Implementation method and system for rapid retrieval of AIoT video monitoring data

Abstract

The invention provides an AIoT video monitoring data rapid retrieval realization method and system, and the method comprises the steps: receiving a video tag inputted by a user through a video tag search bar, or receiving video tags of all levels selected by the user through a tag tree; retrieving monitoring video data meeting conditions according to the video tag input or clicked by the user; listing and displaying a video cover image of the video through a display area, wherein the video cover image displays all preview images of the monitoring video data when being triggered by a user; wherein the video cover image is a representative image of a certain frame of the video stream, and comprises obviously symbolic content or a business target. According to the method, the videos are labeled in advance, and the video covers are treated in advance, so that similar videos are effectively distinguished, and rapid retrieval of the video monitoring data is realized.

Description

Technical Field

[0001] The present invention relates to the field of AIoT video service technology, and in particular to a method and system for implementing rapid retrieval of AIoT video surveillance data. Background Art

[0002] With the continuous maturity of technologies such as the Internet, big data, cloud computing, and artificial intelligence, it provides strong technical support for the intelligent development of all walks of life. It is not only the inevitable result of technological development, but also the need for social and economic development. In many fields such as ecology, water conservancy, cities, and industry, people have an increasing demand for intelligence and automation. Through data analysis and intelligent processing, it provides users with more convenient and efficient services, can collect and process massive amounts of equipment data, and help units and individuals make more accurate and real-time decisions. The construction and application of the AIoT video comprehensive service platform involves multiple industries and fields, such as smart water conservancy, smart environmental protection, smart cities, smart agriculture, etc., which improves the efficiency and level of supervision, and also provides more opportunities for transformation and development for the industry, promoting the process of social intelligence.

[0003] Currently, a large number of video surveillance platforms lack a governance system, which inevitably leads to the problem of scattered and missing video data. The conventional practice is to add a lot of business information to the name of the device, rather than standardizing and structuring the video archive data to form effective video archives and video tags, which leads to the failure of video search in the specific application process to fit the actual application and low query efficiency.

[0004] When there are many video surveillance devices, even if a large amount of business information is added to the name of the video surveillance device, if the business information is similar, it will still be inconvenient to search. The main reason is that when the business information is similar, it is necessary to search through the video content. However, clicking on the video content to play is obviously time-consuming, so it is difficult to quickly retrieve the target video surveillance data. In addition, the video stream of a video surveillance device is long, and the content monitored in different monitoring periods is different. Some specific events may occur within the monitoring range, so the key frames of different monitoring periods will also be different. If they are not distinguished, it is difficult to find the information of the corresponding period. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for realizing rapid retrieval of AIoT video surveillance data, by pre-managing videos by labeling and video covers to effectively distinguish similar surveillance video data and surveillance data of different monitoring periods, thereby realizing rapid retrieval of surveillance video data.

[0006] In a first aspect, the present invention provides a method for implementing rapid retrieval of AIoT video surveillance data, comprising:

[0007] Receive video tags input by the user through the video tag search bar, or receive video tags of various levels selected by the user through the tag tree;

[0008] Retrieve surveillance video data that meets the conditions according to the video tags input or selected by the user;

[0009] The video cover images of the surveillance video data meeting the conditions are listed and displayed in the display area, and the video cover images display all preview images of the surveillance video data when triggered by the user; wherein the video cover image is a representative image of a certain frame of the video stream, and includes obvious iconic content or contains business objectives;

[0010] The process of setting the video cover image includes:

[0011] S1. Connect to the front-end video surveillance equipment, perform real-time analysis on the video surveillance stream, identify key frames of a set period or one or more representative frames of a specific event, and use the captured images as preview images and save them to the preview image path;

[0012] S2. When the user queries each video surveillance device, all preview images under the preview image path are retrieved for preview;

[0013] S3. When the user triggers any preview image, the system captures the video segment corresponding to the preview image for playback, and provides a playback progress bar for the user to operate and play the video segments before and after the video segment corresponding to the preview image.

[0014] In a second aspect, the present invention provides a system for implementing rapid retrieval of AIoT video surveillance data, comprising:

[0015] The tag receiving module is used to receive the video tags input by the user through the video tag search bar, or receive the video tags of each level selected by the user through the tag tree;

[0016] The tag search module is used to retrieve the surveillance video data that meets the conditions according to the video tags input or selected by the user and list them in the display area;

[0017] A video display image, which displays the video cover image of the surveillance video data that meets the conditions through the display area, and the video cover image displays all preview images of the surveillance video data when triggered by the user; wherein the video cover image is a representative image of a certain frame of the video stream, and includes obvious iconic content or contains business objectives;

[0018] The video cover setting module is used to implement the following process:

[0019] S1. Connect to the front-end video surveillance equipment, perform real-time analysis on the video surveillance stream, identify key frames of a set period or one or more representative frames of a specific event, and use the captured images as preview images and save them to the preview image path;

[0020] S2. When the user queries each video surveillance device, all preview images under the preview image path are retrieved for preview;

[0021] S3. When the user triggers any preview image, the system captures the video segment corresponding to the preview image for playback, and provides a playback progress bar for the user to operate and play the video segments before and after the video segment corresponding to the preview image.

[0022] The present invention has the following technical effects or advantages: by pre-managing videos by labeling and covering them, similar surveillance video data can be effectively distinguished. When searching video surveillance equipment, video labels and video covers can be combined for screening without playing video content. The surveillance time period can also be located by historical video covers, thereby realizing rapid retrieval of video surveillance data.

[0023] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention will be further described below in conjunction with embodiments with reference to the accompanying drawings.

[0025] Figure 1 is a flow chart of the method in Embodiment 1 of the present invention;

[0026] Figure 2 A schematic diagram of a video tag search bar and a tag tree according to an embodiment of the present invention;

[0027] Figure 3 A schematic diagram of a state of determining target surveillance video data through a video cover image and all preview images according to an embodiment of the present invention;

[0028] Figure 4 It is a schematic diagram of the structure of the system in the second embodiment of the present invention. DETAILED DESCRIPTION

[0029] The embodiments of the present application provide a method and system for rapid retrieval of AIoT video surveillance data, which pre-manages videos by labeling and video covers to effectively distinguish similar videos and accurately locate monitoring time periods, thereby achieving rapid retrieval of video surveillance data.

[0030] The technical solution in the embodiments of the present application has the following overall idea: by pre-managing the video labels and covers, similar surveillance video data can be effectively distinguished. When searching the video surveillance device, the video labels and video covers can be combined for screening without playing the video content. The surveillance time period can also be located through the historical video covers, thereby realizing rapid retrieval of video surveillance data.

[0031] Embodiment 1

[0032] like Figure 1 As shown, this embodiment provides a method for implementing rapid retrieval of AIoT video surveillance data, including:

[0033] The video tag input by the user is received through the video tag search bar, or the video tags at all levels selected by the user are received through the tag tree; Figure 2 The diagram shows the structure of the video tag search bar and the tag tree; the video tags include the field or industry to which the video belongs, the unit or department to which it belongs, the geographical location, etc.

[0034] Retrieve surveillance video data that meets the conditions according to the video tags input or selected by the user;

[0035] like Figure 3 As shown, the video cover images of the surveillance video data that meet the conditions are listed and displayed in the display area, and all preview images of the surveillance video data are displayed when the user triggers (for example, when the mouse is close to the video cover image); wherein the video cover image is a representative image of a certain frame of the video stream, and includes obvious iconic content or business objectives. The user can further confirm the target surveillance video data through the video cover image and all preview images, thereby realizing rapid retrieval.

[0036] The listing is in the form of picture previews.

[0037] The process of managing video data through video tags is as follows: users create a tag tree based on the actual supervision needs of the department. When creating tags, they support creation operations under multiple sub-tags. Users select the corresponding parent tag and then add the tag. The added tag will automatically become a sub-tag under the parent tag. The upper and lower tags are pre-set to establish a tag tree. For AIoT video surveillance data, commonly used tag trees include the field or industry, the unit or department, the geographical location, etc.

[0038] The process of setting the video cover image includes:

[0039] S1. Connect to the front-end video surveillance equipment, perform real-time analysis on the video surveillance stream, identify key frames of a set period or one or more representative frames of a specific event, and use the captured images as preview images and save them to the preview image path;

[0040] S2. When the user queries each video surveillance device, all preview images under the preview image path are retrieved for preview;

[0041] S3. When the user triggers any preview image, the system captures the video segment corresponding to the preview image for playback, and provides a playback progress bar for the user to operate and play the video segments before and after the video segment corresponding to the preview image.

[0042] Based on the same inventive concept, the present application also provides a device corresponding to the method in Example 1, see Example 2 for details.

[0043] Embodiment 2

[0044] like Figure 4 As shown, in this embodiment, a system for implementing rapid retrieval of AIoT video surveillance data is provided, including:

[0045] The tag receiving module is used to receive the video tags input by the user through the video tag search bar, or receive the video tags of each level selected by the user through the tag tree; Figure 2 The diagram shows the structure of the video tag search bar and the tag tree; the video tags include the field or industry to which the video belongs, the unit or department to which it belongs, the geographical location, etc.

[0046] The tag search module is used to retrieve the surveillance video data that meets the conditions according to the video tags input or selected by the user;

[0047] like Figure 3 As shown, the video display image lists and displays the video cover images of the surveillance video data that meet the conditions through the display area. The video cover image displays all preview images of the surveillance video data when triggered by the user; wherein the video cover image is a representative image of a certain frame of the video stream, and includes obvious iconic content or business objectives. The user can further confirm the target surveillance video data through the video cover image and all preview images, thereby realizing rapid retrieval.

[0048] The listing is in the form of picture previews.

[0049] The process of managing video data through video tags is as follows: users create a tag tree based on the actual supervision needs of the department. When creating tags, they support creation operations under multiple sub-tags. Users select the corresponding parent tag and then add the tag. The added tag will automatically become a sub-tag under the parent tag. The upper and lower tags are pre-set to establish a tag tree. For AIoT video surveillance data, commonly used tag trees include the field or industry, the unit or department, the geographical location, etc.

[0050] The implementation system of this embodiment may include a video cover setting module, and the video cover setting module is used to implement the following process:

[0051] S1. Connect to the front-end video surveillance equipment, perform real-time analysis on the video surveillance stream, identify key frames of a set period or one or more representative frames of a specific event, and use the captured images as preview images and save them to the preview image path;

[0052] S2. When the user queries each video surveillance device, all preview images under the preview image path are retrieved for preview;

[0053] S3. When the user triggers any preview image, the system captures the video segment corresponding to the preview image for playback, and provides a playback progress bar for the user to operate and play the video segments before and after the video segment corresponding to the preview image.

[0054] Since the device introduced in the second embodiment of the present invention is a device used to implement the method of the first embodiment of the present invention, those skilled in the art can understand the specific structure and deformation of the device based on the method introduced in the first embodiment of the present invention, so it is not described here in detail. All devices used in the method of the first embodiment of the present invention belong to the scope of protection of the present invention.

[0055] Although the specific implementation modes of the present invention are described above, those skilled in the art should understand that the specific implementation modes described are only illustrative and are not intended to limit the scope of the present invention. Equivalent modifications and changes made by those skilled in the art in accordance with the spirit of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A method for implementing rapid retrieval of AIoT video surveillance data, characterized in that: include: Receive video tags input by the user through the video tag search bar, or receive video tags of various levels selected by the user through the tag tree; Retrieve surveillance video data that meets the conditions according to the video tags input or selected by the user; The video cover images of the surveillance video data meeting the conditions are listed and displayed in the display area, and the video cover images display all preview images of the surveillance video data when triggered by the user; wherein the video cover image is a representative image of a certain frame of the video stream, and includes obvious iconic content or contains business objectives; The process of setting the video cover image includes: S1. Connect to the front-end video surveillance equipment, perform real-time analysis on the video surveillance stream, identify key frames of a set period or one or more representative frames of a specific event, and use the captured images as preview images and save them to the preview image path; S2. When the user queries each video surveillance device, all preview images under the preview image path are retrieved for preview; S3. When the user triggers any preview image, the system captures the video segment corresponding to the preview image for playback, and provides a playback progress bar for the user to operate and play the video segments before and after the video segment corresponding to the preview image.

2. According to claim 1, a method for realizing rapid retrieval of AIoT video surveillance data is characterized in that: The listing is in the form of picture previews.

3. A system for implementing rapid retrieval of AIoT video surveillance data, characterized in that: include: The tag receiving module is used to receive the video tags input by the user through the video tag search bar, or receive the video tags of each level selected by the user through the tag tree; The tag search module is used to retrieve the surveillance video data that meets the conditions according to the video tags input or selected by the user; A video display image, which lists and displays the video cover images of the surveillance video data that meets the conditions in the display area, and the video cover image displays all preview images of the surveillance video data when triggered by the user; wherein the video cover image is a representative image of a certain frame of the video stream, and includes obvious iconic content or contains business objectives; The video cover setting module is used to implement the following process: S1. Connect to the front-end video surveillance equipment, perform real-time analysis on the video surveillance stream, identify key frames of a set period or one or more representative frames of a specific event, and use the captured images as preview images and save them to the preview image path; S2. When the user queries each video surveillance device, all preview images under the preview image path are retrieved for preview; S3. When the user triggers any preview image, the system captures the video segment corresponding to the preview image for playback, and provides a playback progress bar for the user to operate and play the video segments before and after the video segment corresponding to the preview image.

4. According to claim 3, a system for realizing rapid retrieval of AIoT video surveillance data is characterized in that: The listing is in the form of picture previews.

Images