. Wherein the preliminary step includes stacking the full image negative on the transparent substrate. The irradiating step includes emitting object light and reference light, and the reference light irradiates into the transparent substrate and is totally reflected for multiple times in the thickness of the transparent substrate to form internal total reflection light. The recording step includes that the internal total reflection light and the object light interfere with each other to generate a total image interference pattern and are recorded on the total image negative film in a photosensitive mode.

摘要：【中文】本发明涉及一种定位导盲眼镜，包括镜架，在镜架内设置有主控模块、定位模块、距离感应模块、接收模块和应急模块，所述定位模块包括GPS模组和SIM卡模组，所述距离感应模块与所述主控模块连接，用于检测到障碍物后，将障碍物的距离信息输入所述主控模块，并通过主控模块发送给接收模块，使盲人获取所述障碍物的距离信息；所述定位模块和应急模块分别与所述主控模块连接，所述GPS模组用于获取盲人的当前位置信息，并通过所述应急模块发送至相关联的手机；所述SIM卡模组为所述GPS模组和应急模块提供所需的无线信号和流量。本发明可以在盲人出现迷路等紧急情况时发出报警，以及时获得救援。 【EN】The invention relates to a pair of positioning blind-guiding glasses, which comprises a glasses frame, wherein a main control module, a positioning module, a distance sensing module, a receiving module and an emergency module are arranged in the glasses frame, the positioning module comprises a GPS module and an SIM card module, the distance sensing module is connected with the main control module and is used for inputting the distance information of an obstacle into the main control module after the obstacle is detected and sending the distance information to the receiving module through the main control module, so that a blind person can obtain the distance information of the obstacle; the positioning module and the emergency module are respectively connected with the main control module, and the GPS module is used for acquiring the current position information of the blind person and sending the current position information to the associated mobile phone through the emergency module; the SIM card module provides required wireless signals and flow for the GPS module and the emergency module. The invention can send out alarm when the blind person is lost and other emergency situations occur, and can obtain rescue in time.

摘要：【中文】基于转移概率的车辆风险综合评价方法，其步骤为：1)数据的采集及处理：2)确定指标分类：根据指标对风险影响大小，将指标分为对风险影响大的A类指标和对风险影响小的B类指标；其中A类指标包括防碰撞预警指标和疲劳驾驶报警指标；3)针对预处理过后的车联网报警数据，构建的车辆驾驶风险评估模型。通过上述步骤，本发明提供了一种评价准确性好，使用效果好，适用能力强的基于转移概率的车辆风险综合评价方法。 【EN】A comprehensive evaluation method for vehicle risks based on transition probability comprises the following steps: 1) data acquisition and processing: 2) determining index classification: according to the influence of the indexes on the risk, dividing the indexes into A-type indexes with large influence on the risk and B-type indexes with small influence on the risk; the A-type indexes comprise anti-collision early warning indexes and fatigue driving warning indexes; 3) and aiming at the preprocessed vehicle networking alarm data, a vehicle driving risk assessment model is constructed. Through the steps, the comprehensive vehicle risk evaluation method based on the transition probability is good in evaluation accuracy, good in using effect and strong in applicability.

摘要：【中文】本发明公开了一种基于端到端模型的代码可信执行方法，数据使用方根据数据提供方设定的数据规范进行代码编写并加密传输至数据提供方；数据提供方则为数据使用方提供隔离计算运行环境，由数据提供方启动可信执行环境，数据使用方将加密后的代码与密钥分别发送至可信执行环境内，可信执行环境恢复代码并执行，代码执行过程中通过数据访问代理获取数据并进行处理，结果通过代码执行代理返回给数据使用方。采用本发明可以同时对数据提供方的数据和数据使用方的代码进行保护，使得数据提供方和数据使用方在互不可信的情况下也能够进行信息交互。 【EN】The invention discloses a code credible execution method based on an end-to-end model.A data user writes codes according to data specifications set by a data provider and encrypts and transmits the codes to the data provider; the data provider provides an isolated computing operation environment for the data user, the data provider starts a trusted execution environment, the data user sends the encrypted codes and the encrypted keys to the trusted execution environment respectively, the trusted execution environment recovers the codes and executes the codes, the codes are obtained and processed through the data access agent in the execution process, and the results are returned to the data user through the code execution agent. The invention can protect the data of the data provider and the code of the data user at the same time, so that the data provider and the data user can carry out information interaction under the condition of mutual non-credibility.

摘要：【中文】本发明公开了一种使用语义补全神经网络来解决视频时序定位问题的方法。主要包括如下步骤：1)针对于一个视频和一段自然语言描述，通过一个候选产生模块来生成候选的视频片段及对应的置信度，并使用特定的算法从中挑选出K个优先级最高的候选视频片段。2)通过语义补全模块计算K个候选视频片段的激励，并反馈给候选产生模块。3)重复以上步骤对整个网络进行训练，得到良好的视频时序定位的网络。相比于传统的视频时序定位的学习方法，本发明的学习方法是弱监督的学习方法，仅仅需要粗粒度的视频级别的标注数据集，可以节省昂贵的数据集标注成本。 【EN】The invention discloses a method for solving the problem of video time sequence positioning by using a semantic completion neural network. The method mainly comprises the following steps: 1) aiming at a video and a natural language description, candidate video clips and corresponding confidence degrees are generated through a candidate generation module, and K candidate video clips with the highest priority are selected from the candidate video clips by using a specific algorithm. 2) And calculating the excitation of the K candidate video clips through a semantic completion module, and feeding back the excitation to the candidate generation module. 3) And repeating the steps to train the whole network to obtain the network with good video time sequence positioning. Compared with the traditional learning method for video time sequence positioning, the learning method is a weak supervision learning method, only needs the labeling data set with coarse granularity at the video level, and can save the expensive labeling cost of the data set.

摘要：【中文】本发明公开一种线性滑轨，包含一导轨、一滑块、四个套管、两个端盖、两个循环座及四个滚动单元。滑块以其配合槽可滑动地设置于导轨上。四个套管分别穿设滑块的四个容置通道。两个端盖分别装设于滑块。每个端盖在其面向滑块的部分具有两个结合凸肋。每个循环座包含与相邻的两个套管相接的两个循环部、连接两个循环部的两个连接部及连接两个连接部的一个中保持部。中保持部分别与两个连接部形成有两个结合凹口。每个端盖的两个结合凸肋于几何上与其面向的两个结合凹口互补，且每个端盖的两个结合凸肋与其面向的两个结合凹口无间隙地结合。借此，端盖与循环座之间结合部分的间隙或高低断差能被有效地消除，进而使得滚动单元能够滚动地更佳顺畅。 【EN】The invention discloses a linear sliding rail which comprises a guide rail, a sliding block, four sleeves, two end covers, two circulating seats and four rolling units. The slide block is arranged on the guide rail in a sliding way by the matching groove. The four sleeves are respectively penetrated through the four accommodating channels of the sliding block. Two end covers are respectively arranged on the sliding block. Each end cap has two coupling ribs at a portion thereof facing the slider. Each circulation seat comprises two circulation parts connected with two adjacent bushings, two connecting parts connecting the two circulation parts and a middle holding part connecting the two connecting parts. The middle holding part and the two connecting parts are respectively provided with two combining notches. The two coupling ribs of each end cap are geometrically complementary to the two coupling recesses facing thereto, and the two coupling ribs of each end cap are coupled with the two coupling recesses facing thereto without a gap. Therefore, the gap or the height offset of the combination part between the end cover and the circulating seat can be effectively eliminated, and the rolling unit can roll better and smoothly.

摘要：【中文】本发明公开了一种动态道路交会场景目标车辆散射建模方法，包含：获取目标车辆的数字化模型。以目标车辆数字化模型为输入，开展仿真，获取目标车辆的二维雷达像。提取经仿真的目标车辆的二维雷达像中的目标车辆散射中心。对目标车辆散射中心进行坐标系转换，以将目标车辆散射中心转换至目标车辆本体坐标系下。对动态道路交会场景下的目标车辆与探测雷达的相对运动姿态进行分析，根据探测雷达相对目标车辆的指向，获取各个采样时刻下所对应的目标车辆的散射中心分布模型。调用天线方向图信息对散射中心分布模型进行加权，得到目标车辆在行驶过程中，目标车辆的所有散射中心中随时间变化最强的散射中心的目标车辆散射中心序列。本发明降低虚警率。 【EN】The invention discloses a target vehicle scattering modeling method for a dynamic road intersection scene, which comprises the following steps: a digitized model of the target vehicle is obtained. And (4) taking the digital model of the target vehicle as input, carrying out simulation, and obtaining a two-dimensional radar image of the target vehicle. Target vehicle scattering centers in the simulated two-dimensional radar image of the target vehicle are extracted. And converting the target vehicle scattering center into a target vehicle body coordinate system by using a coordinate system conversion method. And analyzing the relative motion attitude of the target vehicle and the detection radar in the dynamic road intersection scene, and acquiring a scattering center distribution model of the corresponding target vehicle at each sampling moment according to the direction of the detection radar relative to the target vehicle. And calling antenna directional diagram information to weight the scattering center distribution model, and obtaining a target vehicle scattering center sequence of the scattering center which has the strongest change along with time in all the scattering centers of the target vehicle in the driving process of the target vehicle. The invention reduces the false alarm rate.

摘要：【中文】本发明提供一种雾化给药装置的锁定结构，包括第一壳体、第二壳体、螺杆以及扣件。第一壳体具有抵掣部与扣槽，扣槽邻接并形成于抵掣部一侧。第二壳体可旋转地连接第一壳体，借而相对第一壳体沿旋转轴线旋转。螺杆枢设于第二壳体并平行于前述旋转轴线，且螺杆随第二壳体与第一壳体的相对旋转而沿自体中心轴旋转。扣件连接螺杆而受其驱动，且扣件朝向第一壳体的抵掣部凸伸。扣件供抵掣部抵掣以产生位移，以于通过抵掣部后而限位于扣槽。借此，本发明以螺杆及扣件构成锁定机制，可降低结构复杂性与制造成本较低，并防止雾化给药装置被错误地锁定。 【EN】The invention provides a locking structure of an atomization drug delivery device, which comprises a first shell, a second shell, a screw rod and a fastener. The first shell is provided with a pushing part and a buckling groove, and the buckling groove is adjacent to and formed on one side of the pushing part. The second housing is rotatably coupled to the first housing so as to rotate relative to the first housing along the rotation axis. The screw rod is pivoted on the second shell and is parallel to the rotating axis, and the screw rod rotates along the self central shaft along with the relative rotation of the second shell and the first shell. The fastener is connected with the screw rod and driven by the screw rod, and the fastener protrudes towards the abutting part of the first shell. The fastener is used for the abutting part to abut and generate displacement so as to be limited in the buckling groove after passing through the abutting part. Therefore, the invention uses the screw and the fastener to form a locking mechanism, thereby reducing the structural complexity and the manufacturing cost, and preventing the atomization drug delivery device from being locked wrongly.

摘要：【中文】本发明公开了一种低惯量高刚度绳索驱动直线运动装置，该直线运动单元由直线轴承、固定滑轮组、动滑轮组、导向轮、滑动轴、绳索及驱动装置构成。相对于目前的直线运动装置，本发明的机构有效的减小了直线运动单元的体积和重量，采用滑轮组的配合，提高了单条绳索驱动的刚度，因此具有重量轻、惯量低、刚度高的特点。可以适用于有尺寸、重量等有限制的场合，在自动化线、机器人、医疗器械等领域，替代体积重量较大的丝杠螺母、齿轮齿条、液压缸及气压缸等直线运动部件。 【EN】The invention discloses a low-inertia high-rigidity rope-driven linear motion device. Compared with the existing linear motion device, the mechanism of the invention effectively reduces the volume and weight of the linear motion unit, and improves the rigidity of single rope drive by adopting the cooperation of the pulley block, thereby having the characteristics of light weight, low inertia and high rigidity. The linear motion device can be suitable for occasions with limitations on size, weight and the like, and can replace linear motion components such as lead screw nuts, gear racks, hydraulic cylinders, pneumatic cylinders and the like with large volume and weight in the fields of automation lines, robots, medical instruments and the like.

摘要：【中文】本发明涉及医用生物敷料领域，具体地说，是一种可见光共聚交联多孔网状GelMA‑dHAMMA水凝胶的制备方法。具体的制作方法为：明胶通过甲基丙烯酸酐（MA）接枝改性形成甲基丙烯酰胺基明胶（GelMA）；人脱细胞羊膜（dHAM）通过甲基丙烯酸酐（MA）接枝改性形成dHAMMA，将接枝改性后的人脱细胞羊膜dHAMMA研磨成粉状；将形成的粉状GelMA与dHAMMA按质量比2：1混合，加入光引发剂L0290行光交联；将光交联后的复合水凝胶冻干，于γ辐照灭菌，于无菌条件下封装。本发明具有良好的抗炎、抗感染及生物相容性，能够促进口腔黏膜创口快速愈合。 【EN】The invention relates to the field of medical biological dressings, in particular to a preparation method of visible light copolymerization crosslinking porous reticular GelMA-dHAMMA hydrogel. The specific manufacturing method comprises the following steps: grafting and modifying gelatin by Methacrylic Anhydride (MA) to form methacrylamido gelatin (GelMA); grafting and modifying human acellular amniotic membrane (dHAM) by Methacrylic Anhydride (MA) to form dHAMMA, and grinding the grafted and modified human acellular amniotic membrane dHAMMA into powder; and mixing the formed powdery GelMA and dHAMMA according to the mass ratio of 2: 1, mixing, adding a photoinitiator L0290 for photocrosslinking; freeze-drying the photo-crosslinked composite hydrogel, sterilizing by gamma irradiation, and packaging under aseptic condition. The invention has good anti-inflammatory, anti-infection and biocompatibility, and can promote the rapid healing of the oral mucosa wound.