WO2023141900A1 - 新闻图文类数据知识图谱的建立方法、装置、设备及介质 - Google Patents

新闻图文类数据知识图谱的建立方法、装置、设备及介质 Download PDF

Info

Publication number
WO2023141900A1
WO2023141900A1 PCT/CN2022/074424 CN2022074424W WO2023141900A1 WO 2023141900 A1 WO2023141900 A1 WO 2023141900A1 CN 2022074424 W CN2022074424 W CN 2022074424W WO 2023141900 A1 WO2023141900 A1 WO 2023141900A1
Authority
WO
WIPO (PCT)
Prior art keywords
news
key information
text
image
related personnel
Prior art date
Application number
PCT/CN2022/074424
Other languages
English (en)
French (fr)
Inventor
张森
黄学涛
辛伏炎七妹
曾勇华
谭卓
许云侠
唐跃文
袁埜
张润南
Original Assignee
基建通(三亚)国际科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 基建通(三亚)国际科技有限公司 filed Critical 基建通(三亚)国际科技有限公司
Priority to PCT/CN2022/074424 priority Critical patent/WO2023141900A1/zh
Publication of WO2023141900A1 publication Critical patent/WO2023141900A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/26Visual data mining; Browsing structured data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36Creation of semantic tools, e.g. ontology or thesauri

Definitions

  • the present application relates to the field of computer technology, and in particular to a method, device, equipment and medium for establishing a knowledge map of news graphic and text data.
  • the embodiment of the present application provides a method for establishing a knowledge map of news graphic data, including:
  • the news image includes news text and a news scene image that contains the face image of the news-related personnel
  • face recognition is performed on the face image in the news scene image, and the first part of the news text is extracted.
  • Key information and generate news records of news related personnel according to the face recognition result and the first key information;
  • the first key information includes time, location and news topic
  • the second key information includes time, location, news topic and name.
  • the face recognition result is that the news scene image contains a face image of a known first target person
  • the news correlation generated according to the face recognition result and the first key information Press records of personnel, including:
  • the personal information includes name and title.
  • the method further includes:
  • the personal information includes name and title.
  • the method also includes:
  • the personal information of the second target person is associated with the face image corresponding to the second target person and stored.
  • the acquisition of news images and texts recorded by the server includes:
  • the server When the server receives the latest uploaded news graphic from the client, acquiring the latest uploaded news graphic from the server; or
  • the latest uploaded news images and texts of the client are obtained from the server at preset time intervals.
  • the method further includes:
  • the news records contained in the knowledge map of news graphic data are stored in the file directory in chronological order.
  • the performing face recognition on the face image in the news scene image includes:
  • the embodiment of the present application provides a device for establishing a knowledge map of news graphic and text data, including:
  • the obtaining unit is used to obtain news images and texts recorded by the server;
  • a judging unit configured to judge whether the news graphics and texts include news text and news scene images containing face images of news-related personnel
  • the execution unit is used to perform face recognition on the face images in the news scene images when the news images include news text and news scene images containing news-related personnel face images, and extract the news the first key information in the text, and generate news records of news related personnel according to the face recognition result and the first key information;
  • the news graphic includes only news text, extracting second key information in the news text, and generating news records of news related personnel according to the second key information;
  • the first key information includes time, location and news topic
  • the second key information includes time, location, news topic and name.
  • an embodiment of the present application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete communication with each other through the bus;
  • the processor is used to execute the program stored in the memory to realize the following processes:
  • the news image includes news text and a news scene image that contains the face image of the news-related personnel
  • face recognition is performed on the face image in the news scene image, and the first part of the news text is extracted.
  • Key information and generate news records of news related personnel according to the face recognition result and the first key information;
  • the first key information includes time, location and news topic
  • the second key information includes time, location, news topic and name.
  • an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored in the storage medium, and when the computer program is executed by a processor, the following process is implemented:
  • the news image includes news text and a news scene image that contains the face image of the news-related personnel
  • face recognition is performed on the face image in the news scene image, and the first part of the news text is extracted.
  • Key information and generate news records of news related personnel according to the face recognition result and the first key information;
  • the first key information includes time, location and news topic
  • the second key information includes time, location, news topic and name.
  • the news pictures and texts recorded by the server By obtaining the news pictures and texts recorded by the server, if the news pictures and texts include the news text and the news scene images containing the face images of news-related personnel, face recognition is performed on the face images in the news scene images to extract the news text The first key information in the news, and according to the face recognition result and the first key information to generate the news records of the news related personnel; if the news graphic only includes the news text, then extract the second key information in the news text, and according to The second key information generates news records of news related personnel, and then establishes a news graphic data knowledge graph corresponding to news related personnel based on all news records of news related personnel. In this way, face recognition and text extraction can be combined to automatically generate a news graphic data knowledge map containing all news records of enterprise staff, which is convenient for the statistics of news records of enterprise staff participating in various affairs.
  • FIG. 1 is a flow chart of a method for establishing a knowledge map of news graphic and text data provided by an embodiment of the present application.
  • FIG. 2 is a flow chart of a method for establishing a knowledge graph of news graphic data provided by another embodiment of the present application.
  • Fig. 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • Fig. 4 is a schematic structural diagram of a device for establishing a knowledge map of news graphic and text data provided by an embodiment of the present application.
  • the embodiment of the present application provides a method, device, equipment and medium for establishing a knowledge map of news graphic and text data.
  • the method for establishing a knowledge map of news graphic and text data , devices, equipment and media can automatically generate a news graphic data knowledge map of all news records related to enterprise staff, which is convenient for statistics on various affairs that enterprise staff participate in.
  • the method for establishing a knowledge map of news graphic data can be applied to a server or a user terminal, and the user terminal can be, but not limited to, a personal computer, a smart phone, a tablet computer, a personal digital assistant (personal digital assistant) , PDA) etc.
  • FIG. 1 it is a flowchart of a method for establishing a knowledge map of news graphic and text data provided by the embodiment of the present application.
  • the method for establishing a knowledge map of news graphic and text data may include the following steps:
  • Step S101 acquiring news images and texts recorded by the server.
  • the news pictures and texts recorded by the server may be news about important meetings held by enterprise staff or leaders, reports on some important affairs, etc., which are not specifically limited in this embodiment of the application.
  • the news graphics and texts may include news texts, and may also include news texts and news scene images including face images of news-related personnel at the same time.
  • the content in the news text may include information such as the time, place, and news topic corresponding to the firm.
  • the news pictures and texts on the server side can be updated in real time according to the uploads from the client.
  • Text for statistics That is, when the server receives the latest news image and text uploaded by the client, it will obtain the latest uploaded news image and text from the server.
  • the latest news images and texts uploaded by the client can be obtained from the server at preset time intervals.
  • the latest news graphics and texts uploaded by the client can be obtained from the server every other day.
  • Step S102 judging whether the news graphics and texts include news scene images including face images of news-related personnel in addition to the news text, if yes, execute step S103; if not, execute step S104.
  • Step S103 performing face recognition on the face images in the news scene images, extracting the first key information in the news text, and generating news records of news related personnel according to the face recognition results and the first key information.
  • news graphics include news text and news scene images containing news-related personnel face images
  • face recognition is performed on the face images in the news scene images, and the first key information in the news text is extracted.
  • the face recognition results and the first key information generate the news records of the relevant persons in the news.
  • the face feature library with the face features of all personnel of the enterprise is pre-established. After obtaining the news pictures and texts recorded by the server, face recognition can be performed on the face images in the news scene images, so that Identify whether the image of the news scene contains the face image of the known first target person.
  • performing face recognition on face images in live news images may include the following steps S1031-S1035.
  • Step S1031 segment the news scene image to obtain multiple sub-images.
  • cutting the news scene image can be, but not limited to, adopting: python opencv script; that is, using python opencv script to continuously segment the news scene image into small areas, and then store the segmented area as a new small picture , and the stored small pictures are sub-pictures.
  • the number and size of the segmented sub-images can be set according to the actual situation.
  • the sizes of the divided sub-images may be the same or different, and the number of divided sub-images may be 4, 9, 16 and so on.
  • Step S1032 performing face feature extraction on each of the multiple sub-images to obtain at least one face feature map.
  • face images in live news images may be segmented into the same sub-image, and different regions or multiple face images of the same face image may also be segmented into different sub-images.
  • face feature extraction one or more face feature maps can be obtained.
  • Step S1033 performing contrast frequency domain range enhancement processing on at least one face feature map through an adaptive contrast enhancement algorithm.
  • the adaptive contrast enhancement (Adaptive Contrast Enhancement, ACE) algorithm is derived from the retinex algorithm. Its main principle is to correct the final pixel value by differentially calculating the relative shadow relationship between the target point and the surrounding pixels, which has a good enhancement effect. The detailed process It will not be described in detail here.
  • ACE Adaptive Contrast Enhancement
  • Step S1034 perform super-resolution reconstruction on at least one face feature map after enhancement processing.
  • the process of super-resolution reconstruction is to expand the number of channels of the face feature map first, then perform a convolution operation on the feature map corresponding to the expanded channel number, and finally rearrange the pixels to obtain a high-resolution image.
  • the feature map is used for subsequent face recognition to improve the accuracy of face recognition.
  • Step S1035 perform face recognition on at least one face feature map after super-resolution reconstruction.
  • the first key information in the news text may be extracted through a natural language processing (Natural Language Processing, NLP) algorithm.
  • the first key information includes, but is not limited to, time, location, and news topic.
  • the face images in the news scene images may be face images of known persons or face images of unknown persons.
  • the server records the personal information of all staff members.
  • Personal information can include name, work unit and position, and can also include personal location and personal profile and other information.
  • a news record of the first target person may be generated according to the first key information and the personal information of the first target person.
  • the generated news records at least include the name, work unit and position of the first target person, as well as the time, place and news topic of the involved affairs, and may also include the location and personal profile of the first target person.
  • the first target person may be one or more, which is not specifically limited in this embodiment of the present application.
  • the image of the news scene contains known first target persons including the first target person A and the first target person B, and by extracting the key information in the news text, the obtained
  • the time is January 1, 2021
  • the location is CD City
  • the topic of the news is X project research.
  • the name of the first target person A is Zhang San
  • his job title is the chief engineer of the project.
  • the name of the first target person B is Li Si
  • his job title is the project manager.
  • the news records of the first target person A can include as follows:
  • the news record of the first target person B may include the following:
  • the news records of the first target person A and the first target person B may also include other information such as the location of the individual and personal profile.
  • the face recognition process there may also be unrecognizable face images. If it is recognized that the face image of the unknown second target person is included in the news scene image, the face image of the second target person can be extracted, and the user can manually edit the personal information of the second target person and upload it. At this time, the user can receive To the personal information of the second target person uploaded by the user, and then according to the first key information and the personal information of the second target person, a news record of the second target person is generated.
  • the edited personal information of the second target person may include name, work unit and position, and may also include personal location and personal profile and other information.
  • Step S104 extracting the second key information in the news text, and generating news records of news related personnel according to the second key information.
  • the second key information in the news text can be extracted, and the news records of the news related personnel can be generated according to the second key information .
  • the second key information includes, but is not limited to, time, location, news topic and name.
  • the news record of the relevant personnel of the news generation includes the name of the person involved in the transaction, the time and place of the participation in the transaction, and the topic of the news, etc., and may also include the position, location, and personal profile of the person involved in the transaction.
  • the recognized face feature map of the first target person can also be stored in the face feature database, or by identifying The face feature map of the first target person replaces the original face feature map in the face feature database, so as to realize the iterative update of the face feature database.
  • the personal information of the second target person can also be associated with the face image corresponding to the second target person and stored in the face feature database, so as to realize the recognition of the face image.
  • Manual update of the database can also be associated with the face image corresponding to the second target person and stored in the face feature database, so as to realize the recognition of the face image.
  • step S105 a knowledge map of news graphic and text data corresponding to the news related personnel is established according to all news records of the news related personnel.
  • the established knowledge graph of news graphic and text data includes all news records of the relevant news personnel.
  • the news records contained in the news graph-text data knowledge graph can also be stored in the news graph corresponding to the news-related personnel.
  • the news records contained in the news graphic data knowledge map are stored in chronological order in the file directory. In this way, it is convenient to search for various news records of the enterprise staff.
  • FIG. 2 it is a flow chart of another method for establishing a knowledge map of news graphic and text data provided by the embodiment of the present application, which may include the following steps:
  • Step S201 acquiring news images and texts recorded by the server.
  • Step S202 judging whether the news image and text include the news site image including the face image of the news-related person in addition to the news text, if not, go to step S203; if yes, go to step S204.
  • Step S203 extracting the second key information in the news text, and generating news records of news related personnel according to the second key information.
  • Step S204 performing face recognition on the face images in the news scene images, and extracting the first key information in the news text.
  • Step S205 judging whether the face recognition is successful, if yes, execute step S206, if not, execute step S207.
  • Step S206 generating a news record of the relevant person according to the face recognition result and the first key information.
  • Step S207 extract the personal information of the unidentified person through the relationship extraction technology, generate a news record of the relevant person in the news according to the extracted personal information of the unidentified person and the first key information, and store the extracted personal information of the unidentified person stored in the local database.
  • the personal information of the unidentified persons can be extracted through the relationship extraction technology, and based on the extracted personal information of the unidentified persons and the first key information Generate news records of relevant people in the news, and store the extracted personal information and avatars of unidentified people in the local database for subsequent face recognition.
  • the news pictures and texts recorded by the server if the news pictures and texts include news texts and news scene images containing the face images of news-related personnel, face recognition is performed on the face images in the news scene images , extract the first key information in the news text, and generate the news records of the news related personnel according to the face recognition result and the first key information; if the news graphic only includes the news text, then extract the second key information in the news text Key information, and generate news records of news related personnel according to the second key information, and then establish a news graphic data knowledge graph corresponding to news related personnel based on all news records of news related personnel.
  • face recognition and text extraction can be combined to automatically generate a news graphic data knowledge graph containing all news records of enterprise staff, which is convenient for the statistics of enterprise staff's participation in various affairs.
  • multiple sub-images are obtained by segmenting the news scene image, and at least one face feature map is obtained by performing face feature extraction on each sub-image, Then, at least one face feature map is subjected to contrast frequency domain enhancement processing through an adaptive contrast enhancement algorithm, and super-resolution reconstruction is performed during the enhancement process, so that a high-resolution feature map can be obtained for subsequent face Recognition, improve the accuracy of face recognition.
  • the knowledge map of news graphic and text data of each enterprise staff can also be stored separately, so as to facilitate the search of various news records of enterprise staff.
  • the face feature map of the identified first target person can also be stored in the face feature database, or by identifying the face image of the first target person
  • the face feature map replaces the original face feature map in the face feature library, so as to realize the iterative update of the face feature library.
  • manual updating of the face database can also be realized through editing.
  • Fig. 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory.
  • the memory may include a memory, such as a high-speed random-access memory (Random-Access Memory, RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
  • RAM random-Access Memory
  • non-volatile memory such as at least one disk memory.
  • the electronic device may also include hardware required by other services.
  • the processor, the network interface and the memory can be connected to each other through an internal bus, which can be an ISA (Industry Standard Architecture, industry standard architecture) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnection standard) bus or an EISA (Extended Industry Standard Architecture, extended industry standard architecture) bus, etc.
  • the bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one double-headed arrow is used in FIG. 3 , but it does not mean that there is only one bus or one type of bus.
  • Memory for storing programs.
  • the program may include program code, and the program code includes computer operation instructions.
  • Storage which can include internal memory and nonvolatile storage, provides instructions and data to the processor.
  • the processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a device for establishing a knowledge map of news graphic data on a logical level.
  • the processor executes the program stored in the memory, and is specifically used to perform the following operations:
  • the news image includes news text and a news scene image that contains the face image of the news-related personnel
  • face recognition is performed on the face image in the news scene image, and the first part of the news text is extracted.
  • Key information and generate news records of news related personnel according to the face recognition result and the first key information;
  • the first key information includes time, location and news topic
  • the second key information includes time, location, news topic and name.
  • a processor may be an integrated circuit chip with signal processing capabilities.
  • each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software.
  • the above-mentioned processor can be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; it can also be a digital signal processor (Digital Signal Processor, DSP), a dedicated integrated Circuit (Application Specific Integrated Circuit, ASIC), field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • DSP Digital Signal Processor
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • the steps of the method disclosed in conjunction with one or more embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor.
  • the software module may be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register.
  • the storage medium is located in the memory, and the processor reads the information in the memory, and completes the steps of the above method in combination with its hardware.
  • the electronic device can also execute the methods shown in FIG. 1 and FIG. 2 , and realize the functions of the apparatus for establishing knowledge graphs of news graphic and text data in the embodiment shown in FIG. 3 , which will not be repeated here in the embodiments of the present application.
  • the electronic device of the present application does not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is to say, the execution subject of the following processing flow is not limited to each logic unit, It can also be a hardware or logic device.
  • the embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores one or more programs, and the one or more programs include instructions, and when the instructions are used by a portable electronic device including multiple application programs During execution, the portable electronic device can be made to execute the method of the embodiment shown in FIG. 1, and is specifically used to perform the following operations:
  • the news image includes news text and a news scene image that contains the face image of the news-related personnel
  • face recognition is performed on the face image in the news scene image, and the first part of the news text is extracted.
  • Key information and generate news records of news related personnel according to the face recognition result and the first key information;
  • the first key information includes time, place and news topic
  • the second key information includes time, place, news topic and name.
  • Fig. 4 is a schematic structural diagram of an apparatus for establishing a knowledge map of news graphic and text data provided by an embodiment of the present application. Please refer to FIG. 4.
  • the device for establishing a knowledge map of news graphic and text data includes:
  • the obtaining unit is used to obtain news images and texts recorded by the server;
  • a judging unit configured to judge whether the news graphics and texts include news text and news scene images containing face images of news-related personnel
  • the execution unit is used to perform face recognition on the face images in the news scene images when the news images include news text and news scene images containing news-related personnel face images, and extract the news the first key information in the text, and generate news records of news related personnel according to the face recognition result and the first key information;
  • the news graphic includes only news text, extracting second key information in the news text, and generating news records of news related personnel according to the second key information;
  • the first key information includes time, location and news topic
  • the second key information includes time, location, news topic and name.
  • a typical implementing device is a computer.
  • the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or Combinations of any of these devices.
  • Computer-readable media including both permanent and non-permanent, removable and non-removable media, may be implemented by any method or technology for storage of information.
  • Information may be computer readable instructions, data structures, modules of a program, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Flash memory or other memory technology, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cartridge, tape magnetic disk storage or other magnetic storage device or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
  • computer-readable media excludes transitory computer-readable media, such as modulated data signals and carrier waves.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Processing Or Creating Images (AREA)

Abstract

本申请提供一种新闻图文类数据知识图谱的建立方法、装置、设备及介质,涉及计算机技术领域。方法包括获取新闻图文;如果新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对新闻现场图像中的人脸图像进行人脸识别,提取新闻文本中的第一关键信息,根据人脸识别结果和第一关键信息生成新闻相关人员的新闻记录;如果新闻图文中包括新闻文本,提取新闻文本中的第二关键信息,根据第二关键信息生成新闻相关人员的新闻记录,根据新闻相关人员的所有新闻记录建立对应的新闻图文类数据知识图谱。本申请提供的方法、装置、设备及介质可自动生成企业工作人员的新闻图文类数据知识图谱,方便对企业工作人员参与事务的新闻记录进行统计。

Description

新闻图文类数据知识图谱的建立方法、装置、设备及介质 技术领域
本申请涉及计算机技术领域,尤其涉及一种新闻图文类数据知识图谱的建立方法、装置、设备及介质。
背景技术
在企业管理中,企业工作人员开展一些重要事务时,常常会以新闻图文的方式将相关内容记录在企业的系统后台中,并可能在新闻图文中附上相关人员的现场图像,以便对企业工作人员参与的事务进行统计以作为参与工作事务的凭证。
目前,对企业工作人员参与事务的统计,常采用的方式是由后台专门的人员进行人工统计,然而采用这样的方式需要耗费大量的人力资源,给企业工作人员参与事务的统计带来了极大的不便。
因此,如何提供一种有效的方案,以方便对企业工作人员参与的各项事务进行统计,已成为现有技术中一亟待解决的难题。
发明内容
第一方面,本申请实施例提供了一种新闻图文类数据知识图谱的建立方法,包括:
获取服务器记录的新闻图文;
如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
在一个可能的设计中,当人脸识别结果为所述新闻现场图像中包含已知的第一目标人员的人脸图像时,所述根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录,包括:
根据所述第一关键信息和所述第一目标人员的个人信息生成所述第一目标人员的新闻记录;
其中,所述个人信息包括姓名和职务。
在一个可能的设计中,当所述人脸识别结果为所述新闻现场图像中包含未知的第二目标人员的人脸图像时,所述方法还包括:
接收用户上传的所述第二目标人员的个人信息;
根据所述第一关键信息和所述第二目标人员的个人信息生成所述第二目标人员的新闻记录;
其中,所述个人信息包括姓名和职务。
在一个可能的设计中,所述方法还包括:
将所述第二目标人员的个人信息与所述第二目标人员所对应的人脸图像关联后进行存储。
在一个可能的设计中,所述获取服务器记录的新闻图文,包括:
当所述服务器接收到用户端最新上传的新闻图文时,从所述服务器获取所述最新上传的新闻图文;或
每隔预设的时间间隔,从所述服务器获取用户端最新上传的新闻图文。
在一个可能的设计中,在建立与所述新闻相关人员对应的新闻图文类数据知识图谱之后,所述方法还包括:
将新闻图文类数据知识图谱所包含的新闻记录存储至与所述新闻相关人员对应的文件目录中;
其中,新闻图文类数据知识图谱所包含的新闻记录在文件目录中按照时间先后顺序进行存储。
在一个可能的设计中,所述对所述新闻现场图像中的人脸图像进行人脸识别,包括:
对所述新闻现场图像进行分割,得到多张子图像;
对所述多张子图像中的每张子图像进行人脸特征提取,得到至少一张人脸特征图;
通过自适应对比度增强算法对所述至少一张人脸特征图进行对比度频域范围增强处理;
对增强处理后的所述至少一张人脸特征图进行超分辨率重建;
对超分辨率重建后的所述至少一张人脸特征图进行人脸识别。
第二方面,本申请实施例提供了一种新闻图文类数据知识图谱的建立装置,包括:
获取单元,用于获取服务器记录的新闻图文;
判断单元,用于判断所述新闻图文中是否包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像;
执行单元,用于当所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像时,对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
当所述新闻图文中仅包括新闻文本时,提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;以及
根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
第三方面,本申请实施例提供了一种电子设备,包括处理器、通信接口、存储器和通信总线,其中,处理器,通信接口,存储器通过总线完成相互间的通信;
存储器,用于存放计算机程序;
处理器,用于执行存储器上所存放的程序,实现以下流程:
获取服务器记录的新闻图文;
如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
第四方面,本申请实施例提供了一种计算机可读存储介质,所述存储介质内存储有计算机程序,所述计算机程序被处理器执行时实现以下流程:
获取服务器记录的新闻图文;
如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关 键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
本申请一个或多个实施例采用的上述至少一个技术方案能够达到以下有益效果:
通过获取服务器记录的新闻图文,如果新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像时,则对新闻现场图像中的人脸图像进行人脸识别,提取出新闻文本中的第一关键信息,并根据人脸识别结果和第一关键信息生成新闻相关人员的新闻记录;如果新闻图文中仅包括新闻文本,则提取出新闻文本中的第二关键信息,并根据第二关键信息生成新闻相关人员的新闻记录,然后根据新闻相关人员的所有新闻记录建立与新闻相关人员对应的新闻图文类数据知识图谱。如此能够结合人脸识别和文本提取,自动生成包含企业工作人员所有新闻记录的新闻图文类数据知识图谱,方便对企业工作人员参与各项事务的新闻记录进行统计。
附图说明
此处所说明的附图用来提供对本文件的进一步理解,构成本文件的一部分,本文件的示意性实施例及其说明用于解释本文件,并不构成对本文件的不当限定。在附图中:
图1为本申请一个实施例提供的新闻图文类数据知识图谱的建立方法的流程图。
图2为本申请另一个实施例提供的新闻图文类数据知识图谱的建立方法的流程图。
图3为本申请一个实施例提供的电子设备的结构示意图。
图4为本申请一个实施例提供的新闻图文类数据知识图谱的建立装置的结 构示意图。
具体实施方式
为了方便对企业工作人员参与的各项事务进行统计,本申请实施例提供了一种新闻图文类数据知识图谱的建立方法、装置、设备及介质,该新闻图文类数据知识图谱的建立方法、装置、设备及介质能够自动生成与企业工作人员相关的所有新闻记录的新闻图文类数据知识图谱,方便对企业工作人员参与的各项事务进行统计。
本申请实施例提供的新闻图文类数据知识图谱的建立方法可应用于服务器或用户终端,所述用户终端可以是,但不限于个人电脑、智能手机、平板电脑、个人数字助理(personal digital assistant,PDA)等。
下面将对本申请实施例提供的应用于用户终端的新闻图文类数据知识图谱的建立方法进行详细说明。可以理解,所述执行主体并不构成对本申请实施例的限定。
如图1所示,是本申请实施例提供的新闻图文类数据知识图谱的建立方法的流程图,新闻图文类数据知识图谱的建立方法可以包括如下步骤:
步骤S101,获取服务器记录的新闻图文。
本申请实施例中,服务器记录的新闻图文可以是,企业工作人员或领导开展重要会议的新闻、开展一些重要事务的报告等,本申请实施例中不做具体限定。
其中,所述新闻图文中可以包括新闻文本,也可以同时包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像。新闻文本中的内容可以包括事务所对应的时间、地点以及新闻主题等信息。
本申请实施例中,服务器端的新闻图文可根据客户端的上传实时更新,为便于及时对新闻记录进行统计,可以是服务器端每接收到一条上传的新闻图文,就对该最新上传的新闻图文进行统计。即当服务器接收到用户端最新上传的新闻图文时,就会从服务器获取该最新上传的新闻图文。
可以理解的,在其他的一些实施例中,也可以每隔预设的时间间隔,从服务器获取用户端最新上传的新闻图文。例如,可以每隔一天从服务器获取一次用户端最新上传的新闻图文。
步骤S102,判断新闻图文中除新闻文本外,是否还包括包含新闻相关人员人脸图像的新闻现场图像,如果是,则执行步骤S103;如果否,则执行步骤S104。
步骤S103,对新闻现场图像中的人脸图像进行人脸识别,提取出新闻文本中的第一关键信息,并根据人脸识别结果和第一关键信息生成新闻相关人员的新闻记录。
如果新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对新闻现场图像中的人脸图像进行人脸识别,提取出新闻文本中的第一关键信息,并根据人脸识别结果和第一关键信息生成新闻相关人员的新闻记录。
本申请实施例中,预先建立有企业所有人员人脸特征的人脸特征库,在获取到服务器记录的新闻图文后,可对其中的新闻现场图像中的人脸图像进行人脸识别,从而识别出新闻现场图像中是否包含已知的第一目标人员的人脸图像。
本申请实施例中,对新闻现场图像中的人脸图像进行人脸识别可以包括如下的步骤S1031-S1035。
步骤S1031,对新闻现场图像中进行分割,得到多张子图像。
在本实施例中,对新闻现场图像进行切割可以但不限于采用:python opencv脚本;即利用python opencv脚本对新闻现场图像不断的切分小区域,然后将切分的区域存储为新的小图片,存储的小图片则为子图像。
对新闻现场图像进行分割时,分割的子图像的数量和尺寸可根据实际情况设定。分割的子图像的尺寸可以相同或不同,分割的子图像的数量可以是4张、9张、16张等。
步骤S1032,对多张子图像中的每张子图像进行人脸特征提取,得到至少一张人脸特征图。
在进行图像分割时,新闻现场图像中的人脸图像可能会分割至同一张子图像中,同一人脸图像的不同区域或多个人脸图像也可能被分割至不同的子图像中,因此在进行人脸特征提取时,可得到一张或多种人脸特征图。
步骤S1033,通过自适应对比度增强算法对至少一张人脸特征图进行对比度频域范围增强处理。
自适应对比度增强(Adaptive Contrast Enhancement,ACE)算法源自retinex算法,其主要原理是通过差分计算目标点与周围像素点的相对阴暗关系来矫正最终像素值,具有很好的增强效果,其详细过程在此不再具体说明。
步骤S1034,对增强处理后的至少一张人脸特征图进行超分辨率重建。
超分辨率重建的过程为,先对人脸特征图进行通道数扩展,然后对扩展后的通道数对应的特征图进行卷积操作,最后再对像素进行重新排列,从而可得到高分辨率的特征图,以便用于后续的人脸识别,提高人脸识别精度。
步骤S1035,对超分辨率重建后的至少一张人脸特征图进行人脸识别。
本申请实施例中,在提取出所述新闻文本中的第一关键信息时,可通过自然语言处理(Natural Language Processing,NLP)算法提取出新闻文本中的第一关键信息。所述第一关键信息包括,但不限于时间、地点和新闻主题等。
对新闻现场图像中的人脸图像进行人脸识别时,新闻现场图像中的人脸图像可以是已知人员的人脸图像,也可以是未知人员的人脸图像。
服务器记录有所有工作人员的个人信息,个人信息可以包括姓名、工作单位和职务,还可以包括个人所在地区和个人简介等信息。
如果识别出新闻现场图像中包括已知的第一目标人员的人脸图像,则可根据第一关键信息和第一目标人员的个人信息生成第一目标人员的新闻记录。生成的新闻记录中至少包括有第一目标人员的姓名、工作单位和职务,以及参与事务的时间、地点和新闻主题等内容,还可以包括第一目标人员所在地区和个 人简介等。其中,第一目标人员可以是一个或多个,本申请实施例中不做具体限定。
例如,在一个实施例中,通过人脸识别识别出新闻现场图像中包含已知的第一目标人员包括第一目标人员A和第一目标人员B,通过提取出新闻文本中的关键信息,得到的时间为2021年1月1日,地点为CD市,新闻主题为X项目调研。而第一目标人员A的姓名为张三,职务为项目总工,第一目标人员B的名称为李四,职务为项目经理。则第一目标人员A的新闻记录可以包括如下:
姓名:张三;
职务:项目总工;
时间:2021年1月1日;
地点:CD市;
新闻主题:X项目调研。
第一目标人员B的新闻记录可以包括如下:
姓名:李四;
职务:项目经理;
时间:2021年1月1日;
地点:CD市;
新闻主题:X项目调研。
可以理解的,第一目标人员A和第一目标人员B的新闻记录中还可以包括个人所在地区和个人简介等其他信息。
在人脸识别过程中,还可能存在不能识别的人脸图像。如果识别出新闻现场图像中包括未知的第二目标人员的人脸图像,则可以提取出第二目标人员的人脸图像,用户可人为编辑第二目标人员的个人信息并上传,此时可接收到用户上传的第二目标人员的个人信息,然后根据第一关键信息和第二目标人员的个人信息,生成第二目标人员的新闻记录。其中,编辑的第二目标人员的个人 信息可以包括姓名、工作单位和职务,还可以包括个人所在地区和个人简介等信息。
步骤S104,提取出新闻文本中的第二关键信息,并根据第二关键信息生成新闻相关人员的新闻记录。
如果新闻图文中只包括新闻文本,不包括包含新闻相关人员人脸图像的新闻现场图像,则可提取出新闻文本中的第二关键信息,并根据第二关键信息生成新闻相关人员的新闻记录。其中,第二关键信息包括,但不限于时间、地点、新闻主题和姓名。根据第二关键信息生成新闻相关人员的新闻记录中包括参与事务人员的姓名、参与事务的时间、地点和新闻主题等,还可以包括参与事务人员的职务、所在地区和个人简介等。
在一个可能的设计中,在识别出已知的第一目标人员的人脸图像后,还可以将识别出的第一目标人员的人脸特征图储至人脸特征库中,或者通过识别出第一目标人员的人脸特征图替换掉人脸特征库中原始的人脸特征图,从而实现人脸特征库的迭代更新。
另外,在用户人为编辑第二目标人员的个人信息后,还可以将第二目标人员的个人信息与第二目标人员所对应的人脸图像关联并通过人脸特征库进行存储,从而实现人脸数据库的手动更新。
步骤S105,根据新闻相关人员的所有新闻记录建立与新闻相关人员对应的新闻图文类数据知识图谱。
其中,建立的新闻图文类数据知识图谱中包含有所述新闻相关人员的所有新闻记录。
在一个可能的设计中,在建立与所述新闻相关人员对应的新闻图文类数据知识图谱之后,还可将新闻图文类数据知识图谱所包含的新闻记录存储至与所述新闻相关人员对应的文件目录中,其中,新闻图文类数据知识图谱所包含的新闻记录在文件目录中按照时间先后顺序进行存储。如此可方便对企业工作人员的各项新闻记录进行查找。
如图2所示,是本申请实施例提供的另一新闻图文类数据知识图谱的建立方法的流程图,其可以包括如下步骤:
步骤S201,获取服务器记录的新闻图文。
步骤S202,判断新闻图文中除新闻文本外,是否还包括包含新闻相关人员人脸图像的新闻现场图像,如果否,则执行步骤S203;如果是,则执行步骤S204。
步骤S203,提取出新闻文本中的第二关键信息,并根据第二关键信息生成新闻相关人员的新闻记录。
步骤S204,对新闻现场图像中的人脸图像进行人脸识别,并提取出新闻文本中的第一关键信息。
步骤S205,判断人脸识别是否成功,如果是,则执行步骤S206,如果否,则执行步骤S207。
步骤S206,根据人脸识别结果和第一关键信息生成新闻相关人员的新闻记录。
步骤S207,通过关系抽取技术提取出未识别人员的个人信息,根据提取出的未识别人员的个人信息和第一关键信息生成新闻相关人员的新闻记录,并将提取出的未识别人员的个人信息存储至本地数据库。
本申请实施例中,新闻现场图像中可能在未能识别的人员,此时可以通过关系抽取技术提取出未识别人员的个人信息,并根据提取出的未识别人员的个人信息和第一关键信息生成新闻相关人员的新闻记录,同时还可以将提取出的未识别人员的个人信息和头像存储至本地数据库,以便用于后续的人脸识别。
综上所述,通过获取服务器记录的新闻图文,如果新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像时,则对新闻现场图像中的人脸图像进行人脸识别,提取出新闻文本中的第一关键信息,并根据人脸识别结果和第一关键信息生成新闻相关人员的新闻记录;如果新闻图文中仅包括新闻文本,则提取出新闻文本中的第二关键信息,并根据第二关键信息生成新闻相 关人员的新闻记录,然后根据新闻相关人员的所有新闻记录建立与新闻相关人员对应的新闻图文类数据知识图谱。如此能够结合人脸识别和文本提取,自动生成包含企业工作人员所有新闻记录的新闻图文类数据知识图谱,方便对企业工作人员参与各项事务进行统计。同时,在对新闻现场图像中的人脸图像进行人脸识别时,通过对新闻现场图像进行分割得到多张子图像,对每张子图像进行人脸特征提取得到至少一张人脸特征图,然后通过自适应对比度增强算法对至少一张人脸特征图进行对比度频域范围增强处理,并在增强处理进行超分辨率重建,如此可得到高分辨率的特征图,以便用于后续的人脸识别,提高人脸识别精度。其次,还可针对每个企业工作人员的新闻图文类数据知识图谱单独进行存储,方便对企业工作人员的各项新闻记录进行查找。另外,在识别出已知的第一目标人员的人脸图像,还可以将识别出的第一目标人员的人脸特征图储至人脸特征库中,或者通过识别出第一目标人员的人脸特征图替换掉人脸特征库中原始的人脸特征图,从而实现人脸特征库的迭代更新。最后,对于不能识别的人脸图像,还可以通过编辑实现人脸数据库的手动更新。
图3是本申请的一个实施例提供的电子设备的结构示意图。请参考图3,在硬件层面,该电子设备包括处理器,可选地还包括内部总线、网络接口、存储器。其中,存储器可能包含内存,例如高速随机存取存储器(Random-Access Memory,RAM),也可能还包括非易失性存储器(non-volatile memory),例如至少1个磁盘存储器等。当然,该电子设备还可能包括其他业务所需要的硬件。
处理器、网络接口和存储器可以通过内部总线相互连接,该内部总线可以是ISA(Industry Standard Architecture,工业标准体系结构)总线、PCI(Peripheral Component Interconnect,外设部件互连标准)总线或EISA(Extended Industry Standard Architecture,扩展工业标准结构)总线等。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示,图3中仅用一个双向箭头表示,但并不表示仅有一根总线或一种类型的总线。
存储器,用于存放程序。具体地,程序可以包括程序代码,所述程序代码包括计算机操作指令。存储器可以包括内存和非易失性存储器,并向处理器提供指令和数据。
处理器从非易失性存储器中读取对应的计算机程序到内存中然后运行,在逻辑层面上形成新闻图文类数据知识图谱的建立装置。处理器,执行存储器所存放的程序,并具体用于执行以下操作:
获取服务器记录的新闻图文;
如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
上述如本申请图3所示实施例揭示的新闻图文类数据知识图谱的建立装置执行的方法可以应用于处理器中,或者由处理器实现。处理器可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器,包括中央处理器(Central Processing Unit,CPU)、网络处理器(Network Processor,NP)等;还可以是数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现场可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本 申请一个或多个实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请一个或多个实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器,处理器读取存储器中的信息,结合其硬件完成上述方法的步骤。
该电子设备还可执行图1和图2的方法,并实现新闻图文类数据知识图谱的建立装置在图3所示实施例的功能,本申请实施例在此不再赘述。
当然,除了软件实现方式之外,本申请的电子设备并不排除其他实现方式,比如逻辑器件抑或软硬件结合的方式等等,也就是说以下处理流程的执行主体并不限定于各个逻辑单元,也可以是硬件或逻辑器件。
本申请实施例还提出了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的便携式电子设备执行时,能够使该便携式电子设备执行图1所示实施例的方法,并具体用于执行以下操作:
获取服务器记录的新闻图文;
如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息 包括时间、地点、新闻主题和姓名。
图4是本申请的一个实施例提供的新闻图文类数据知识图谱的建立装置的结构示意图。请参阅图4,在一种软件实施方式中,新闻图文类数据知识图谱的建立装置包括:
获取单元,用于获取服务器记录的新闻图文;
判断单元,用于判断所述新闻图文中是否包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像;
执行单元,用于当所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像时,对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
当所述新闻图文中仅包括新闻文本时,提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;以及
根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
总之,以上所述仅为本文件的较佳实施例而已,并非用于限定本文件的保护范围。凡在本文件的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本文件的保护范围之内。
上述实施例阐明的系统、装置、模块或单元,具体可以由计算机芯片或实体实现,或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的,计算机例如可以为个人计算机、膝上型计算机、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备的组合。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任 何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
本文件中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于系统实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。

Claims (10)

  1. 一种新闻图文类数据知识图谱的建立方法,其特征在于,包括:
    获取服务器记录的新闻图文;
    如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
    如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
    根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
    其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
  2. 根据权利要求1所述的方法,其特征在于,当人脸识别结果为所述新闻现场图像中包含已知的第一目标人员的人脸图像时,所述根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录,包括:
    根据所述第一关键信息和所述第一目标人员的个人信息生成所述第一目标人员的新闻记录;
    其中,所述个人信息包括姓名和职务。
  3. 根据权利要求1所述的方法,其特征在于,当所述人脸识别结果为所述新闻现场图像中包含未知的第二目标人员的人脸图像时,所述方法还包括:
    接收用户上传的所述第二目标人员的个人信息;
    根据所述第一关键信息和所述第二目标人员的个人信息生成所述第二目标人员的新闻记录;
    其中,所述个人信息包括姓名和职务。
  4. 根据权利要求3所述的方法,其特征在于,所述方法还包括:
    将所述第二目标人员的个人信息与所述第二目标人员所对应的人脸图像关联后进行存储。
  5. 根据权利要求1所述的方法,其特征在于,所述获取服务器记录的新闻图文,包括:
    当所述服务器接收到用户端最新上传的新闻图文时,从所述服务器获取所述最新上传的新闻图文;或
    每隔预设的时间间隔,从所述服务器获取用户端最新上传的新闻图文。
  6. 根据权利要求1所述的方法,其特征在于,在建立与所述新闻相关人员对应的新闻图文类数据知识图谱之后,所述方法还包括:
    将新闻图文类数据知识图谱所包含的新闻记录存储至与所述新闻相关人员对应的文件目录中;
    其中,新闻图文类数据知识图谱所包含的新闻记录在文件目录中按照时间先后顺序进行存储。
  7. 根据权利要求1所述的方法,其特征在于,所述对所述新闻现场图像中的人脸图像进行人脸识别,包括:
    对所述新闻现场图像进行分割,得到多张子图像;
    对所述多张子图像中的每张子图像进行人脸特征提取,得到至少一张人脸特征图;
    通过自适应对比度增强算法对所述至少一张人脸特征图进行对比度频域范围增强处理;
    对增强处理后的所述至少一张人脸特征图进行超分辨率重建;
    对超分辨率重建后的所述至少一张人脸特征图进行人脸识别。
  8. 一种新闻图文类数据知识图谱的建立装置,其特征在于,包括:
    获取单元,用于获取服务器记录的新闻图文;
    判断单元,用于判断所述新闻图文中是否包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像;
    执行单元,用于当所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像时,对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
    当所述新闻图文中仅包括新闻文本时,提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;以及
    根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
    其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
  9. 一种电子设备,其特征在于,包括处理器、通信接口、存储器和通信总线,其中,处理器,通信接口,存储器通过总线完成相互间的通信;
    存储器,用于存放计算机程序;
    处理器,用于执行存储器上所存放的程序,实现以下流程:
    获取服务器记录的新闻图文;
    如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
    如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
    根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
    其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
  10. 一种计算机可读存储介质,其特征在于,所述存储介质内存储有计算机程序,所述计算机程序被处理器执行时实现以下流程:
    获取服务器记录的新闻图文;
    如果所述新闻图文中包括新闻文本和包含新闻相关人员人脸图像的新闻现场图像,则对所述新闻现场图像中的人脸图像进行人脸识别,提取出所述新闻文本中的第一关键信息,并根据人脸识别结果和所述第一关键信息生成新闻相关人员的新闻记录;
    如果所述新闻图文中仅包括新闻文本,则提取出所述新闻文本中的第二关键信息,并根据所述第二关键信息生成新闻相关人员的新闻记录;
    根据新闻相关人员的所有新闻记录建立与所述新闻相关人员对应的新闻图文类数据知识图谱;
    其中,所述第一关键信息包括时间、地点和新闻主题,所述第二关键信息包括时间、地点、新闻主题和姓名。
PCT/CN2022/074424 2022-01-27 2022-01-27 新闻图文类数据知识图谱的建立方法、装置、设备及介质 WO2023141900A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/074424 WO2023141900A1 (zh) 2022-01-27 2022-01-27 新闻图文类数据知识图谱的建立方法、装置、设备及介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/074424 WO2023141900A1 (zh) 2022-01-27 2022-01-27 新闻图文类数据知识图谱的建立方法、装置、设备及介质

Publications (1)

Publication Number Publication Date
WO2023141900A1 true WO2023141900A1 (zh) 2023-08-03

Family

ID=87469930

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/074424 WO2023141900A1 (zh) 2022-01-27 2022-01-27 新闻图文类数据知识图谱的建立方法、装置、设备及介质

Country Status (1)

Country Link
WO (1) WO2023141900A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117453225A (zh) * 2023-10-26 2024-01-26 中国电子科技集团公司第十五研究所 新闻流生成方法、装置、设备和介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104091164A (zh) * 2014-07-28 2014-10-08 北京奇虎科技有限公司 人脸图片人名识别方法和系统
US20170024375A1 (en) * 2015-07-26 2017-01-26 Microsoft Technology Licensing, Llc Personal knowledge graph population from declarative user utterances
CN109754224A (zh) * 2018-12-29 2019-05-14 贵州小爱机器人科技有限公司 人事关系图谱构建方法、装置以及计算机存储介质
CN110245244A (zh) * 2019-06-20 2019-09-17 贵州电网有限责任公司 一种基于海量文本数据的人事关系知识图谱构建方法
CN110750656A (zh) * 2019-10-29 2020-02-04 上海德拓信息技术股份有限公司 一种基于知识图谱的多媒体检测方法
CN112364091A (zh) * 2020-11-09 2021-02-12 北京工商大学 一种基于知识图谱的人物关系可视化查询方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104091164A (zh) * 2014-07-28 2014-10-08 北京奇虎科技有限公司 人脸图片人名识别方法和系统
US20170024375A1 (en) * 2015-07-26 2017-01-26 Microsoft Technology Licensing, Llc Personal knowledge graph population from declarative user utterances
CN109754224A (zh) * 2018-12-29 2019-05-14 贵州小爱机器人科技有限公司 人事关系图谱构建方法、装置以及计算机存储介质
CN110245244A (zh) * 2019-06-20 2019-09-17 贵州电网有限责任公司 一种基于海量文本数据的人事关系知识图谱构建方法
CN110750656A (zh) * 2019-10-29 2020-02-04 上海德拓信息技术股份有限公司 一种基于知识图谱的多媒体检测方法
CN112364091A (zh) * 2020-11-09 2021-02-12 北京工商大学 一种基于知识图谱的人物关系可视化查询方法及系统

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117453225A (zh) * 2023-10-26 2024-01-26 中国电子科技集团公司第十五研究所 新闻流生成方法、装置、设备和介质

Similar Documents

Publication Publication Date Title
CN111062871B (zh) 一种图像处理方法、装置、计算机设备及可读存储介质
US8725490B2 (en) Virtual universal translator for a mobile device with a camera
TWI760671B (zh) 一種音視訊訊息處理方法及裝置、電子設備和電腦可讀儲存介質
US20140348394A1 (en) Photograph digitization through the use of video photography and computer vision technology
US20180025215A1 (en) Anonymous live image search
CN111275784A (zh) 生成图像的方法和装置
US20180336931A1 (en) Automatic and intelligent video sorting
US9665773B2 (en) Searching for events by attendants
CN113205047A (zh) 药名识别方法、装置、计算机设备和存储介质
WO2023141900A1 (zh) 新闻图文类数据知识图谱的建立方法、装置、设备及介质
EP3437015A1 (en) Image action based on automatic feature extraction
CN111292333B (zh) 用于分割图像的方法和装置
US10158598B2 (en) Social interaction data preservation for augmented photos
CN112102145B (zh) 图像处理方法及装置
CN111914068A (zh) 试题知识点的提取方法
WO2020244076A1 (zh) 人脸识别方法、装置、电子设备及存储介质
CN115333879B (zh) 一种远程会议方法及系统
CN114155841A (zh) 语音识别方法、装置、设备及存储介质
US20180189602A1 (en) Method of and system for determining and selecting media representing event diversity
CN113741864A (zh) 基于自然语言处理的语义化服务接口自动设计方法与系统
CN112711718A (zh) 一种评论信息的审核方法、装置、介质及电子设备
WO2021114824A1 (zh) 演示文稿生成方法、装置、设备及介质
Nguyen et al. Unmasking The Artist: Discriminating Human-Drawn And AI-Generated Human Face Art Through Facial Feature Analysis
CN112036350B (zh) 一种基于政务云的用户调查方法和系统
CN114154465B (zh) 结构图的结构重构方法、装置、电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22922746

Country of ref document: EP

Kind code of ref document: A1