当前查询到87条专利与查询词 "Deng Xiaobing"相关,搜索用时0.4999529秒!排序方式:
发明专利:38实用新型: 44外观设计: 5
38 条,当前第 1-10 条 下一页 最后一页 返回搜索页
申请号:202010021008.1 公开号:CN111274369A 主分类号:G06F16/332
摘要:【中文】本申请适用于计算机技术领域,提供了一种英文单词的识别方法,包括:获取待识别的第一目标图像;其中,所述第一目标图像包括英文信息;将所述第一目标图像输入已训练的英文定位模型,定位所述第一目标图像包括的所述英文信息,得到所述英文信息对应的第二目标图像;将所述第二目标图像输入已训练的英文识别模型进行识别,得到所述第二目标图像对应的英文字符串;基于所述第二目标图像对应的英文字符串,确定所述英文字符串对应的英文单词。相对于现有技术,本申请提出的英文单词的识别方法提高了英文单词识别的准确率。 【EN】The application is suitable for the technical field of computers, and provides an English word recognition method, which comprises the following steps: acquiring a first target image to be identified; wherein the first target image comprises English information; inputting the first target image into a trained English positioning model, positioning the English information included in the first target image, and obtaining a second target image corresponding to the English information; inputting the second target image into a trained English recognition model for recognition to obtain an English character string corresponding to the second target image; and determining an English word corresponding to the English character string based on the English character string corresponding to the second target image. Compared with the prior art, the English word recognition method improves the accuracy of English word recognition.
详细信息 下载全文

申请号:201911275567.9 公开号:CN111027533A 主分类号:G06K9/20
摘要:【中文】本发明属于智能设备领域,公开了一种点读坐标的变换方法、系统、终端设备及存储介质,其方法包括获取点读书本图像;识别所述点读书本图像中的指示体对应的点读坐标;在数据库中查找与所述点读书本图像匹配的目标页面图像;将所述点读坐标、所述点读书本图像和所述目标页面图像输入预先训练好的深度学习坐标变换模型中,得到所述指示体在所述目标页面图像中的点读变换坐标。本发明无需进行三个阶段的任务计算,减少计算任务,且不需要进行手动提取特征,也不用手动进行参数调整,特别是在点读书本存在环境影响与书本成像问题上,具有抗干扰性,鲁棒性强,可提高点读的成功率。 【EN】The invention belongs to the field of intelligent equipment, and discloses a conversion method and a system of reading coordinates, terminal equipment and a storage medium, wherein the method comprises the steps of acquiring a reading book image; recognizing reading coordinates corresponding to an indicator in the reading book image; searching a target page image matched with the point-reading book image in a database; inputting the reading coordinates, the reading book image and the target page image into a pre-trained deep learning coordinate transformation model to obtain the reading transformation coordinates of the indicator in the target page image. The invention does not need to carry out task calculation in three stages, reduces calculation tasks, does not need to carry out manual feature extraction, does not need to carry out manual parameter adjustment, particularly has anti-interference performance and strong robustness on the problems of environmental influence and book imaging in reading books, and can improve the success rate of reading.
详细信息 下载全文

申请号:202010075902.7 公开号:CN111152464A 主分类号:B29C65/08
摘要:【中文】本发明提供了一种装配式数粒条及其制作方法,装配式数粒条包括数粒板,所述数粒板包括彼此上下扣合的上数粒板和下数粒板,所述上数粒板的两侧均设置有上数粒板通槽,所述下数粒板的两侧均设置有下数粒板通槽,所述上数粒板上设置有上数粒孔,所述下数粒板上设置有下数粒孔,所述上数粒孔与所述下数粒孔上下贯通。本发明采用多个数粒板依次首尾连接而成,可以制作出长度超长的数粒条,数粒条上具有使药丸等颗粒状的药品或食品逐个通过的数粒孔,数粒条可以根据需要尽可能地加长,而不再受大尺寸塑料零件加工精度低、易变形的限制。 【EN】The invention provides an assembled grain counting strip and a manufacturing method thereof, wherein the assembled grain counting strip comprises a grain counting plate, the grain counting plate comprises an upper grain counting plate and a lower grain counting plate which are vertically buckled with each other, an upper grain counting plate through groove is formed in each of two sides of the upper grain counting plate, a lower grain counting plate through groove is formed in each of two sides of the lower grain counting plate, an upper grain counting hole is formed in the upper grain counting plate, a lower grain counting hole is formed in the lower grain counting plate, and the upper grain counting hole is vertically communicated with the lower grain counting hole. The invention is formed by connecting a plurality of particle counting plates end to end in sequence, can manufacture a plurality of particles with super-long length, the particle counting strips are provided with particle counting holes which enable granular medicines or foods such as pills to pass through one by one, and the particle counting strips can be lengthened as much as possible according to the requirements without being limited by low processing precision and easy deformation of large-size plastic parts.
详细信息 下载全文

申请号:201911293084.1 公开号:CN111079771A 主分类号:G06K9/46
摘要:【中文】本发明属于点读识别领域,公开了一种点读图像的特征提取方法、系统、终端设备及存储介质。其方法包括:获取点读页面的点读图像;将所述点读图像变换为Umat对象变量;调用Opencl程序接口对所述Umat对象变量进行并行处理,提取所述点读图像中的特征点和所述特征点的属性。本发明将Opencl异构平台应用于点读特征的提取,充分利用现有硬件资源进行加速,在提取特征点时通过并行处理,可解决高分辨率图像特征点提取的耗时问题,而且不会减少点读特征点的提取数量,便于后续的搜索匹配和点读坐标变换,提高点读效果。 【EN】The invention belongs to the field of reading identification, and discloses a method and a system for extracting characteristics of a reading image, terminal equipment and a storage medium. The method comprises the following steps: acquiring a reading image of a reading page; converting the point-reading image into a Umat object variable; and calling an Opencl program interface to perform parallel processing on the Umat object variable, and extracting the feature points in the point reading image and the attributes of the feature points. The invention applies the Opencl heterogeneous platform to the extraction of the click-to-read characteristics, fully utilizes the existing hardware resources for acceleration, solves the time-consuming problem of the extraction of the high-resolution image characteristic points by parallel processing when extracting the characteristic points, does not reduce the extraction quantity of the click-to-read characteristic points, facilitates the subsequent search matching and click-to-read coordinate transformation, and improves the click-to-read effect.
详细信息 下载全文

申请号:201811067748.8 公开号:CN110896588A 主分类号:H05K1/11
申请人:【中文】深圳市广和通无线股份有限公司【EN】FIBOCOM WIRELESS INC. 申请日:2018.09.13 公开日:2020.03.20
摘要:【中文】本发明涉及电路板技术领域,特别是涉及一种印制电路板,包括金手指和安装板,金手指设置于安装板的一侧,安装板用于放置电子元件,金手指和安装板均为多层结构,安装板的层数小于金手指的层数,因此安装板的厚度小于所述金手指的厚度,从而让金手指与安装板的连接处呈现阶梯状,这样既能满足标准金手指插卡需要,又能满足安装板安装了电子元器件后的限高要求。 【EN】The invention relates to the technical field of circuit boards, in particular to a printed circuit board which comprises a golden finger and a mounting plate, wherein the golden finger is arranged on one side of the mounting plate, the mounting plate is used for placing electronic elements, the golden finger and the mounting plate are of a multilayer structure, and the number of layers of the mounting plate is smaller than that of the golden finger, so that the thickness of the mounting plate is smaller than that of the golden finger, and the connecting part of the golden finger and the mounting plate is in a step shape, so that the requirements of standard golden finger inserting and inserting can be met, and the requirement of height limitation after electronic elements are mounted on.
详细信息 下载全文

申请号:201911108498.2 公开号:CN111008295A 主分类号:G06F16/583
摘要:【中文】本申请实施例公开了一种书页检索方法、装置、电子设备和存储介质,该方法包括:获取封面图像和当前书页图像;确定封面图像的矢量数据和当前书页图像的矢量数据,其中,封面图像的矢量数据包括封面全局特征数据和封面局部特征数据,当前书页图像的矢量数据包括书页全局特征数据和书页局部特征数据;将封面图像的矢量数据输入至预先存储的封面数据库确定目标书籍;将当前书页图像的矢量数据输入至预先存储的目标书籍对应的书页数据库进行检索,以确定目标书页。提高了书页检索的检索速度、准确率和应用范围。 【EN】The embodiment of the application discloses a page retrieval method, a page retrieval device, electronic equipment and a storage medium, wherein the page retrieval method comprises the following steps: acquiring a cover image and a current page image; determining vector data of a cover image and vector data of a current page image, wherein the vector data of the cover image comprises global cover characteristic data and local cover characteristic data, and the vector data of the current page image comprises global page characteristic data and local page characteristic data; inputting the vector data of the cover image into a pre-stored cover database to determine a target book; and inputting the vector data of the current page image into a page database corresponding to a pre-stored target book for retrieval so as to determine the target page. The retrieval speed, the accuracy and the application range of page retrieval are improved.
详细信息 下载全文

申请号:201911267512.3 公开号:CN111177441A 主分类号:G06F16/532
摘要:【中文】本发明实施例涉及教育技术领域,公开了一种搜题方法及电子设备,该方法包括:对终端设备上传的题目照片进行裁剪,获得题目照片中的图片的文件流;分别按照第一转换方式和第二转换方式将该文件流转换成图片对应的第一类编码和第二类编码;再分别以图片对应的第一类编码和第二类编码为依据进行搜索,各获得若干图片标识码和每一图片标识码对应的图片置信度;从搜索获得的图片标识码中确定出指定数量的图片标识码为目标图片标识码;获取与目标图片标识码相匹配的题目信息。可见,电子设备利用题目照片中的图片进行搜题,无需依赖文字进行搜题,能够提高搜题结果的准确率。 【EN】The embodiment of the invention relates to the technical field of education, and discloses a question searching method and electronic equipment, wherein the method comprises the following steps: cutting a subject photo uploaded by a terminal device to obtain a file stream of pictures in the subject photo; converting the file stream into a first type of code and a second type of code corresponding to the picture according to a first conversion mode and a second conversion mode respectively; searching according to the first type of codes and the second type of codes corresponding to the pictures respectively to obtain a plurality of picture identification codes and picture confidence degrees corresponding to the picture identification codes; determining a specified number of picture identification codes as target picture identification codes from the searched picture identification codes; and acquiring the subject information matched with the identification code of the target picture. Therefore, the electronic equipment searches for the questions by using the pictures in the question photos without depending on characters, and the accuracy of the question searching results can be improved.
详细信息 下载全文

申请号:201911390525.X 公开号:CN111079310A 主分类号:G06F30/20
摘要:【中文】本发明公开了一种湍流区域识别方法,涉及计算流体动力学领域,包括以下步骤:步骤1:获取需要进行流动分析的流场;步骤2:计算所述流场的湍流区域识别值σ;步骤3:判断所述湍流区域识别值σ是否为0,若是则该流场为湍流区域,若否则不是;所述湍流区域识别值σ基于局部雷诺数与涡拉伸函数得到,本发明提供了一种不依赖涡粘性系数的湍流区域识别方法,使得湍流区域识别的方法具有更广泛的适用范围。 【EN】The invention discloses a turbulent flow region identification method, which relates to the field of computational fluid dynamics and comprises the following steps: step 1: acquiring a flow field to be subjected to flow analysis; step 2: calculating a turbulence region identification value sigma of the flow field; and step 3: judging whether the identification value sigma of the turbulent flow area is 0, if so, the flow field is the turbulent flow area, otherwise, the flow field is not the turbulent flow area; the turbulence area identification value sigma is obtained based on the local Reynolds number and the vortex stretch function, and the invention provides the turbulence area identification method independent of the vortex viscosity coefficient, so that the turbulence area identification method has a wider application range.
详细信息 下载全文

申请号:201911218179.7 公开号:CN110984809A 主分类号:E06B5/16
摘要:【中文】本发明公开了一种具有避难功能的防火窗,包括第一安装箱,所述第一安装箱的内部底端通过螺栓安装有温度传感器,所述温度传感器电性连接有中央处理器,所述中央处理器的电性连接有第一伺服电机,所述中央处理器和第一伺服电机均通过螺栓固定在第一安装箱的内部底端,所述第一伺服电机的输出轴端部固定有安装框,所述安装框内安装有玻璃,所述安装框远离第一伺服电机的一端端部固定有第一转轴,所述第一转轴的上端端部在安装架内转动,所述安装架位于安装框的外侧,所述安装架的上端端部设置有挡火机构。发生火灾时,实现窗户自动关闭,外侧有防火布阻挡火焰,为人们提供更长的逃生时间,起到较好的避难效。 【EN】The invention discloses a fireproof window with a refuge function, which comprises a first installation box, wherein a temperature sensor is installed at the inner bottom end of the first installation box through a bolt, the temperature sensor is electrically connected with a central processing unit, the central processing unit is electrically connected with a first servo motor, the central processing unit and the first servo motor are both fixed at the inner bottom end of the first installation box through bolts, an installation frame is fixed at the end part of an output shaft of the first servo motor, glass is installed in the installation frame, a first rotating shaft is fixed at the end part, far away from the first servo motor, of the installation frame, the upper end part of the first rotating shaft is arranged in an installation frame, the installation frame is positioned on the outer side of the installation frame in a rotating manner, and a fire blocking mechanism is arranged at the upper end part of the installation. When a fire disaster happens, the window is automatically closed, the fire-proof cloth is arranged on the outer side of the window to block flames, so that longer escape time is provided for people, and a better refuge effect is achieved.
详细信息 下载全文

申请号:202010191241.4 公开号:CN111236543A 主分类号:E04D3/16
摘要:【中文】本发明公开了一种金属屋面抗风防坠落的施工方法及结构,包括至少两块需要依次连接一起的金属屋面板,在金属屋面板下方设置有高强铝合金支座,该高强铝合金支座通过不锈钢自攻自钻钉固定在钢结构梁上,本发明确保板端铝合金夹具与金属屋面之间存在一定的空间,不影响屋面排水,同时可确保整个屋面板端部不因强风原因而掀起坠落,施工较为方便快捷安全,整个屋面板更加稳固和安全。 【EN】The invention discloses a construction method and a structure for wind resistance and falling prevention of a metal roof, which comprises at least two metal roof boards which need to be connected together in sequence, wherein a high-strength aluminum alloy support is arranged below the metal roof boards and is fixed on a steel structure beam through stainless steel self-tapping self-drilling nails.
详细信息 下载全文

38 条,当前第 1-10 条 下一页 最后一页 返回搜索页