WO2023061019A1 - 图像处理方法、装置、设备和计算机可读存储介质 - Google Patents

图像处理方法、装置、设备和计算机可读存储介质 Download PDF

Info

Publication number
WO2023061019A1
WO2023061019A1 PCT/CN2022/111792 CN2022111792W WO2023061019A1 WO 2023061019 A1 WO2023061019 A1 WO 2023061019A1 CN 2022111792 W CN2022111792 W CN 2022111792W WO 2023061019 A1 WO2023061019 A1 WO 2023061019A1
Authority
WO
WIPO (PCT)
Prior art keywords
aging
target image
information
image information
present application
Prior art date
Application number
PCT/CN2022/111792
Other languages
English (en)
French (fr)
Inventor
刘杭雨
Original Assignee
中兴通讯股份有限公司
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 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2023061019A1 publication Critical patent/WO2023061019A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/76—Television signal recording
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/76—Television signal recording
    • H04N5/78—Television signal recording using magnetic recording
    • H04N5/781—Television signal recording using magnetic recording on disks or drums

Definitions

  • the embodiments of the present application relate to, but are not limited to, the technical field of image processing, and in particular, relate to an image processing method, an image processing apparatus, an image processing device, and a computer-readable storage medium.
  • the aging processing strategy for video files in the current video surveillance system is usually to set an aging period. After the aging period, the expired video files will be deleted. Wherein, the length of the aging period is determined by the size of the disk where the video files are stored.
  • Embodiments of the present application provide an image processing method, an image processing device, an image processing device, and a computer-readable storage medium.
  • the embodiment of the present application provides an image processing method, including: acquiring target image information, and inputting the target image information into a trained image recognition model; when the target is determined by the image recognition model The image information satisfies a preset identification rule, and the target image information is marked so that the target image information carries label information corresponding to the preset identification rule, wherein the label information corresponds to an aging strategy one-to-one; Execute the aging policy corresponding to the tag information on the target image information according to the tag information.
  • the embodiment of the present application also provides an image processing device, including: an acquisition unit, configured to acquire target image information, and input the target image information into a trained image recognition model; a marking unit, configured to When it is determined by the image recognition model that the target image information satisfies a preset recognition rule, marking the target image information so that the target image information carries tag information corresponding to the preset recognition rule , wherein the tag information corresponds to an aging strategy one by one; the aging unit is configured to execute the aging strategy corresponding to the tag information on the target image information according to the tag information.
  • the embodiment of the present application also provides an image processing device, including: a memory, a processor, and a computer program stored in the memory and operable on the processor, and the processor executes the The computer program realizes the image processing method as described in the first aspect above.
  • the embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions, and the computer-executable instructions are used to execute the image processing method as described in the above-mentioned first aspect.
  • Fig. 1 is a schematic diagram of a system architecture platform for executing an image processing method provided by an embodiment of the present application
  • FIG. 2 is a flowchart of an image processing method provided by an embodiment of the present application.
  • FIG. 3 is a flow chart of performing aging processing on target image information when the aging strategy includes the number of aging times in the image processing method provided by an embodiment of the present application;
  • FIG. 4 is a flow chart of performing the first aging process in the image processing method provided by an embodiment of the present application.
  • FIG. 5 is a flow chart of the second aging process after the first aging process does not set the aging time information in the image processing method provided by an embodiment of the present application;
  • FIG. 6 is a flow chart of querying image information through tags after the tags are marked in the image processing method provided by an embodiment of the present application;
  • Fig. 7 is an architecture diagram of an image recognition model provided by an embodiment of the present application.
  • FIG. 8 is an overall schematic diagram of an image processing method provided by an embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of an image processing device provided by an embodiment of the present application.
  • the aging processing strategy for video files in the current video surveillance system is usually to set an aging period. After the aging period, the expired video files will be deleted. Wherein, the length of the aging period is determined by the size of the disk where the video files are stored.
  • Most of the methods currently used to avoid video aging are to avoid video aging by saving video space, for example, saving video space by using a Gaussian mixture model, and avoiding aging by saving redundant information.
  • most of the methods for avoiding video aging are to constantly scan the video that is about to be aged, and then further process it. Most of these methods start from the way of saving storage space, rather than directly labeling them when recording, and this method usually uses a unified aging strategy to perform unified aging processing on all videos, which is less flexible.
  • an embodiment of the present application provides an image processing method, an image processing device, an image processing device, and a computer-readable storage medium.
  • the image processing method includes but is not limited to the following steps: first, acquire target image information, and The target image information is input into the trained image recognition model; then, when the image recognition model determines that the target image information satisfies the preset recognition rules, the target image information is marked so that the target image information carries the label corresponding to the preset recognition rules information, wherein the label information corresponds to the aging policy one by one; finally, according to the label information, the aging policy corresponding to the label information is executed on the target image information.
  • the embodiment of the present application can mark the target image information that meets the preset recognition rules, classify the marked image information, and implement different aging strategies for the image information carrying different label information , strong flexibility, can save key image information for a longer time, and improve product quality and user experience.
  • FIG. 1 is a schematic diagram of a system architecture platform 100 for executing an image processing method provided by an embodiment of the present application.
  • the system architecture platform 100 is provided with a processor 110 and a memory 120 , wherein the processor 110 and the memory 120 may be connected via a bus or in other ways.
  • connection via a bus is taken as an example.
  • the memory 120 can be used to store non-transitory software programs and non-transitory computer-executable programs.
  • the memory 120 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices.
  • the memory 120 may optionally include memory 120 located remotely relative to the processor 110, and these remote memories may be connected to the system architecture platform through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • system architecture platform can be applied to LTE communication network systems, 5G communication network systems, and subsequent evolved mobile communication network systems, etc., which is not specifically limited in this embodiment.
  • FIG. 1 does not constitute a limitation to the embodiment of the present application, and may include more or less components than those shown in the illustration, or combine some components, or have different Part placement.
  • the processor 110 can call the image processing program stored in the memory 120 to execute the image processing method.
  • FIG. 2 is a flowchart of an image processing method provided by an embodiment of the present application, and the method includes but is not limited to step S100 , step S200 and step S300 .
  • Step S100 acquiring target image information, and inputting the target image information into the trained image recognition model
  • Step S200 when it is determined by the image recognition model that the target image information satisfies the preset recognition rules, mark the target image information so that the target image information carries label information corresponding to the preset recognition rules, wherein the label information and the aging policy are one by one correspond;
  • Step S300 according to the tag information, execute an aging policy corresponding to the tag information on the target image information.
  • the target image information is obtained, and the target image information is input into the trained image recognition model; then, when the image recognition model determines that the target image information satisfies the preset recognition rules, the target The image information is marked so that the target image information carries label information corresponding to the preset identification rules, wherein the label information corresponds to the aging strategy one by one; finally, according to the label information, the aging strategy corresponding to the label information is executed on the target image information .
  • the embodiment of the present application can mark the target image information that meets the preset recognition rules, classify the marked image information, and implement different aging strategies for the image information carrying different label information , strong flexibility, can save key image information for a longer time, and improve product quality and user experience.
  • the image recognition model in the embodiment of the present application involves image recognition processing technology, wherein the image recognition processing technology refers to the use of computers to process, analyze and understand images to identify targets and objects in various patterns
  • image recognition technology is generally divided into face recognition and commodity recognition. Face recognition is mainly used in security inspection, identity verification and mobile payment.
  • the embodiment of the present application uses deep learning to train a matching model including the matching relationship between videos and rules based on the image recognition technology algorithm and combined with the preset intelligent rules. That is, in the process of video recording, when a certain frame meets the intelligent rules, the tag matched by the rules will be used to label the video and its corresponding position.
  • the above-mentioned trained image recognition model can be trained from sample image information and label information corresponding to preset recognition rules.
  • image information may include at least one of graphics and images.
  • FIG. 3 is a flow chart of performing aging processing on target image information in an image processing method provided by an embodiment of the present application when the aging strategy includes aging times.
  • the aging policy includes the number of aging times
  • the implementation of the aging policy corresponding to the label information on the target image information in the above step S300 includes but is not limited to step S400.
  • Step S400 perform aging processing on the target image information, wherein the number of aging processing times is the number of aging times in the aging strategy.
  • the number of times of performing aging processing on the target image information in this embodiment of the present application corresponds to the number of times of aging in the aging policy.
  • one preset recognition rule corresponds to one level of label information
  • different levels of label information correspond to different aging times.
  • the level of people falling in the nursing home is level 1
  • the level of people walking in the nursing home is level 2, wherein level 1 is the highest and level 2 is next.
  • the video when the video includes a scene where a person falls, the video will be marked as a level 1 marked video; similarly, when the video includes a scene where a person is walking, the video will be marked as a level 2 marked video; then , because the video recording of a person falling is of high importance, so in the embodiment of the present application, it is possible to choose not to perform video aging on the video recording of the first level mark, and to set the aging times for the video recording of the second level mark.
  • FIG. 4 is a flow chart of performing the first aging process in the image processing method provided by an embodiment of the present application.
  • performing aging processing on the target image information in step S400 above it includes but not limited to step S510 and step S520.
  • Step S510 in the case of performing the first aging process on the target image information, generate an aging time request instruction
  • Step S520 when aging time information corresponding to the aging time request instruction is received, perform aging processing on the target image information according to the aging time information.
  • the embodiment of the present application when the first aging process is performed on the target image information, the embodiment of the present application will generate an aging time request command in response, and then the user can input the aging time information according to the aging time request command, and then the embodiment of the present application will The information performs aging processing on the target image information, that is, the user is reminded that the aging time can be set during the first aging processing.
  • the user may be reminded to set the aging time by popping up an input box, or the user may be reminded to set the aging time by voice.
  • Figure 5 is a flowchart of the second aging process after the first aging process does not set the aging time information in the image processing method provided by an embodiment of the present application.
  • Step S400 it also includes but not limited to step S610 and step S620.
  • Step S610 in the case of performing the first aging process on the target image information, generate an aging time request instruction
  • Step S620 when the aging time information corresponding to the aging time request instruction is not received, stop the first aging process, and directly execute the aging process on the target image information during the second aging process.
  • the embodiment of the present application when the first aging process is performed on the target image information, the embodiment of the present application will generate an aging time request command in response, and then the user can input the aging time information according to the aging time request command. If the aging time information is input, the embodiment of the present application will directly perform aging processing on the target image information during the second aging processing.
  • the user may be reminded to set the aging time by popping up an input box, or the user may be reminded to set the aging time by voice.
  • FIG. 6 is a flow chart of querying image information through tags after tags are marked in an image processing method provided by an embodiment of the present application. After the above step S200, it also includes but not limited to step S710 and step S720.
  • Step S710 receiving a query instruction corresponding to the tag information
  • Step S720 calling and displaying the target image information corresponding to the label information according to the query instruction.
  • the embodiment of the present application can be based on the
  • the query instruction corresponding to a certain tag information filters out the target image information corresponding to the tag information from a plurality of stored image information.
  • the video surveillance system may be applied to a scene where key video recordings need to be viewed later.
  • key video recordings need to be viewed later.
  • a doctor diagnoses a patient in a nursing home, he may need to use key videos to help diagnose, and the doctor can view the video more conveniently through the set rules.
  • it may be necessary to use the key video marked as entering a certain room to catch the suspect and recover the stolen items.
  • FIG. 7 is an architecture diagram of an image recognition model provided by an embodiment of the present application; specifically, it may include an input layer 210, a convolutional layer 220, a pooling layer 230, a fully connected layer 240, and Output layer 250.
  • the embodiment of the present application mainly uses the image recognition algorithm. When the rule scene preset by the user is hit by the algorithm, the video is marked according to the marking level of the rule, and then aging processing of the corresponding level is performed on the video.
  • FIG. 8 is an overall schematic diagram of an image processing method provided by an embodiment of the present application.
  • the image processing method of the embodiment of the present application includes but is not limited to the following steps: first, the user presets corresponding rules and rule levels on the video surveillance platform; then, trains a large number of rule scenes through image recognition algorithms; When using the image recognition algorithm to hit the rule scene, mark the video hit by the rule; finally, implement different aging strategies for different videos.
  • FIG. 9 is a schematic structural diagram of an image processing device provided by an embodiment of the present application.
  • the image processing apparatus 300 of the embodiment of the present application includes, but is not limited to, an acquisition unit 310 , a marking unit 320 and an aging unit 330 .
  • the obtaining unit 310 is used to obtain target image information, and input the target image information into the trained image recognition model; the marking unit 320 is used to determine that the target image information meets the preset recognition rules through the image recognition model, Marking the target image information so that the target image information carries label information corresponding to the preset recognition rules, wherein the label information corresponds to the aging strategy one by one; The aging policy corresponding to the information.
  • the aging unit 330 is further configured to perform aging processing on the target image information, wherein the aging processing times are the aging times in the aging strategy.
  • the aging unit 330 is also used to generate an aging time request instruction when the first aging process is performed on the target image information; Information performs aging processing.
  • the aging unit 330 is also used to generate an aging time request instruction when the first aging processing is performed on the target image information; when the aging time information corresponding to the aging time request instruction is not received, stop the first aging processing , and perform aging processing directly on the target image information during the second aging processing.
  • the image processing device 300 of the embodiment of the present application includes but is not limited to a query unit 340, the query unit 340 is used to receive the query instruction corresponding to the label information, and call and display the target image corresponding to the label information according to the query instruction information.
  • an embodiment of the present application also provides an image processing device, which includes: a memory, a processor, and a computer program stored in the memory and operable on the processor.
  • the processor and memory can be connected by a bus or other means.
  • the image processing device in this embodiment can be applied to the system architecture platform in the embodiment shown in Figure 1, and the image processing device in this embodiment can constitute the system in the embodiment shown in Figure 1 As a part of the architecture platform, the two belong to the same inventive concept, so they have the same realization principle and beneficial effects, which will not be described in detail here.
  • the non-transitory software programs and instructions required to realize the image processing method of the above-mentioned embodiment are stored in the memory, and when executed by the processor, the image processing method of the above-mentioned embodiment is executed, for example, executing the above-described Fig. 2 to Fig. 6 method steps in .
  • an embodiment of the present application also provides a computer-readable storage medium, where the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to execute the above-mentioned image processing method. For example, the method steps in FIGS. 2 to 6 described above are performed.
  • the embodiment of the present application includes: firstly, acquiring target image information, and inputting the target image information into a trained image recognition model; then, when the image recognition model determines that the target image information satisfies preset recognition rules, marking the target image information so that the target image information carries label information corresponding to the preset identification rule, wherein the label information corresponds to an aging policy one by one; finally, according to the label information, Executing the aging policy corresponding to the label information on the target image information.
  • the embodiment of the present application can mark the target image information that meets the preset recognition rules, classify the marked image information, and implement different aging strategies for the image information carrying different label information , strong flexibility, can save key image information for a longer time, and improve product quality and user experience.
  • Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cartridges, tape, magnetic disk storage or other magnetic storage devices, or can Any other medium used to store desired information and which can be accessed by a computer.
  • communication media typically embody computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Signal Processing (AREA)
  • Image Analysis (AREA)

Abstract

本申请实施例提供了一种图像处理方法、装置、设备和计算机可读存储介质，包括：获取目标图像信息，并将目标图像信息输入至训练好的图像识别模型(S100)；当通过图像识别模型确定目标图像信息满足预设识别规则，对目标图像信息进行标记以使目标图像信息携带有与预设识别规则对应的标签信息，其中，标签信息和老化策略一一对应(S200)；根据标签信息，对目标图像信息执行与标签信息对应的老化策略(S300)。

Description

图像处理方法、装置、设备和计算机可读存储介质
相关申请的交叉引用
本申请基于申请号为202111188970.5、申请日为2021年10月12日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请实施例涉及但不限于图像处理技术领域，尤其涉及一种图像处理方法、图像处理装置、图像处理设备和计算机可读存储介质。
背景技术
目前，视频监控系统发展日渐成熟，录像是其中比较重要的模块。而当前视频监控系统中对录像文件的老化处理策略通常为设定一个老化周期，到了老化周期之后，就将过期的录像文件删除，其中，老化周期的长短由录像文件保存的磁盘大小确定。
目前采用的避免录像老化的方法大多是通过节约录像空间，以此避免录像老化，具体地，其大多数都是不断去扫描即将老化的录像，然后对其进行进一步的处理，该方式主要从节约存储空间的方向出发，但是该方式通常采用统一的老化策略对所有录像进行统一老化处理，灵活性较差。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本申请实施例提供了一种图像处理方法、图像处理装置、图像处理设备和计算机可读存储介质。
第一方面，本申请实施例提供了一种图像处理方法，包括：获取目标图像信息，并将所述目标图像信息输入至训练好的图像识别模型；当通过所述图像识别模型确定所述目标图像信息满足预设识别规则，对所述目标图像信息进行标记以使所述目标图像信息携带有与所述预设识别规则对应的标签信息，其中，所述标签信息和老化策略一一对应；根据所述标签信息，对所述目标图像信息执行与所述标签信息对应的所述老化策略。
第二方面，本申请实施例还提供了一种图像处理装置，包括：获取单元，用于获取目标图像信息，并将所述目标图像信息输入至训练好的图像识别模型；标记单元，用于在通过所述图像识别模型确定所述目标图像信息满足预设识别规则的情况下，对所述目标图像信息进行标记以使所述目标图像信息携带有与所述预设识别规则对应的标签信息，其中，所述标签信息和老化策略一一对应；老化单元，用于根据所述标签信息，对所述目标图像信息执行与所述标签信息对应的所述老化策略。
第三方面，本申请实施例还提供了一种图像处理设备，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如上述第一方面所述的图像处理方法。
第四方面，本申请实施例还提供了一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令用于执行如上述第一方面所述的图像处理方法。
本申请的其它特征和优点将在随后的说明书中阐述，并且，部分地从说明书中变得显而易见，或者通过实施本申请而了解。本申请的目的和其他优点可通过在说明书、权利要求书以及附图中所特别指出的结构来实现和获得。
附图说明
附图用来提供对本申请技术方案的进一步理解，并且构成说明书的一部分，与本申请的实施例一起用于解释本申请的技术方案，并不构成对本申请技术方案的限制。
图1是本申请一个实施例提供的用于执行图像处理方法的系统架构平台的示意图；
图2是本申请一个实施例提供的图像处理方法的流程图；
图3是本申请一个实施例提供的图像处理方法中当老化策略包括老化次数时对目标图像信息执行老化处理的流程图；
图4是本申请一个实施例提供的图像处理方法中执行第一次老化处理的流程图；
图5是本申请一个实施例提供的图像处理方法中当第一次老化处理没有设置老化时间信息之后第二次老化处理的流程图；
图6是本申请一个实施例提供的图像处理方法中在标记完标签之后通过标签对图像信息进行查询的流程图；
图7是本申请一个实施例提供的图像识别模型的架构图；
图8是本申请一个实施例提供的图像处理方法的整体示意图；
图9是本申请一个实施例提供的图像处理装置的结构示意图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅用以解释本申请，并不用于限定本申请。
需要说明的是，虽然在装置示意图中进行了功能模块划分，在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于装置中的模块划分，或流程图中的顺序执行所示出或描述的步骤。说明书、权利要求书或上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。
在相关技术中，视频监控系统发展日渐成熟，录像是其中比较重要的模块。而当前视频监控系统中对录像文件的老化处理策略通常为设定一个老化周期，到了老化周期之后，就将过期的录像文件删除，其中，老化周期的长短由录像文件保存的磁盘大小确定。
目前采用的避免录像老化的方法大多是通过节约录像空间，以此避免录像老化，示例性地，通过高斯混合模型节约录像空间，通过节约冗余信息的方式避免老化。具体地，其绝大多数的避免录像老化的方法，都是不断去扫描即将老化的录像，然后对其进行进一步的处理。这种方式大多从节约存储空间的方式出发，而不是在录像时直接对其打上标签，并且这种方式通常采用统一的老化策略对所有录像进行统一老化处理，灵活性较差。
基于上述情况，本申请实施例提供了一种图像处理方法、图像处理装置、图像处理设备 和计算机可读存储介质，该图像处理方法包括但不限于如下步骤：首先，获取目标图像信息，并将目标图像信息输入至训练好的图像识别模型；接着，当通过图像识别模型确定目标图像信息满足预设识别规则，对目标图像信息进行标记以使目标图像信息携带有与预设识别规则对应的标签信息，其中，标签信息和老化策略一一对应；最后，根据标签信息，对目标图像信息执行与标签信息对应的老化策略。根据本申请实施例的技术方案，本申请实施例能够对满足预设识别规则的目标图像信息进行标记，并对标记的图像信息进行分类，针对携带有不同标签信息的图像信息执行不同的老化策略，灵活性强，能够使得关键图像信息的保存时间更久，便于提升产品质量和用户体验。
下面结合附图，对本申请实施例作进一步阐述。
如图1所示，图1是本申请一个实施例提供的用于执行图像处理方法的系统架构平台100的示意图。
在图1的示例中，该系统架构平台100设置有处理器110和存储器120，其中，处理器110和存储器120可以通过总线或者其他方式连接，图1中以通过总线连接为例。
存储器120作为一种非暂态计算机可读存储介质，可用于存储非暂态软件程序以及非暂态性计算机可执行程序。此外，存储器120可以包括高速随机存取存储器，还可以包括非暂态存储器，例如至少一个磁盘存储器件、闪存器件、或其他非暂态固态存储器件。在一些实施方式中，存储器120可选包括相对于处理器110远程设置的存储器120，这些远程存储器可以通过网络连接至该系统架构平台。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
本领域技术人员可以理解的是，该系统架构平台可以应用于LTE通信网络系统、5G通信网络系统以及后续演进的移动通信网络系统等，本实施例对此并不作具体限定。
本领域技术人员可以理解的是，图1中示出的系统架构平台并不构成对本申请实施例的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
在图1所示的系统架构平台中，处理器110可以调用储存在存储器120中的图像处理程序，从而执行图像处理方法。
基于上述系统架构平台，下面提出本申请的图像处理方法的各个实施例。
如图2所示，图2是本申请一个实施例提供的图像处理方法的流程图，该方法包括但不限于有步骤S100、步骤S200和步骤S300。
步骤S100、获取目标图像信息，并将目标图像信息输入至训练好的图像识别模型；
步骤S200、当通过图像识别模型确定目标图像信息满足预设识别规则，对目标图像信息进行标记以使目标图像信息携带有与预设识别规则对应的标签信息，其中，标签信息和老化策略一一对应；
步骤S300、根据标签信息，对目标图像信息执行与标签信息对应的老化策略。
具体地，在本申请实施例中，首先，获取目标图像信息，并将目标图像信息输入至训练好的图像识别模型；接着，当通过图像识别模型确定目标图像信息满足预设识别规则，对目标图像信息进行标记以使目标图像信息携带有与预设识别规则对应的标签信息，其中，标签信息和老化策略一一对应；最后，根据标签信息，对目标图像信息执行与标签信息对应的老化策略。根据本申请实施例的技术方案，本申请实施例能够对满足预设识别规则的目标图像信息进行标记，并对标记的图像信息进行分类，针对携带有不同标签信息的图像信息执行不 同的老化策略，灵活性强，能够使得关键图像信息的保存时间更久，便于提升产品质量和用户体验。
需要说明的是，本申请实施例中的图像识别模型涉及图像识别处理技术，其中，图像识别处理技术是指利用计算机对图像进行处理、分析和理解，以识别各种不同模式的目标和对象的技术，是应用深度学习算法的一种实践应用。现阶段图像识别技术一般分为人脸识别与商品识别，人脸识别主要运用在安全检查、身份核验与移动支付中。
因此，本申请实施例在图像识别技术算法的基础上，结合预设的智能规则，利用深度学习训练出包括有录像和规则之间匹配关系的匹配模型。即在录像的过程中，当某一帧画面满足智能规则时，通过规则匹配的标记为该段录像以及其对应位置打上标签。
另外，可以理解的是，关于上述训练好的图像识别模型，可以由与预设识别规则对应的样本图像信息和标签信息训练得到。
另外，可以理解的是，关于上述的图像信息，可以包括图形和影像中的至少一种。
另外，如图3所示，图3是本申请一个实施例提供的图像处理方法中当老化策略包括老化次数时对目标图像信息执行老化处理的流程图。当老化策略包括老化次数，关于上述步骤S300中的对目标图像信息执行与标签信息对应的老化策略，包括但不限于有步骤S400。
步骤S400、对目标图像信息执行老化处理，其中，老化处理的次数为老化策略中的老化次数。
具体地，当老化策略包括老化次数，那么，本申请实施例对目标图像信息执行老化处理的次数就对应为老化策略中的老化次数。
需要说明的是，图像识别模型内设置有多个预设识别规则，一个预设识别规则对应一个等级的标签信息，不同等级的标签信息对应不同的老化次数。
示例性地，在养老院时人员跌倒的等级为1级，在养老院时人员行走的等级为2级，其中，1级为最高，2级次之。那么，当录像中包括有人员跌倒的场景，那么会将该录像标记为1级标记录像；同理，当录像中包括有人员行走的场景，那么会将该录像标记为2级标记录像；接着，由于人员跌倒的录像的重要程度较高，所以，本申请实施例可以选择对1级标记录像不进行录像老化，选择对2级标记录像设置老化次数。
另外，如图4所示，图4是本申请一个实施例提供的图像处理方法中执行第一次老化处理的流程图。关于上述步骤S400中的对目标图像信息执行老化处理，包括但不限于有步骤S510和步骤S520。
步骤S510、在对目标图像信息执行第一次老化处理的情况下，生成老化时间请求指令；
步骤S520、当接收到与老化时间请求指令对应的老化时间信息，根据老化时间信息对目标图像信息执行老化处理。
具体地，当对目标图像信息执行第一次老化处理时，本申请实施例会响应生成老化时间请求指令，接着用户可以根据老化时间请求指令输入老化时间信息，然后本申请实施例就会根据老化时间信息对目标图像信息执行老化处理，即在首次老化处理时提醒用户可以设置老化时间。
可以理解的是，关于上述的老化时间请求指令的表现形式，可以通过弹出输入框来提醒用户设置老化时间，也可以通过语音方式来提醒用户设置老化时间。
另外，如图5所示，图5是本申请一个实施例提供的图像处理方法中当第一次老化处理 没有设置老化时间信息之后第二次老化处理的流程图。关于上述步骤S400中的对目标图像信息执行老化处理，还包括但不限于有步骤S610和步骤S620。
步骤S610、在对目标图像信息执行第一次老化处理的情况下，生成老化时间请求指令；
步骤S620、当没有接收到与老化时间请求指令对应的老化时间信息，停止第一次老化处理，并在第二次老化处理时直接对目标图像信息执行老化处理。
具体地，当对目标图像信息执行第一次老化处理时，本申请实施例会响应生成老化时间请求指令，接着用户可以根据老化时间请求指令输入老化时间信息，若用户在第一次老化处理时没有输入老化时间信息，那么本申请实施例就会在第二次老化处理时直接对目标图像信息执行老化处理。
可以理解的是，关于上述的老化时间请求指令的表现形式，可以通过弹出输入框来提醒用户设置老化时间，也可以通过语音方式来提醒用户设置老化时间。
另外，如图6所示，图6是本申请一个实施例提供的图像处理方法中在标记完标签之后通过标签对图像信息进行查询的流程图。在上述步骤S200之后，还包括但不限于有步骤S710和步骤S720。
步骤S710、接收与标签信息对应的查询指令；
步骤S720、根据查询指令，调用并显示与标签信息对应的目标图像信息。
具体地，在对目标图像信息进行标记以使目标图像信息携带有与预设识别规则对应的标签信息之后，当后期需要调用查看某个标签信息对应的图像信息时，本申请实施例可以根据与某个标签信息对应的查询指令，从多个存储的图像信息中筛选出与标签信息对应的目标图像信息。
示例性地，当本申请实施例应用于视频监控系统，那么该视频监控系统可以应用于后期需要查看关键录像的场景。例如：医生给养老院的病人诊断时，可能需要借助关键录像帮助诊断，而且通过设置的规则医生能够更方便的查看录像。或者，银行发生盗窃视频时，可能需要借助标记为进入某个房间的关键录像，便于抓捕嫌疑人，追回盗窃物品。
另外，如图7所示，图7是本申请一个实施例提供的图像识别模型的架构图；具体地，可以包括有输入层210、卷积层220、池化层230、全连接层240和输出层250。本申请实施例主要通过图像识别算法，当用户预设的规则情景被算法命中时，通过该规则的标记等级针对该录像进行相应的标记，然后对这个的录像执行相应等级的老化处理。
基于上述的图像处理方法，下面提出本申请的图像处理方法的整体实施例。
如图8所示，图8是本申请一个实施例提供的图像处理方法的整体示意图。本申请实施例的图像处理方法包括但不限于有如下步骤：首先，用户在视频监控平台预设相应的规则以及规则等级；然后，通过图像识别算法训练大量的规则场景；接着，当在录像进行时利用图像识别算法进行命中规则场景，针对被规则命中的录像进行标记；最后，再针对不同的录像执行不同的老化策略。
基于上述的图像处理方法，下面提出本申请的图像处理装置的各个实施例。
如图9所示，图9是本申请一个实施例提供的图像处理装置的结构示意图。本申请实施例的图像处理装置300包括但不限于有获取单元310、标记单元320和老化单元330。
具体地，获取单元310用于获取目标图像信息，并将目标图像信息输入至训练好的图像识别模型；标记单元320用于在通过图像识别模型确定目标图像信息满足预设识别规则的情 况下，对目标图像信息进行标记以使目标图像信息携带有与预设识别规则对应的标签信息，其中，标签信息和老化策略一一对应；老化单元330用于根据标签信息，对目标图像信息执行与标签信息对应的老化策略。
另外，当老化策略包括老化次数，老化单元330还用于对目标图像信息执行老化处理，其中，老化处理的次数为老化策略中的老化次数。
另外，老化单元330还用于在对目标图像信息执行第一次老化处理的情况下，生成老化时间请求指令；当接收到与老化时间请求指令对应的老化时间信息，根据老化时间信息对目标图像信息执行老化处理。
另外，老化单元330还用于在对目标图像信息执行第一次老化处理的情况下，生成老化时间请求指令；当没有接收到与老化时间请求指令对应的老化时间信息，停止第一次老化处理，并在第二次老化处理时直接对目标图像信息执行老化处理。
另外，本申请实施例的图像处理装置300包括但不限于有查询单元340，该查询单元340用于接收与标签信息对应的查询指令，并根据查询指令，调用并显示与标签信息对应的目标图像信息。
基于上述的图像处理方法，下面分别提出本申请的图像处理设备和计算机可读存储介质的各个实施例。
另外，本申请的一个实施例还提供了一种图像处理设备，该图像处理设备包括：存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序。
处理器和存储器可以通过总线或者其他方式连接。
需要说明的是，本实施例中的图像处理设备，可以应用于如图1所示实施例中的系统架构平台，本实施例中的图像处理设备，能够构成图1所示实施例中的系统架构平台的一部分，两者属于相同的发明构思，因此两者具有相同的实现原理以及有益效果，此处不再详述。
实现上述实施例的图像处理方法所需的非暂态软件程序以及指令存储在存储器中，当被处理器执行时，执行上述实施例的图像处理方法，例如，执行以上描述的图2至图6中的方法步骤。
此外，本申请的一个实施例还提供了一种计算机可读存储介质，该计算机可读存储介质存储有计算机可执行指令，计算机可执行指令用于执行上述的图像处理方法。例如，执行以上描述的图2至图6中的方法步骤。
本申请实施例包括：首先，获取目标图像信息，并将所述目标图像信息输入至训练好的图像识别模型；接着，当通过所述图像识别模型确定所述目标图像信息满足预设识别规则，对所述目标图像信息进行标记以使所述目标图像信息携带有与所述预设识别规则对应的标签信息，其中，所述标签信息和老化策略一一对应；最后，根据所述标签信息，对所述目标图像信息执行与所述标签信息对应的所述老化策略。根据本申请实施例的技术方案，本申请实施例能够对满足预设识别规则的目标图像信息进行标记，并对标记的图像信息进行分类，针对携带有不同标签信息的图像信息执行不同的老化策略，灵活性强，能够使得关键图像信息的保存时间更久，便于提升产品质量和用户体验。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统可以被实施为软件、固件、硬件及其适当的组合。某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实 施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域普通技术人员公知的是，通信介质通常包括计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。
以上是对本申请的一些实施进行了具体说明，但本申请并不局限于上述实施方式，熟悉本领域的技术人员在不违背本申请精神的共享条件下还可作出种种等同的变形或替换，这些等同的变形或替换均包括在本申请权利要求所限定的范围内。

Claims (10)

  1. 一种图像处理方法，包括：
    获取目标图像信息，并将所述目标图像信息输入至训练好的图像识别模型；
    当通过所述图像识别模型确定所述目标图像信息满足预设识别规则，对所述目标图像信息进行标记以使所述目标图像信息携带有与所述预设识别规则对应的标签信息，其中，所述标签信息和老化策略一一对应；
    根据所述标签信息，对所述目标图像信息执行与所述标签信息对应的所述老化策略。
  2. 根据权利要求1所述的方法，其中，所述老化策略包括老化次数，所述对所述目标图像信息执行与所述标签信息对应的所述老化策略，包括：
    对所述目标图像信息执行老化处理，其中，所述老化处理的次数为所述老化策略中的所述老化次数。
  3. 根据权利要求2所述的方法，其中，所述对所述目标图像信息执行老化处理，包括：
    在对所述目标图像信息执行第一次老化处理的情况下，生成老化时间请求指令；
    当接收到与所述老化时间请求指令对应的老化时间信息，根据所述老化时间信息对所述目标图像信息执行老化处理。
  4. 根据权利要求3所述的方法，其中，所述对所述目标图像信息执行老化处理，还包括：
    当没有接收到与所述老化时间请求指令对应的老化时间信息，停止第一次老化处理，并在第二次老化处理时直接对所述目标图像信息执行老化处理。
  5. 根据权利要求2所述的方法，其中，所述图像识别模型内设置有多个所述预设识别规则，一个所述预设识别规则对应一个等级的所述标签信息，不同等级的所述标签信息对应不同的所述老化次数。
  6. 根据权利要求1所述的方法，其中，在所述对所述目标图像信息进行标记以使所述目标图像信息携带有与所述预设识别规则对应的标签信息之后，所述方法还包括：
    接收与所述标签信息对应的查询指令；
    根据所述查询指令，调用并显示与所述标签信息对应的所述目标图像信息。
  7. 根据权利要求1至6中任意一项所述的方法，其中，所述图像识别模型由与所述预设识别规则对应的样本图像信息和标签信息训练得到。
  8. 一种图像处理装置，其中，包括：
    获取单元，用于获取目标图像信息，并将所述目标图像信息输入至训练好的图像识别模型；
    标记单元，用于在通过所述图像识别模型确定所述目标图像信息满足预设识别规则的情况下，对所述目标图像信息进行标记以使所述目标图像信息携带有与所述预设识别规则对应的标签信息，其中，所述标签信息和老化策略一一对应；
    老化单元，用于根据所述标签信息，对所述目标图像信息执行与所述标签信息对应的所述老化策略。
  9. 一种图像处理设备，其中，包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如权利要求1至7中任意一项所述的图像处理方法。
  10. 一种计算机可读存储介质，其中，存储有计算机可执行指令，所述计算机可执行指令用于执行如权利要求1至7中任意一项所述的图像处理方法。
PCT/CN2022/111792 2021-10-12 2022-08-11 图像处理方法、装置、设备和计算机可读存储介质 WO2023061019A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111188970.5 2021-10-12
CN202111188970.5A CN115965903A (zh) 2021-10-12 2021-10-12 图像处理方法、装置、设备和计算机可读存储介质

Publications (1)

Publication Number Publication Date
WO2023061019A1 true WO2023061019A1 (zh) 2023-04-20

Family

ID=85889945

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/111792 WO2023061019A1 (zh) 2021-10-12 2022-08-11 图像处理方法、装置、设备和计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN115965903A (zh)
WO (1) WO2023061019A1 (zh)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006059185A (ja) * 2004-08-20 2006-03-02 Rise Corp データ管理システム及び方法
JP2006323896A (ja) * 2005-05-17 2006-11-30 Toshiba Corp 記録済み情報の自動削除機能を備えた記録再生装置、及びこの自動削除機能を実現する削除方法
CN102945678A (zh) * 2012-11-09 2013-02-27 中兴通讯股份有限公司 录像文件的存储方法、系统和录像监测装置
CN103458209A (zh) * 2012-06-04 2013-12-18 中兴通讯股份有限公司 监控录像老化处理方法及装置
CN104639860A (zh) * 2014-12-31 2015-05-20 安科智慧城市技术（中国）有限公司 一种监控录像的存储方法和装置
CN111857551A (zh) * 2019-04-29 2020-10-30 杭州海康威视数字技术股份有限公司 一种录像数据老化方法及装置
US20210304575A1 (en) * 2018-08-10 2021-09-30 Sharp Kabushiki Kaisha Monitoring information recording apparatus, monitoring information recording system, control method for monitoring information recording apparatus, and recording medium

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006059185A (ja) * 2004-08-20 2006-03-02 Rise Corp データ管理システム及び方法
JP2006323896A (ja) * 2005-05-17 2006-11-30 Toshiba Corp 記録済み情報の自動削除機能を備えた記録再生装置、及びこの自動削除機能を実現する削除方法
CN103458209A (zh) * 2012-06-04 2013-12-18 中兴通讯股份有限公司 监控录像老化处理方法及装置
CN102945678A (zh) * 2012-11-09 2013-02-27 中兴通讯股份有限公司 录像文件的存储方法、系统和录像监测装置
CN104639860A (zh) * 2014-12-31 2015-05-20 安科智慧城市技术（中国）有限公司 一种监控录像的存储方法和装置
US20210304575A1 (en) * 2018-08-10 2021-09-30 Sharp Kabushiki Kaisha Monitoring information recording apparatus, monitoring information recording system, control method for monitoring information recording apparatus, and recording medium
CN111857551A (zh) * 2019-04-29 2020-10-30 杭州海康威视数字技术股份有限公司 一种录像数据老化方法及装置

Also Published As

Publication number Publication date
CN115965903A (zh) 2023-04-14

Similar Documents

Publication Publication Date Title
CN110276366A (zh) 使用弱监督模型来检测对象
US20190340746A1 (en) Stationary object detecting method, apparatus and electronic device
US20210382933A1 (en) Method and device for archive application, and storage medium
WO2021232594A1 (zh) 语音情绪识别方法、装置、电子设备及存储介质
WO2022048209A1 (zh) 车牌识别方法、装置、电子设备及存储介质
WO2018035667A1 (zh) 显示方法、装置、电子设备、计算机程序产品和非暂态计算机可读存储介质
US20120114198A1 (en) Facial image gender identification system and method thereof
WO2020038138A1 (zh) 样本标注方法及装置、损伤类别的识别方法及装置
WO2021057148A1 (zh) 基于神经网络的脑组织分层方法、装置、计算机设备
CN112287994A (zh) 伪标签处理方法、装置、设备及计算机可读存储介质
CN110619252B (zh) 识别图片中表单数据的方法、装置、设备及存储介质
KR20200068043A (ko) 영상 기계학습을 위한 객체 gt 정보 생성 방법 및 시스템
CN110619656B (zh) 基于双目摄像头的人脸检测跟踪方法、装置及电子设备
TW201448585A (zh) 利用行動電話及雲端可視化搜尋引擎之即時物體掃描
CN110309337B (zh) 一种针对多种目标识别算法的特征值集中存储方法及装置
WO2022213540A1 (zh) 目标检测、属性识别与跟踪方法及系统
Taverriti et al. Real-time wearable computer vision system for improved museum experience
CN108734003A (zh) 身份验证方法、装置、设备、存储介质及程序
CN106846625A (zh) 身份信息识别处理装置
WO2020001616A1 (zh) 本体构建方法及装置
WO2020103462A1 (zh) 视觉搜索方法、装置、计算机设备及存储介质
WO2020001614A1 (zh) 本体构建方法及装置
WO2023061019A1 (zh) 图像处理方法、装置、设备和计算机可读存储介质
US11705239B2 (en) Method and apparatus for generating image reports
US11706381B2 (en) Selective obfuscation of objects in media content

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: 22879966

Country of ref document: EP

Kind code of ref document: A1