CN113507642A - Video segmentation method, system, storage medium and device - Google Patents
Video segmentation method, system, storage medium and device Download PDFInfo
- Publication number
- CN113507642A CN113507642A CN202111058621.1A CN202111058621A CN113507642A CN 113507642 A CN113507642 A CN 113507642A CN 202111058621 A CN202111058621 A CN 202111058621A CN 113507642 A CN113507642 A CN 113507642A
- Authority
- CN
- China
- Prior art keywords
- image
- node
- video
- business process
- images
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
Abstract
The invention provides a video segmentation method, a video segmentation system, a storage medium and a video segmentation device, wherein the method comprises the following steps: acquiring and segmenting video data to obtain a frame image; and comparing the frame images to obtain a service node image, obtaining video starting and ending nodes of a target video record corresponding to the service flow according to the image and service flow corresponding rules, and segmenting the video data according to the video starting nodes and the video ending nodes to obtain the target video record. According to the video segmentation method, the video segmentation system, the storage medium and the video segmentation equipment, the video data are segmented into the frame images to obtain the plurality of frame images, so that the video images are segmented into the plurality of static behavior images and the images are compared and analyzed with the business process corresponding rule, when the related business process video needs to be obtained, only the target video corresponding to the business process needs to be directly called, the video does not need to be played back from the beginning, and the technical problem that the video cannot be quickly tracked and played back in the prior art is solved.
Description
Technical Field
The present invention relates to the field of video segmentation technologies, and in particular, to a video segmentation method, a video segmentation system, a storage medium, and a video segmentation device.
Background
In the case handling business of the criminal conviction and the penalty of the inspection institution, cases which are voluntarily convinced and the penalty need to be summoned to an inspection officer, an on-duty lawyer and a criminal suspect for carrying out the negotiation and communication of the debate, and in order to guarantee the right of the criminal suspect, the whole-process recording is needed in the period. However, the process of acquainting and penalizing includes a plurality of nodes, and in the prior art, video monitoring cannot distinguish which video stream corresponds to which node, so that subsequent video recording cannot be quickly tracked and played back.
Disclosure of Invention
Based on this, the present invention provides a video segmentation method, a video segmentation system, a storage medium, and a video segmentation device, which are used to solve the technical problem that a video recording cannot be tracked and played back quickly in the prior art.
One aspect of the present invention provides a video segmentation method, including:
acquiring video data, and carrying out frame picture segmentation on the video data to obtain a plurality of frame images;
comparing and analyzing the frame image through the pre-trained image and the business process corresponding rule to obtain a plurality of business node images;
according to the image and business process corresponding rule, the business node image and the business process are associated and corresponding, and the starting or ending time node of the business node image which is associated and corresponding to the business process is the starting or ending time node of the business node in the business process;
and according to the starting time node of the service node in the service flow and the ending time node of the service node in the service flow, acquiring the video starting node and the video ending node of the target video corresponding to the service flow, and segmenting the video data according to the video starting node and the video ending node to obtain the target video corresponding to the service flow.
The video segmentation method obtains a plurality of frame images by segmenting the frame images of the video data, thereby segmenting the video images into a plurality of static behavior images and comparing and analyzing the images with the corresponding rules of the business process, improving the accuracy of video analysis, avoiding the condition that the analysis result is inaccurate because the dynamic video is directly analyzed, further, obtaining the start and end time nodes of the business node in the business process video by correlating and corresponding the business node images with the business process, so as to obtain the start and end nodes of the video of the target video corresponding to the business process, thereby segmenting the video data according to the start node and the end node of the video to obtain the target video, when the related business process video needs to be obtained, only needing to directly call the target video corresponding to the business process, the video is not required to be played back from the beginning, so that the video record is quickly tracked and played back, and the technical problem that the video record cannot be quickly tracked and played back in the prior art is solved.
In addition, the video segmentation method according to the present invention may further have the following additional technical features:
further, the step of comparing and analyzing the frame image according to the pre-trained image and the business process corresponding rule to obtain a plurality of business node images includes:
comparing and analyzing the frame image through a pre-trained image and business process corresponding rule to screen out a plurality of target node images, wherein the image and business process corresponding rule comprises a plurality of standard images, and each standard image corresponds to a business process node;
and carrying out image comparison on the target node image and the standard image through an image comparison technology to obtain a plurality of service node images.
Further, the step of comparing the target node image with the standard image by an image comparison technique to obtain a plurality of service node images includes:
performing full-image fuzzy comparison on the target node image and the standard image, and judging whether the result of the full-image fuzzy comparison of the target node image is greater than 70%;
if the result after the full-image fuzzy comparison is more than 70%, performing local feature point comparison on target node images of which the result after the full-image fuzzy comparison is more than 70%, and judging whether the result after the local feature point comparison is more than 90%;
and if the result of the comparison of the local characteristic points is more than 90%, determining the target node image of which the result of the comparison of the local characteristic points is more than 90% as the service node image.
Further, the image and business process correspondence rule is as follows:
inputting a portrait behavior image related to an initial node according to the initial node of a target business process;
and when the portrait behavior image appears in the frame image, judging that the current frame image is the service node image of the target service process.
Further, the step of comparing the target node image with the standard image by an image comparison technique to obtain a plurality of service node images includes:
carrying out image comparison on the target node image and the standard image through an image comparison technology;
judging whether the image comparison result is qualified or not;
if not, extracting a next target node image, and carrying out image comparison on the next target node image and the standard image through an image comparison technology;
and if so, associating and corresponding the target node image with the qualified image comparison result with the corresponding business process.
Further, the step of segmenting the video data according to the video start node and the video end node to obtain the target video record includes:
acquiring a video node of the target video record;
and marking a service serial number corresponding to the service process on the video node.
Further, the step of performing comparative analysis on the frame image through the pre-trained image and the business process corresponding rule to screen out a plurality of target node images comprises;
and carrying out contrast analysis on the frame images in the acquisition sequence of the frame images.
In another aspect, the present invention provides a video segmentation system, which includes:
the acquisition module is used for acquiring video data and carrying out frame image segmentation on the video data to obtain a plurality of frame images;
the comparison analysis module is used for comparing and analyzing the frame image through the pre-trained image and the business process corresponding rule to obtain a plurality of business node images;
the correlation module is used for correlating and corresponding the business node images with the business process according to the image and business process corresponding rules, and the start or end time nodes of the business node images after correlation and corresponding to the business process are the start or end time nodes of the business nodes in the business process;
and the segmentation module is used for acquiring video starting and ending nodes of a target video record corresponding to the business process according to the starting time node of the business node in the business process and the ending time node of the business node in the business process, and segmenting the video data according to the video starting node and the video ending node to obtain the target video record corresponding to the business process.
The video segmentation system obtains a plurality of frame images by segmenting the frame images of the video data, thereby segmenting the video images into a plurality of static behavior images and comparing and analyzing the images with the corresponding rules of the business process, improving the accuracy of video analysis, avoiding the condition that the analysis result is inaccurate because the dynamic video is directly analyzed, further, obtaining the start and end time nodes of the business node in the business process video by correlating and corresponding the business node images with the business process, so as to obtain the start and end nodes of the video of the target video corresponding to the business process, thereby segmenting the video data according to the start node and the end node of the video to obtain the target video, when the related business process video needs to be obtained, only needing to directly call the target video corresponding to the business process, the video is not required to be played back from the beginning, so that the video record is quickly tracked and played back, and the technical problem that the video record cannot be quickly tracked and played back in the prior art is solved.
Another aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a video slicing method as described above.
Another aspect of the present invention also provides a data processing apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the video slicing method as described above when executing the program.
Drawings
FIG. 1 is a flow chart of a video slicing method according to a first embodiment of the present invention;
FIG. 2 is a flowchart of the first embodiment of the present invention after step S103 is detailed;
FIG. 3 is a flowchart of the first embodiment of the present invention after step S103 is refined again;
fig. 4 is a block diagram of a video slicing system according to a second embodiment of the present invention.
The following detailed description will further illustrate the invention in conjunction with the above-described figures.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Several embodiments of the invention are presented in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
The invention utilizes the image characteristic recognition technology to carry out frame picture segmentation on the video, comprehensively masters the behaviors of dynamic articles, static articles and the like in the video, and carries out combined recognition analysis on each frame picture according to the preset behavior rules. The video pictures are split into static behavior images for comparative analysis, so that the fault tolerance rate can be improved, and the effective analysis of the whole process behavior is supported. And secondly, the analysis accuracy is improved, the multi-behavior combination analysis is supported, and the detailed comparison of the characteristic points can be carried out.
The method is characterized in that a plurality of specific behavior actions in the process of the acquiescence and the punishment are associated with a service flow, the acquiescence and the punishment synchronous audio-video record is identified by utilizing an image identification technology, when a set behavior action occurs, node segmentation is carried out, the association with the service flow is automatically carried out, each section of video stream of the acquiescence and the punishment synchronous audio-video record is determined to correspond to a corresponding flow node, and the follow-up backtracking can be quickly tracked and played back on site.
Example one
Referring to fig. 1, a video slicing method according to a first embodiment of the present invention is shown, and the method includes steps S101 to S104:
s101, video data are obtained, and frame image segmentation is carried out on the video data to obtain a plurality of frame images.
In the case handling business of the criminal conviction and the penalty of the inspection institution, cases which are voluntarily convinced and the penalty need to be summoned to an inspection officer, an on-duty lawyer and a criminal suspect for carrying out the negotiation and communication of the debate, and in order to guarantee the right of the criminal suspect, the whole-process recording is needed in the period.
The invention utilizes the image characteristic recognition technology to carry out frame picture segmentation on the video, comprehensively masters the behaviors of dynamic articles, static articles and the like in the video, and carries out combined recognition analysis on each frame picture according to the preset behavior rules. The video pictures are divided into static behavior images for comparison and analysis, so that the analysis accuracy is improved, multi-behavior combination analysis is supported, and detailed comparison of characteristic points can be performed.
S102, comparing and analyzing the frame images through the pre-trained images and the business process corresponding rules, and screening out a plurality of target node images.
Specifically, the image and business process corresponding rule is that a portrait behavior image related to an initial node is input according to the initial node of a target business process; and when the portrait behavior image appears in the frame image, judging that the current frame image is the service node image of the target service process.
The image and business process corresponding rule comprises a plurality of standard images, and each standard image corresponds to a business process node. Before comparing frame images, different images need to be established according to different business processes, and image and business process corresponding rules are established for the images and the business processes, so that when the images which are the same as the business process nodes appear, time points can be obtained in time, and videos corresponding to different business processes can be classified conveniently.
In this embodiment, according to the inspection business process, the right obligation notification, evidence presentation, observation of crime advice suggestion, final statement of criminal suspect, and signing of the crime and penalty instrument conclusion are respectively performed. In this embodiment, the initial behavior images are set for the five business links respectively according to different initial behaviors of different links, for example, when a scouting officer stands up and holds a document to read, the image is determined as a right obligation informing link; when the criminal suspect holds the sign pen by hand to sign the document, the criminal suspect is determined to be a link of signing, convincing, penalizing and closing the document. According to the synchronous audio and video recording videos, images are automatically extracted for comparison and analysis, and the videos are effectively segmented according to the business process.
Specifically, the business process of the "right obligation informing link" is taken as an example, when a frame image of "the inspector seat stands up and the document is held by hand for interpretation" appears, the frame image is taken as a starting node of the "right obligation informing link" at the moment until the "inspector seat of the last frame is taken down and the document is handed over to the criminal suspect seat for reading and signing", so that the frame image can be understood as an ending node of the "right obligation informing link", and during the recording period of the recorded video, the video is divided into the "right obligation informing link" video, so that the direct calling of the "right obligation informing link" video in the later period is facilitated, and the calling efficiency is increased.
Furthermore, when the frame images are compared and analyzed, the frame images are compared and analyzed according to the acquisition sequence of the frame images, so that the video content between the starting node and the ending node of the business process is ensured to be all the content of the business process, and the content of other business processes is avoided.
In this step, a first round of screening is performed on a large number of frame images according to the rule that the images correspond to the service flows, so that frame images conforming to any one service flow are obtained, that is, a plurality of target node images are selected, and then, secondary screening is performed on the frame images satisfying the service flows, so that service node images corresponding to the service flows one by one are obtained, that is, corresponding service node images are obtained from the target node images according to the service flows.
S103, comparing the target node image with the standard image through an image comparison technology to obtain a plurality of service node images, associating and corresponding the service node images with the service flow according to the image and service flow corresponding rules, wherein the start or end time node of the service node image associated and corresponding to the service flow is the start or end time node of the service node in the service flow.
The video data is subjected to frame picture segmentation to obtain a plurality of frame images, so that the video pictures are split into a plurality of static behavior images and standard images to be compared and analyzed, the accuracy of video analysis is improved, and the condition that the analysis result is inaccurate due to the fact that dynamic videos are directly analyzed is avoided.
In this embodiment, the service node images are associated and correspond to the service flows, so as to obtain a service node corresponding to each service flow, that is, a first frame image of the local service flow is a start time node of the local service flow, and a last frame image of the local service flow is an end time node of the local service flow.
Specifically, as shown in fig. 2, step S103 includes steps S10311 to S10313:
and S10311, carrying out full-image fuzzy comparison on the target node image and the standard image, and judging whether the result of the full-image fuzzy comparison of the target node image is greater than 70%.
And S10312, if the result of the full map fuzzy comparison is larger than 70%, performing local feature point comparison on the target node image of which the result of the full map fuzzy comparison is larger than 70%, and judging whether the result of the local feature point comparison is larger than 90%.
And S10313, if the result of the comparison of the local feature points is greater than 90%, determining the target node image of which the result of the comparison of the local feature points is greater than 90% as the service node image.
In this embodiment, the full-map fuzzy comparison is performed on the target node image, then, the local feature point comparison is performed on the target node image which meets the preset requirement in the full-map fuzzy comparison, when the local feature point comparison meets the preset requirement, the target node image can be determined to be the service node image, and the service node image is subjected to multiple comparison, so that the accuracy of the service node image is further ensured, the analysis accuracy is improved, and the fault tolerance is also improved.
Further, as shown in fig. 3, step S103 further includes steps S10321 to S10324:
and S10321, comparing the target node image with the standard image through an image comparison technology.
S10322, judging whether the image comparison result is qualified.
If not, go to step S10323;
if yes, go to step S10324;
s10323, extracting a next target node image, and carrying out image comparison on the next target node image and the standard image through an image comparison technology.
S10324, the target node images with qualified image comparison results are associated and correspond to the corresponding business processes.
As a specific example, in this step, when the acquisition of the start node and the end node of a service flow is completed, that is, when the video acquisition of the service flow is completed, the system automatically performs image comparison of the next service flow, so that in the video recording process, the target node image and the standard image can be continuously compared, thereby avoiding that the comparison is interrupted, which results in failure of the target video acquisition of the service flow.
In this embodiment, the process of comparing the target node image with the standard image through the image comparison technique belongs to a dynamic loop process, and the next target node image is automatically extracted for comparison. The comparison is stopped only after the video recording is finished and the recording equipment is closed, otherwise, the process is continued all the time.
S104, acquiring video starting and ending nodes of a target video record corresponding to the business process according to the starting time node of the business node in the business process and the ending time node of the business node in the business process, and segmenting the video data according to the video starting node and the video ending node to obtain the target video record corresponding to the business process.
Further, according to the start time node and the end time node of the frame image, the video start node and the end node of the target video record corresponding to the service flow may be converted, and the video content between the video start node and the video end node is the target video record corresponding to the service flow.
In order to further improve the calling efficiency of the target video record, in some optional embodiments, a video node of the target video record may be further obtained, a service serial number corresponding to the service flow is marked on the video node, the target video record is played and output in the MP4 format and is accompanied by the service serial number, when the target video record corresponding to the service flow needs to be extracted, the target video record can be quickly extracted according to the service serial number, and the efficiency of quickly tracking and playing back the video record is further improved.
In summary, in the video segmentation method in the above embodiments of the present invention, the video data is segmented into the frame images to obtain the plurality of frame images, so that the video images are segmented into the plurality of static behavior images and the images are compared and analyzed with the corresponding rules of the business process, the accuracy of video analysis is improved, and the situation that the analysis result is inaccurate due to the direct analysis of the dynamic video is avoided, further, the start and end time nodes of the business node in the business process video are obtained by associating and corresponding the business node image with the business process, so that the video start and end nodes of the target video corresponding to the business process are obtained, so that the video data is segmented according to the video start node and the video end node to obtain the target video, when the related business process video needs to be obtained, only the target video corresponding to the business process needs to be directly called, the video is not required to be played back from the beginning, so that the video record is quickly tracked and played back, and the technical problem that the video record cannot be quickly tracked and played back in the prior art is solved.
Example two
Referring to fig. 4, a video slicing system according to a second embodiment of the present invention is shown, the system comprising:
the acquisition module is used for acquiring video data and carrying out frame image segmentation on the video data to obtain a plurality of frame images;
the comparison analysis module is used for comparing and analyzing the frame image through the pre-trained image and the business process corresponding rule to obtain a plurality of business node images;
the correlation module is used for correlating and corresponding the business node images with the business process according to the image and business process corresponding rules, and the start or end time nodes of the business node images after correlation and corresponding with the business process are the start or end time nodes of the business nodes in the business process;
and the segmentation module is used for acquiring video starting and ending nodes of a target video record corresponding to the business process according to the starting time node of the business node in the business process and the ending time node of the business node in the business process, and segmenting the video data according to the video starting node and the video ending node to obtain the target video record corresponding to the business process.
In some alternative embodiments, the comparative analysis module comprises:
the screening unit is used for carrying out comparative analysis on the frame image through a pre-trained image and a business process corresponding rule to screen out a plurality of target node images, wherein the image and business process corresponding rule comprises a plurality of standard images, and each standard image corresponds to a business process node;
and the image comparison unit is used for carrying out image comparison on the target node image and the standard image through an image comparison technology to obtain a plurality of service node images.
In some optional embodiments, the image alignment unit may further include:
the full-image fuzzy comparison unit is used for performing full-image fuzzy comparison on the target node image and the standard image and judging whether the result of the full-image fuzzy comparison of the target node image is greater than 70%;
the local feature point comparison unit is used for comparing local feature points of the target node images with the overall image fuzzy comparison result of more than 70% if the overall image fuzzy comparison result is more than 70%, and judging whether the comparison result of the local feature points is more than 90%;
and the judging unit is used for judging the target node image with the local characteristic point comparison result being more than 90% as the service node image if the local characteristic point comparison result is more than 90%.
Further, in some optional embodiments, the image alignment unit may further include:
the image comparison subunit is used for carrying out image comparison on the target node image and the standard image through an image comparison technology;
the judging subunit is used for judging whether the image comparison result is qualified or not;
the first execution subunit is used for extracting a next target node image if the image comparison result is unqualified, and performing image comparison on the next target node image and the standard image through an image comparison technology;
and the second execution subunit is used for associating and corresponding the target node image with the qualified image comparison result with the corresponding business process if the image comparison result is qualified.
Further, in some optional embodiments, the method may further include:
an obtaining unit, configured to obtain a video node of the target video record;
and the marking unit is used for marking the service serial number corresponding to the service flow on the video node.
Further, in some optional embodiments, the screening unit may further include:
and the contrast analysis subunit is used for performing contrast analysis on the frame images according to the acquisition sequence of the frame images.
In summary, in the video segmentation system in the above embodiments of the present invention, the video data is segmented into the frame images to obtain the plurality of frame images, so that the video images are segmented into the plurality of static behavior images and the images are compared and analyzed with the corresponding rules of the business process, the accuracy of video analysis is improved, and the situation that the analysis result is inaccurate due to the direct analysis of the dynamic video is avoided, further, the start and end time nodes of the business node in the business process video are obtained by associating and corresponding the business node image with the business process, so that the video start and end nodes of the target video corresponding to the business process are obtained, so that the video data is segmented according to the video start node and the video end node to obtain the target video, when the related business process video needs to be obtained, only the target video corresponding to the business process needs to be directly called, the video is not required to be played back from the beginning, so that the video record is quickly tracked and played back, and the technical problem that the video record cannot be quickly tracked and played back in the prior art is solved.
Furthermore, an embodiment of the present invention also proposes a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method in the above-described embodiment.
Furthermore, an embodiment of the present invention also provides a data processing apparatus, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the method in the above-mentioned embodiment.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims (9)
1. A method for video segmentation, the method comprising:
acquiring video data, and carrying out frame picture segmentation on the video data to obtain a plurality of frame images;
comparing and analyzing the frame image through the pre-trained image and the business process corresponding rule to obtain a plurality of business node images;
according to the image and business process corresponding rule, the business node image and the business process are associated and corresponding, and the starting or ending time node of the business node image which is associated and corresponding to the business process is the starting or ending time node of the business node in the business process;
acquiring video starting and ending nodes of a target video record corresponding to the business process according to a starting time node of the business node in the business process and an ending time node of the business node in the business process, and segmenting the video data according to the video starting node and the video ending node to obtain the target video record corresponding to the business process;
the step of comparing and analyzing the frame image through the pre-trained image and the business process corresponding rule to obtain a plurality of business node images comprises the following steps:
comparing and analyzing the frame image through the pre-trained image and business process corresponding rule to screen out a plurality of target node images, wherein the image and business process corresponding rule comprises a plurality of standard images, and each standard image corresponds to a business process node;
and carrying out image comparison on the target node image and the standard image through an image comparison technology to obtain a plurality of service node images.
2. The video slicing method according to claim 1, wherein the step of comparing the target node image with the standard image by an image comparison technique to obtain a plurality of service node images comprises:
performing full-image fuzzy comparison on the target node image and the standard image, and judging whether the result of the full-image fuzzy comparison of the target node image is greater than 70%;
if the result after the full-image fuzzy comparison is more than 70%, performing local feature point comparison on target node images of which the result after the full-image fuzzy comparison is more than 70%, and judging whether the result after the local feature point comparison is more than 90%;
and if the result of the comparison of the local characteristic points is more than 90%, determining the target node image of which the result of the comparison of the local characteristic points is more than 90% as the service node image.
3. The video segmentation method according to claim 1, wherein the image-to-business process correspondence rule is:
inputting a portrait behavior image related to an initial node according to the initial node of a target business process;
and when the portrait behavior image appears in the frame image, judging that the current frame image is the service node image of the target service process.
4. The video slicing method according to claim 1, wherein the step of comparing the target node image with the standard image by an image comparison technique to obtain a plurality of service node images comprises:
carrying out image comparison on the target node image and the standard image through an image comparison technology;
judging whether the image comparison result is qualified or not;
if not, extracting a next target node image, and carrying out image comparison on the next target node image and the standard image through an image comparison technology;
and if so, associating and corresponding the target node image with the qualified image comparison result with the corresponding business process.
5. The video segmentation method according to claim 1, wherein the step of segmenting the video data according to the video start node and the video end node to obtain a target video corresponding to the business process comprises:
acquiring a video node of the target video record;
and marking a service serial number corresponding to the service process on the video node.
6. The video segmentation method according to claim 1, wherein the step of screening out a plurality of target node images by performing comparative analysis on the frame images according to the pre-trained image and the business process correspondence rule comprises;
and carrying out contrast analysis on the frame images in the acquisition sequence of the frame images.
7. A video slicing system, said system comprising:
the acquisition module is used for acquiring video data and carrying out frame image segmentation on the video data to obtain a plurality of frame images;
the comparison analysis module is used for comparing and analyzing the frame image through the pre-trained image and the business process corresponding rule to obtain a plurality of business node images;
the correlation module is used for correlating and corresponding the business node images with the business process according to the image and business process corresponding rules, and the start or end time nodes of the business node images after correlation and corresponding with the business process are the start or end time nodes of the business nodes in the business process;
the segmentation module is used for acquiring video starting and ending nodes of a target video record corresponding to the business process according to a starting time node of the business node in the business process and an ending time node of the business node in the business process, and segmenting the video data according to the video starting node and the video ending node to obtain the target video record corresponding to the business process;
the comparative analysis module comprises:
the screening unit is used for carrying out comparative analysis on the frame image through the pre-trained image and business process corresponding rule to screen out a plurality of target node images, wherein the image and business process corresponding rule comprises a plurality of standard images, and each standard image corresponds to a business process node;
and the image comparison unit is used for carrying out image comparison on the target node image and the standard image through an image comparison technology to obtain a plurality of service node images.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a video slicing method according to any one of claims 1 to 6.
9. A data processing apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the video slicing method of any one of claims 1 to 6 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111058621.1A CN113507642A (en) | 2021-09-10 | 2021-09-10 | Video segmentation method, system, storage medium and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111058621.1A CN113507642A (en) | 2021-09-10 | 2021-09-10 | Video segmentation method, system, storage medium and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113507642A true CN113507642A (en) | 2021-10-15 |
Family
ID=78017088
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111058621.1A Pending CN113507642A (en) | 2021-09-10 | 2021-09-10 | Video segmentation method, system, storage medium and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113507642A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2224357A1 (en) * | 2009-02-27 | 2010-09-01 | BRITISH TELECOMMUNICATIONS public limited company | Video segmentation |
| CN104077568A (en) * | 2014-06-23 | 2014-10-01 | 北京理工大学珠海学院 | High-accuracy driver behavior recognition and monitoring method and system |
| CN106202333A (en) * | 2016-07-02 | 2016-12-07 | 北京工业大学 | A kind of method of store in a warehouse Video processing and quick-searching |
| CN108363713A (en) * | 2017-12-20 | 2018-08-03 | 武汉烽火众智数字技术有限责任公司 | Video image information resolver, system and method |
| CN112261477A (en) * | 2020-10-22 | 2021-01-22 | 新东方教育科技集团有限公司 | Video processing method and device, training method and storage medium |
| CN112333549A (en) * | 2021-01-04 | 2021-02-05 | 深圳壹账通智能科技有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN113095202A (en) * | 2021-04-07 | 2021-07-09 | 中国工商银行股份有限公司 | Data segmentation method and device in double-record data quality inspection |
-
2021
- 2021-09-10 CN CN202111058621.1A patent/CN113507642A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2224357A1 (en) * | 2009-02-27 | 2010-09-01 | BRITISH TELECOMMUNICATIONS public limited company | Video segmentation |
| CN104077568A (en) * | 2014-06-23 | 2014-10-01 | 北京理工大学珠海学院 | High-accuracy driver behavior recognition and monitoring method and system |
| CN106202333A (en) * | 2016-07-02 | 2016-12-07 | 北京工业大学 | A kind of method of store in a warehouse Video processing and quick-searching |
| CN108363713A (en) * | 2017-12-20 | 2018-08-03 | 武汉烽火众智数字技术有限责任公司 | Video image information resolver, system and method |
| CN112261477A (en) * | 2020-10-22 | 2021-01-22 | 新东方教育科技集团有限公司 | Video processing method and device, training method and storage medium |
| CN112333549A (en) * | 2021-01-04 | 2021-02-05 | 深圳壹账通智能科技有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN113095202A (en) * | 2021-04-07 | 2021-07-09 | 中国工商银行股份有限公司 | Data segmentation method and device in double-record data quality inspection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9858340B1 (en) | Systems and methods for queryable graph representations of videos | |
| CN106530320B (en) | End-to-end image segmentation processing method and system | |
| JP2015087903A (en) | Apparatus and method for information processing | |
| US11205260B2 (en) | Generating synthetic defect images for new feature combinations | |
| CN112017056B (en) | Intelligent double-recording method and system | |
| CN110347866B (en) | Information processing method, information processing device, storage medium and electronic equipment | |
| CN110348392B (en) | Vehicle matching method and device | |
| CN110570348A (en) | Face image replacement method and device | |
| CN114758271A (en) | Video processing method, device, computer equipment and storage medium | |
| Ma et al. | Pandrol track fastener defect detection based on local convolutional neural networks | |
| CN113516251B (en) | Machine learning system and model training method | |
| CN113095204B (en) | Double-recording data quality inspection method, device and system | |
| CN113342942A (en) | Corpus automatic acquisition method and device, computer equipment and storage medium | |
| CN113486856A (en) | Driver irregular behavior detection method based on semantic segmentation and convolutional neural network | |
| CN113158773A (en) | Training method and training device for living body detection model | |
| CN113507642A (en) | Video segmentation method, system, storage medium and device | |
| CN116596895A (en) | Substation equipment image defect identification method and system | |
| CN111126493A (en) | Deep learning model training method and device, electronic equipment and storage medium | |
| CN116126740A (en) | Model in-loop test method and device, computer equipment and storage medium | |
| CN112884866B (en) | Coloring method, device, equipment and storage medium for black-and-white video | |
| CN111461901B (en) | Method and device for outputting vehicle insurance claim information | |
| CN113900935A (en) | Automatic defect identification method and device, computer equipment and storage medium | |
| CN110032933B (en) | Image data acquisition method and device, terminal and storage medium | |
| CN113971402A (en) | Content identification method, device, medium and electronic equipment | |
| Ai et al. | An anisotropic heat diffusion model for enhancing the extraction of underground rail track fasteners under extremely low and uneven illumination conditions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211015 |