CN113239236B - Video processing method and device, electronic equipment and storage medium - Google Patents
Video processing method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113239236B CN113239236B CN202110789911.7A CN202110789911A CN113239236B CN 113239236 B CN113239236 B CN 113239236B CN 202110789911 A CN202110789911 A CN 202110789911A CN 113239236 B CN113239236 B CN 113239236B
- Authority
- CN
- China
- Prior art keywords
- video
- videos
- trend
- distribution trend
- abnormal
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/75—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N17/004—Diagnosis, testing or measuring for television systems or their details for digital television systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The present disclosure relates to a video processing method, apparatus, electronic device, and storage medium, the method comprising: acquiring a first characteristic of each video in a plurality of videos; dividing the plurality of videos into a plurality of first video packets based on the first characteristic of each video; identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet; determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet; and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining the second distribution trend as a video processing result.
Description
Technical Field
The present disclosure relates to the field of internet, and in particular, to a video processing method and apparatus, an electronic device, and a storage medium.
Background
The distribution trend of abnormal videos in a plurality of videos in the video platform is information which is often required to be acquired in the operation and maintenance process of the video platform and is used for providing operation personnel so that the operation personnel can know the distribution situation of the abnormal videos in the video platform.
In the related technology, when the distribution trend of abnormal videos in a plurality of videos needs to be acquired, the whole amount of videos of a video platform are randomly sampled, the plurality of videos are randomly extracted, the abnormal videos in the extracted plurality of videos are identified, and the distribution trend of the abnormal videos of the videos is acquired according to the identification result. However, due to the characteristics of the random extraction method, the degree of association between the randomly extracted videos is low, and accordingly, the degree of association between the identified abnormal videos in the randomly extracted videos is low, which results in low accuracy of the distribution trend of the abnormal videos in the obtained videos.
Disclosure of Invention
The present disclosure provides a video processing method, apparatus, terminal, and storage medium to at least solve a problem of low accuracy of a distribution trend of an abnormal video among a plurality of video videos in the related art. The technical scheme of the disclosure is as follows:
according to a first aspect of the embodiments of the present disclosure, there is provided a video processing method, including:
acquiring a first characteristic of each video in a plurality of videos;
dividing the plurality of videos into a plurality of first video packets based on the first feature of each video;
identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet;
determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet;
and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining that the second distribution trend is a video processing result.
In some embodiments, after determining that the second distribution trend is a video processing result, the method further comprises:
determining an abnormal proportion corresponding to the plurality of videos based on the second distribution trend, wherein the abnormal proportion corresponding to the plurality of videos is the proportion of the number of abnormal videos in the plurality of videos to the number of the plurality of videos.
In some embodiments, the method further comprises:
under the condition that the trend trends of the second distribution trend and the first distribution trend do not meet the consistency condition, acquiring a second feature of each video in the plurality of videos, wherein the second feature is different from the first feature;
dividing the plurality of videos into a plurality of first video packets based on the second characteristic of each video;
identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet;
determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet;
and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining that the second distribution trend is a video processing result.
In some embodiments, determining a plurality of second video packets from the first video packets comprises:
determining a plurality of key points on a curve representing the first distribution trend, wherein each key point corresponds to one first video packet, the abscissa of each key point is a sequence number indicating the position of the first video packet in the plurality of first video packets, and the ordinate of each key point is the proportion of the first video packets corresponding to the key point, and the proportion is the proportion of the number of abnormal videos in the first video packets corresponding to the key point to the number of videos in the first video packets corresponding to the key point;
and determining the first video packet corresponding to each key point as a second video packet.
In some embodiments, dividing the plurality of videos into a plurality of first video packets based on the first feature of each video comprises:
sorting the plurality of videos according to a first characteristic of the video;
dividing each preset number of videos into a first video packet.
In some embodiments, identifying the abnormal videos in each of the first video packets, and determining the first distribution trend of the abnormal videos in the plurality of first video packets based on the first identification result of each of the first video packets comprises:
for each first video packet, sampling a first to-be-identified video from the first video packet, identifying the sampled first to-be-identified video, and obtaining a sampling evaluation result corresponding to the first video packet;
and obtaining the first trend distribution based on the sampling evaluation result corresponding to each first video packet.
In some embodiments, identifying anomalous video in each of the second video packets comprises:
for each of the second video packets, sampling a second video to be identified from the second video packet; and identifying the sampled second video to be identified.
In some embodiments, the consistency condition is that the degree of similarity of the trend of the second distribution trend with the trend of the first distribution trend is greater than a similarity threshold.
According to a second aspect of the embodiments of the present disclosure, there is provided a video processing apparatus including:
an acquisition module configured to acquire a first feature of each of a plurality of videos;
a dividing module configured to divide the plurality of videos into a plurality of first video packets based on a first feature of each of the videos;
a first determining module configured to identify abnormal videos in each of the first video packets, and determine a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each of the first video packets;
a second determining module configured to determine a plurality of second video packets from the first video packets, identify abnormal videos in each of the second video packets, and determine a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each of the second video packets;
and the third determining module is configured to determine the second distribution trend as a video processing result under the condition that the trend trends of the second distribution trend and the first distribution trend meet a consistency condition.
In some embodiments, the apparatus further comprises:
a fourth determining module configured to determine, after determining that the second distribution trend is a video processing result, an abnormal proportion corresponding to the plurality of videos based on the second distribution trend, wherein the abnormal proportion corresponding to the plurality of videos is a proportion of the number of abnormal videos in the plurality of videos to the number of the plurality of videos.
In some embodiments, the apparatus further comprises:
the re-execution module is configured to acquire a second feature of each of the plurality of videos under the condition that the trend of the second distribution trend and the trend of the first distribution trend do not meet a consistency condition, wherein the second feature is different from the first feature; dividing the plurality of videos into a plurality of first video packets based on the second characteristic of each video; identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet; determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet; and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining that the second distribution trend is a video processing result.
In some embodiments, the second determination module is further configured to determine a plurality of key points on the curve representing the first distribution trend, wherein each key point corresponds to one first video packet, the abscissa of the key point is a sequence number indicating the position of the first video packet in the plurality of first video packets, and the ordinate of the key point is a proportion of the first video packet corresponding to the key point, the proportion being a proportion of the number of abnormal videos in the first video packet corresponding to the key point to the number of videos in the first video packet corresponding to the key point; and determining the first video packet corresponding to each key point as a second video packet.
In some embodiments, the partitioning module is further configured to sort the plurality of videos according to a first characteristic of the videos; dividing each preset number of videos into a first video packet.
In some embodiments, the first determining module is further configured to, for each of the first video packets, sample a first to-be-identified video from the first video packet, identify the sampled first to-be-identified video, and obtain a sampling evaluation result corresponding to the first video packet; and obtaining the first trend distribution based on the sampling evaluation result corresponding to each first video packet.
In some embodiments, the second determining module is further configured to, for each of the second video packets, sample a second video to be identified from the second video packet; and identifying the sampled second video to be identified.
In some embodiments, the consistency condition is that the degree of similarity of the trend of the second distribution trend with the trend of the first distribution trend is greater than a similarity threshold.
According to a third aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including: a processor;
wherein the processor is configured to execute the instructions to implement the method as in any one of the first aspect.
According to a fourth aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium, in which instructions, which, when executed by a processor of an electronic device, enable the electronic device to perform the method of any one of the first aspects.
According to a fifth aspect of embodiments of the present disclosure, there is provided a computer program product comprising computer readable code which, when run on an electronic device, causes the electronic device to perform the method according to any one of the first aspect.
The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects:
the video packets are divided according to the first characteristics of the videos, the videos with similar first characteristics can be divided into the same first video packet, each first video packet comprises a plurality of videos with similar first characteristics, the association degree of the videos in each first video packet is high, the second video packets are determined from the first video packets, the association degree of the videos in each second video packet is high, correspondingly, the association degree of the abnormal videos in each second video packet is high, meanwhile, as the second video packets are determined from the first video packets, each second video packet is identified in the process of identifying the abnormal videos in each first video packet, the identification of the abnormal videos in each second video packet is equivalent to the identification of the abnormal videos in each second video packet again, the abnormal videos in the second video packets are identified for multiple times, the probability that the abnormal videos in the second video packets are identified can be improved, the second identification result of each second video packet is more accurate compared with the first identification result of each first video packet obtained through one-time identification, and the accuracy of a video processing result, namely the second distribution trend of the abnormal videos in the second video packets, constructed by the second identification result of each second video packet is high.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
FIG. 1 is a flow diagram illustrating one embodiment of a video processing method in accordance with an exemplary embodiment;
fig. 2 shows an effect diagram in which trend trends of the second distribution trend and the first distribution trend satisfy a consistency condition;
FIG. 3 is a block diagram illustrating the structure of a video processing apparatus according to an exemplary embodiment;
fig. 4 is a block diagram illustrating another video processing apparatus according to an example embodiment;
fig. 5 is a block diagram illustrating a structure of an electronic device according to an example embodiment.
Detailed Description
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
Fig. 1 is a flow diagram illustrating one embodiment of a video processing method, which may be performed by an electronic device, according to an example embodiment, including the following steps 101-105.
In the present disclosure, a first feature of each of a plurality of videos may be acquired. For example,
the first feature of the video may be a violation risk value of the video, which may indicate a probability that the video is an abnormal video. For any video, a plurality of key frames are extracted from the video, and the extracted key frames are input into a neural network which is trained in advance and used for predicting violation risk values, such as a convolutional neural network, so that the violation risk value of the video, which is output by the neural network which is trained in advance and used for predicting the violation risk value, can be obtained.
In the present disclosure, the plurality of videos may be divided into a plurality of first video packets based on the first feature of each video.
For example, the first feature of the video is the violation risk value of the video, and a plurality of preset violation risk value intervals may be preset. The preset violation risk value interval in which each of the multiple videos is located can be determined, and videos in which the violation risk values in the multiple videos are located in the same preset violation risk value interval can be divided into a first video group.
In some embodiments, dividing the plurality of videos into a plurality of first video packets based on the first characteristic of each video comprises: sorting the plurality of videos according to a first feature of the videos; each preset number of videos is divided into a first video packet.
For example, the first feature of the video is the violation risk value of the video, and the videos are sorted according to the violation risk value of the video from high to low. The preset number is n, after the ordering is performed, the 1 st to n th videos are divided into the 1 st first video packet, the (n + 1) th to 2n th videos are divided into the 2 nd first video packet, and so on.
In the disclosure, a plurality of videos are sequenced according to a first feature of the videos, and each preset number of videos is divided into one first video group, so that the plurality of videos with similar first features can be divided into the same first video group, the association degree of the videos in each first video group is higher, correspondingly, the association degree of abnormal videos in each first video group is higher, and the abnormal videos in each first video group are suitable for constructing a first distribution trend of the abnormal videos in the plurality of first video groups, so that the accuracy of the first distribution trend of the abnormal videos in the plurality of constructed first video groups is high.
And 103, identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on the first identification result of each first video packet.
In the disclosure, for each first video packet, the abnormal video in the first video packet may be identified, a first identification result of the first video packet is obtained, and the identified abnormal video in the first video packet may be determined according to the first identification result of the first video packet.
In the present disclosure, for each first video packet, the number of identified abnormal videos in the first video packet may be divided by the number of videos targeted by step 101 to obtain an abnormal proportion corresponding to the first video packet, and a first distribution trend of the abnormal videos is determined based on the abnormal proportion. In one embodiment, a first data point corresponding to the first video packet may be added to the coordinate system, where an abscissa of the first data point corresponding to the first video packet is a sequence number indicating a position of the first video packet in a plurality of first video packets, and an ordinate of the first data point corresponding to the first video packet is an abnormal proportion corresponding to the first video packet.
For example, the number of the plurality of videos is x, and the plurality of first video packets sequentially includes a first video packet 1, a first video packet 2, and a first video packet 3. The abscissa of the first data point corresponding to the first video packet 1 is 1, the ordinate corresponding to the first video packet 1 is the abnormal proportion corresponding to the first video packet 1, that is, the number of identified abnormal videos in the first video packet 1 is divided by x, the abscissa of the first data point corresponding to the first video packet 2 is 2, the ordinate corresponding to the first video packet 2 is the abnormal proportion corresponding to the first video packet 2, that is, the number of identified abnormal videos in the first video packet 2 is divided by x, the abscissa of the first data point corresponding to the first video packet 3 is 3, the ordinate corresponding to the first video packet 3 is the abnormal proportion corresponding to the first video packet 3, that is, the number of identified abnormal videos in the first video packet 3 is divided by x, and so on.
And fitting the first data points corresponding to all the first video packets to obtain a curve representing the first distribution trend of the abnormal videos in the plurality of first video packets, thereby obtaining the first distribution trend representing the abnormal videos in the plurality of first video packets.
In some embodiments, identifying the anomalous video in each of the first video packets, and based on the first identification of each of the first video packets, determining a first distribution trend for the anomalous video in the plurality of first video packets comprises: for each first video packet, sampling a first to-be-identified video from the first video packet, identifying the sampled first to-be-identified video, and obtaining a sampling evaluation result corresponding to the first video packet; a first trend distribution is obtained based on the sample evaluation result corresponding to each of the first video packets.
In this disclosure, for each first video packet, a first to-be-identified video may be sampled from the first video packet, and a ratio of the number of the first to-be-identified videos sampled from the first video packet to the number of videos in the first video packet may be a preset ratio. For example, the preset ratio is 0.5, and for a first video packet, the number of videos in the first video packet is 100, and the number of videos extracted from the first video packet is 50, which is the product of 100 and the preset ratio.
For each first video packet, abnormal videos in all first videos to be identified extracted from the first video packet can be identified, sampling evaluation results corresponding to the first video packet are obtained, and the identified abnormal videos in all first videos to be identified extracted from the first video packet can be determined according to the sampling evaluation results corresponding to the first video packet.
For each first video packet, the number of identified abnormal videos in all first videos to be identified extracted from the first video packet may be divided by the number of videos targeted in step 101 to obtain a sampling abnormal proportion corresponding to the first video packet, and a first distribution trend of the abnormal videos is determined based on the abnormal proportion. In one embodiment, a second data point corresponding to the first video packet may be added to the coordinate system, where an abscissa of the second data point corresponding to the first video packet is a sequence number indicating a position of the first video packet in a plurality of first video packets, and an ordinate of the second data point corresponding to the first video packet is a sampling anomaly ratio corresponding to the first video packet. And fitting the second data points corresponding to all the first video packets to obtain a curve representing the first distribution trend of the abnormal videos in the plurality of first video packets, thereby obtaining the first distribution trend representing the abnormal videos in the plurality of first video packets.
In the present disclosure, when determining the first distribution trend of the abnormal videos in the plurality of first video packets, the abnormal videos in all the first videos to be identified sampled from the first video packets may be identified, and the first distribution trend of the abnormal videos in the plurality of first video packets is obtained by using the identified abnormal videos in all the first videos to be identified sampled from the first video packets. Compared with the identification of the abnormal videos in all the videos in the first video packet, the resource consumed by the identification of the abnormal videos can be reduced, and further the resource consumed by obtaining the first distribution trend of the abnormal videos in the plurality of first video packets is reduced.
And 104, determining a plurality of second video packets from the plurality of first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet.
In the present disclosure, a plurality of second video packets may be determined from the plurality of first video packets.
In one embodiment, the plurality of second video packets may be randomly extracted from the plurality of first video packets.
In another embodiment, the plurality of first video packets are sorted according to the first characteristic, and a first video packet with a preset number, which is sorted at the top, a first video packet with a preset number, which is sorted at the tail, and a plurality of first video packets, which are sorted in the middle, are selected as the second video packet. For example, the plurality of first video packets have an order, each of a first preset number of first video packets in the plurality of first video packets may be determined as a second video packet, and each of a second preset number of first video packets in the plurality of first video packets may be determined as a second video packet. A plurality of first video packets may be extracted from all remaining first video packets except for a first video packet of a preset number and a second video packet of a preset number, and each of the extracted first video packets may be regarded as a second video packet. In extracting the first video packet, the first video packet may be extracted at a preset interval indicating the number of first video packets between the currently extracted first video packet and the next extracted first video packet. For another example, if there are 100 first video packets, which are ordered as packet 1, packet 2, packet 3 … …, packets 50-60, packets 70-80 … …, packet 98, packet 99, packet 100, and the predetermined number is 3, then the selected second video packets may be packet 1, packet 2, packet 3, packet 50, packet 60, packet 70, packet 80, packet 98, packet 99, packet 100.
In some embodiments, determining the plurality of second video packets from the first video packets comprises: determining a plurality of key points on a curve representing the first distribution trend, wherein each key point corresponds to one first video packet, the abscissa of the key point is a serial number indicating the position of the first video packet in the plurality of first video packets, and the ordinate of the key point is the proportion of the first video packet corresponding to the key point, and the proportion is the proportion of the number of abnormal videos in the first video packet corresponding to the key point to the number of videos in the first video packet corresponding to the key point; and determining the first video packet corresponding to each key point as a second video packet.
For example, a curve representing a first distribution trend of the abnormal videos in the plurality of first video packets is obtained by fitting first data points corresponding to each first video packet, and the first data points corresponding to each first video packet are located on the curve representing the first distribution trend of the abnormal videos in the plurality of first video packets. Each of the first data points of the first preset number of points of all the first data points may be determined as a keypoint, and the following types of first data points of all the first data points may be determined as keypoints: head type, tail type, inflection type. The head type first data point may refer to a first data point of a previous preset number of first data points among all the first data points, and the tail type first data point may refer to a first data point of a next preset number of first data points among all the first data points.
In the present disclosure, a plurality of key points on a curve representing the first distribution trend may be determined, and the first video packet corresponding to each key point is determined as the second video packet. The key points are points suitable for constructing the distribution trend, correspondingly, the video groups corresponding to the key points are video groups suitable for constructing the distribution trend, the first video group corresponding to each key point is determined to be a second video group, each second video group is a video group suitable for constructing the distribution trend, and therefore the accuracy of the second distribution trend of abnormal videos in a plurality of second video groups constructed by each second video group is high.
In the disclosure, for each second video packet, the abnormal video in the second video packet may be identified, a second identification result of the second video packet is obtained, and the identified abnormal video in the second video packet may be determined according to the second identification result of the second video packet.
In some embodiments, identifying the anomalous video in each second video packet comprises: for each second video packet, sampling a second video to be identified from the second video packet; and identifying the sampled second video to be identified.
In this disclosure, for each second video packet, a second to-be-identified video may be sampled from the second video packet, and a ratio of the number of the second to-be-identified videos sampled from the second video packet to the number of videos in the second video packet may be a preset ratio.
For example, the preset ratio is 0.6, and for a second video packet, the number of videos in the second video packet is 100, and the number of videos extracted from the second video packet is 60, which is the product of 100 and the preset ratio.
For each second video packet, identifying the sampled second to-be-identified video, obtaining a sampling evaluation result corresponding to the second video packet, and determining the identified abnormal video in all the second to-be-identified videos extracted from the second video packet according to the sampling evaluation result corresponding to the second video packet.
In the present disclosure, when determining the second distribution trend of the abnormal video in the plurality of second video packets, the abnormal video in all the second videos to be identified sampled in the second video packets may be identified, and the second distribution trend of the abnormal video in the plurality of second video packets may be obtained by using the identified abnormal video in all the second videos to be identified sampled in the second video packets. Resources consumed in identifying the abnormal video can be reduced relative to identifying the abnormal video in all videos in the second video packet, and thus resources consumed in obtaining the second distribution trend of the abnormal video in the plurality of second video packets can be reduced.
For each second video packet, the number of identified abnormal videos in the second video packet may be divided by the number of videos targeted by step 101 to obtain an abnormal proportion corresponding to the second video packet, and a second distribution trend of the abnormal videos may be determined based on the abnormal proportion. In one embodiment, for each second video packet, if an abnormal video in the second video packet is identified, the abnormal proportion corresponding to the second video packet may be the number of the identified abnormal videos in the second video packet divided by the number of the videos targeted in step 101, and if a second video to be identified sampled from the video packet is identified, the abnormal proportion corresponding to the second video packet may be the number of the identified abnormal videos in all the second videos to be identified extracted from the second video packet divided by the number of the videos.
A third data point may be generated for each second video packet. In the coordinate system, the abscissa of the third data point corresponding to the second video packet is a sequence number indicating the position of the second video packet in the plurality of second video packets, and the ordinate of the third data point corresponding to the second video packet is an abnormal proportion corresponding to the second video packet. The third data points corresponding to all the second video packets may be fitted to obtain a curve representing a second distribution trend of the abnormal video in the plurality of second video packets, so as to determine the second distribution trend of the abnormal video in the plurality of second video packets.
And 105, under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining the second distribution trend as a video processing result.
In the present disclosure, the trend of the first distribution tendency is described by a curve representing the first distribution tendency, and the trend of the second distribution tendency is described by a curve representing the second distribution tendency. The degree of difference between the curve representing the first distribution trend and the curve representing the second distribution trend may be calculated, for example, the Longest Common subsequence (LCSS), Dynamic Time Warping (DTW), and the like may be used to calculate the degree of difference between the two curves, and the degree of difference between the curve representing the first distribution trend and the curve representing the second distribution trend is calculated by using one of the above algorithms, and the degree of difference between the curve representing the first distribution trend and the curve representing the second distribution trend is the degree of difference between the trend of the first distribution trend and the trend of the second distribution trend. The degree of difference between the trend of the first distribution trend and the trend of the second distribution trend is a numerical value within the interval of 0,1, if the degree of difference between the trend of the first distribution trend and the trend of the second distribution trend is 0, the trend of the first distribution trend is consistent with the trend of the second distribution trend, that is, the first distribution trend is the same as the second distribution trend, and if the degree of difference between the trend of the first distribution trend and the trend of the second distribution trend is 1, the trend of the first distribution trend is not associated with the trend of the second distribution trend, that is, the first distribution trend is not associated with the second distribution trend.
In one embodiment of the present disclosure, the consistency condition may be: the degree of difference between the trend of the first distribution trend and the trend of the second distribution trend is smaller than the threshold value of the degree of difference. After the difference degree between the trend of the first distribution trend and the trend of the second distribution trend is calculated, the difference degree between the trend of the first distribution trend and the trend of the second distribution trend can be compared with a difference degree threshold, and if the difference degree between the trend of the first distribution trend and the trend of the second distribution trend is smaller than the difference degree threshold, the trend of the second distribution trend and the trend of the first distribution trend can be determined to meet the consistency condition.
In the present disclosure, in a case where the trend of the second distribution trend and the trend of the first distribution trend satisfy a consistency condition, the second distribution trend is determined as a video processing result.
Please refer to fig. 2, which shows an effect diagram of the trend of the second distribution trend and the first distribution trend satisfying the consistency condition.
In fig. 2, a coordinate system, a curve representing a first distribution trend, a curve representing a second distribution trend, a part of data points on the curve representing the first distribution trend, a part of data points on the curve representing the second distribution trend are shown. Data points are represented by black dots. The value of the X axis of the coordinate system is the serial number of the video packet, and the value of the Y axis is the abnormal proportion corresponding to the video packet. Each data point on the curve representing the first distribution trend corresponds to a respective one of the first video packets, and each data point on the curve representing the second distribution trend corresponds to a respective one of the second video packets. As can be seen from fig. 2, the curve representing the first distribution trend and the curve representing the second distribution trend are decreasing curves, and accordingly, the trend of the first distribution trend and the trend of the second distribution trend are decreasing trends, and the difference between the trend of the first distribution trend and the trend of the second distribution trend is small. The difference degree between the trend of the first distribution trend and the trend of the second distribution trend can be calculated, the difference degree between the trend of the first distribution trend and the trend of the second distribution trend is determined to be smaller than a difference degree threshold value, and the trend trends of the second distribution trend and the first distribution trend are determined to meet the consistency condition.
In the disclosure, each of the first video packets is divided according to the first feature of the video, the videos with similar first features may be divided into the same first video packet, each of the first video packets includes a plurality of videos with similar first features, so that the relevance of the videos in each of the first video packets is high, a plurality of second video packets are determined from the plurality of first video packets, the relevance of the videos in each of the second video packets is high, and accordingly, the relevance of the abnormal videos in each of the second video packets is high, and at the same time, since each of the second video packets is determined from the plurality of first video packets, each of the second video packets has been identified in the process of identifying the abnormal videos in each of the first video packets, identifying the abnormal videos in each of the second video packets is equivalent to identifying the abnormal videos in each of the second video packets again, the abnormal videos in the second video packets are identified for multiple times, the probability that the abnormal videos in the second video packets are identified can be improved, the second identification result of each second video packet is more accurate compared with the first identification result of each first video packet obtained through one-time identification, and the accuracy of a video processing result constructed by the second identification result in each second video packet, namely the second distribution trend of the abnormal videos in the second video packets, is high.
In some embodiments, the consistency condition is: the similarity between the trend of the second distribution trend and the trend of the first distribution trend is larger than a similarity threshold.
In the present disclosure, the degree of difference between the trend of the first distribution trend and the trend of the second distribution trend may be subtracted from 1 to obtain the similarity between the trend of the second distribution trend and the trend of the first distribution trend. Then, the similarity between the trend of the second distribution trend and the trend of the first distribution trend is compared with a similarity threshold, wherein the similarity threshold is a preset numerical value smaller than 1, such as 0.8, and if the similarity between the trend of the second distribution trend and the trend of the first distribution trend is larger than the similarity threshold, the trend of the second distribution trend and the trend of the first distribution trend are determined to meet the consistency condition.
The consistency condition may be used to determine whether the degree of similarity between the trend of the second distribution trend and the trend of the first distribution trend is higher, and if the trend of the second distribution trend and the trend of the first distribution trend satisfy the consistency condition, it is proved that the degree of similarity between the trend of the second distribution trend and the trend of the first distribution trend is higher. The consistency condition may also be used to determine whether the degree of difference between the trend of the second distribution trend and the trend of the first distribution trend is low, and if the trend of the second distribution trend and the trend of the first distribution trend satisfy the consistency condition, it is proved that the degree of difference between the trend of the second distribution trend and the trend of the first distribution trend is low.
In the present disclosure, it may be determined as a consistency condition that the degree of similarity between the trend of the second distribution trend and the trend of the first distribution trend is greater than the similarity threshold or the degree of difference between the trend of the first distribution trend and the trend of the second distribution trend is less than the difference threshold, so that it may be accurately determined whether the degree of similarity between the trend of the second distribution trend and the trend of the first distribution trend is higher. In one embodiment, the threshold degree of difference is 0.2.
In some embodiments, after determining that the second distribution trend is a video processing result, the method further comprises: and determining an abnormal proportion corresponding to the plurality of videos based on the second distribution trend, wherein the abnormal proportion corresponding to the plurality of videos is the proportion of the number of abnormal videos in the plurality of videos to the number of the plurality of videos. The curve representing the second distribution trend in the coordinate system may be integrated, and an area enclosed by the curve representing the second distribution trend and an x-axis in the coordinate system and an area enclosed by the curve representing the second distribution trend and the x-axis in the coordinate system may be calculated as the abnormal proportion corresponding to the plurality of videos.
In the present disclosure, after determining that the second distribution trend is a video processing result, the abnormal proportion corresponding to the plurality of videos may also be determined based on the second distribution trend, so that the proportion of abnormal videos in the plurality of videos may be further known through the abnormal proportion corresponding to the plurality of videos.
In some embodiments, further comprising: under the condition that the trend trends of the second distribution trend and the first distribution trend do not meet the consistency condition, acquiring a second feature of each video in the plurality of videos, wherein the second feature is different from the first feature; dividing the plurality of videos into a plurality of first video packets based on the second characteristic of each video; identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet; determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet; and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining the second distribution trend as a video processing result.
If the trend of the second distribution trend and the trend of the first distribution trend do not satisfy the consistency condition, the accuracy of the second distribution trend may not satisfy the requirement, in the present disclosure, it is considered that the fact that the accuracy of the second distribution trend may not satisfy the requirement may be caused by the first feature being not suitable for obtaining the second distribution trend, if the trend of the second distribution trend and the trend of the first distribution trend does not satisfy the consistency condition, the second distribution trend may be continuously tried to be obtained by using the second feature of each of the plurality of videos, and the second distribution trend may be continuously tried to be obtained by using different features until the second distribution trend is successfully obtained. Therefore, the situation that the first utilized characteristic, namely the first characteristic, is not suitable for obtaining the second distribution trend so that the second distribution trend cannot be obtained can be avoided, and the flexibility of the process of obtaining the second distribution trend is enhanced.
Fig. 3 is a block diagram illustrating a configuration of a video processing apparatus according to an exemplary embodiment. Referring to fig. 3, the video processing apparatus includes: the device comprises an acquisition module 301, a division module 302, a first determination module 303, a second determination module 304 and a third determination module 305.
The obtaining module 301 is configured to obtain a first feature of each of a plurality of videos;
the dividing module 302 is configured to divide the plurality of videos into a plurality of first video packets based on the first characteristic of each video;
the first determining module 303 is configured to identify abnormal videos in each of the first video packets, and determine a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each of the first video packets;
the second determining module 304 is configured to determine a plurality of second video packets from the first video packets, identify abnormal videos in each of the second video packets, and determine a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each of the second video packets;
the third determining module 305 is configured to determine the second distribution trend as a video processing result if the trend of the second distribution trend and the first distribution trend satisfies a consistency condition.
In some embodiments, the second determining module 304 is further configured to determine a plurality of key points on the curve representing the first distribution trend, wherein each key point corresponds to one first video packet, the abscissa of the key point is a sequence number indicating the position of the first video packet in the plurality of first video packets, and the ordinate of the key point is a proportion of the first video packet corresponding to the key point, the proportion is a proportion of the number of abnormal videos in the first video packet corresponding to the key point to the number of videos in the first video packet corresponding to the key point; and determining the first video packet corresponding to each key point as a second video packet.
In some embodiments, the partitioning module 301 is further configured to sort the plurality of videos according to a first characteristic of the videos; dividing each preset number of videos into a first video packet.
In some embodiments, the first determining module 303 is further configured to, for each of the first video packets, sample a first to-be-identified video from the first video packet, identify the first to-be-identified video that is sampled, and obtain a sampling evaluation result corresponding to the first video packet; and obtaining the first trend distribution based on the sampling evaluation result corresponding to each first video packet.
In some embodiments, the second determining module 304 is further configured to, for each of the second video packets, sample a second video to be identified from the second video packet; and identifying the sampled second video to be identified.
In some embodiments, the consistency condition is one of: the similarity of the trend of the second distribution trend and the trend of the first distribution trend is larger than a similarity threshold, and the difference degree between the trend of the second distribution trend and the trend of the first distribution trend is smaller than a difference degree threshold.
Referring to fig. 4, a block diagram of another video processing apparatus according to an exemplary embodiment is shown. Referring to fig. 4, the video processing apparatus includes: the device comprises an acquisition module 401, a division module 402, a first determination module 403, a second determination module 404, a third determination module 405, a fourth determination module 406 and a re-execution module 407.
The acquiring module 401 has the same function as the acquiring module 301, the dividing module 402 has the same function as the dividing module 302, the first determining module 403 has the same function as the first determining module 303, the second determining module 404 has the same function as the second determining module 304, and the third determining module 405 has the same function as the third determining module 305.
The fourth determining module 406 is configured to determine, after determining that the second distribution trend is a video processing result, an abnormal proportion corresponding to the plurality of videos based on the second distribution trend, wherein the abnormal proportion corresponding to the plurality of videos is a proportion of the number of abnormal videos in the plurality of videos to the number of the plurality of videos.
The re-execution module 407 is configured to obtain a second feature of each of the plurality of videos if the trend of the second distribution trend and the trend of the first distribution trend do not satisfy a consistency condition, where the second feature is different from the first feature; dividing the plurality of videos into a plurality of first video packets based on the second characteristic of each video; identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet; determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet; and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining that the second distribution trend is a video processing result.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 5 is a block diagram illustrating a structure of an electronic device according to an example embodiment. Referring to fig. 5, the electronic device includes a processing component 522 that further includes one or more processors and memory resources, represented by memory 532, for storing instructions, such as application programs, that are executable by the processing component 522. The application programs stored in memory 532 may include one or more modules that each correspond to a set of instructions. Further, the processing component 522 is configured to execute instructions to perform the above-described methods.
The electronic device may also include a power supply component 526 configured to perform power management of the electronic device, a wired or wireless network interface 550 configured to connect the electronic device to a network, and an input/output (I/O) interface 558. The electronic device may operate based on an operating system stored in memory 332, such as Windows Server, MacOS XTM, UnixTM, LinuxTM, FreeBSDTM, or the like.
In an exemplary embodiment, a computer-readable storage medium comprising instructions, such as a memory comprising instructions, executable by an electronic device to perform the video processing method described above is also provided. Alternatively, the computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
In an exemplary embodiment, the present application further provides a computer program product comprising computer readable code which, when run on an electronic device, causes the electronic device to perform a video processing method.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (16)
1. A method of video processing, the method comprising:
acquiring a first characteristic of each video in a plurality of videos;
dividing the plurality of videos into a plurality of first video packets based on the first feature of each video;
identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet;
determining a plurality of second video packets from each of the first video packets, identifying abnormal videos in each of the second video packets, and determining a second distribution trend of abnormal videos in the plurality of second video packets based on a second identification result of each of the second video packets, the plurality of second video packets being different from each of the first video packets;
under the condition that the trend trends of the second distribution trend and the first distribution trend meet a consistency condition, determining that the second distribution trend is a video processing result, wherein the consistency condition is one of the following conditions: the similarity of the trend of the second distribution trend and the trend of the first distribution trend is larger than a similarity threshold, and the difference degree between the trend of the second distribution trend and the trend of the first distribution trend is smaller than a difference degree threshold.
2. The method of claim 1, wherein after determining that the second distribution trend is a video processing result, the method further comprises:
determining an abnormal proportion corresponding to the plurality of videos based on the second distribution trend, wherein the abnormal proportion corresponding to the plurality of videos is the proportion of the number of abnormal videos in the plurality of videos to the number of the plurality of videos.
3. The method of claim 1, further comprising:
under the condition that the trend trends of the second distribution trend and the first distribution trend do not meet the consistency condition, acquiring a second feature of each video in the plurality of videos, wherein the second feature is different from the first feature;
dividing the plurality of videos into a plurality of first video packets based on the second characteristic of each video;
identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet;
determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet;
and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining that the second distribution trend is a video processing result.
4. The method of claim 1, wherein determining a plurality of second video packets from the first video packets comprises:
determining a plurality of key points on a curve representing the first distribution trend, wherein each key point corresponds to one first video packet, the abscissa of each key point is a sequence number indicating the position of the first video packet in the plurality of first video packets, and the ordinate of each key point is the proportion of the first video packets corresponding to the key point, and the proportion is the proportion of the number of abnormal videos in the first video packets corresponding to the key point to the number of videos in the first video packets corresponding to the key point;
and determining the first video packet corresponding to each key point as a second video packet.
5. The method of claim 1, wherein dividing the plurality of videos into a plurality of first video packets based on the first characteristic of each video comprises:
sorting the plurality of videos according to a first characteristic of the video;
dividing each preset number of videos into a first video packet.
6. The method of claim 1, wherein identifying abnormal videos in each of the first video packets, and wherein determining a first distribution trend of abnormal videos in the plurality of first video packets based on the first identification result of each of the first video packets comprises:
for each first video packet, sampling a first to-be-identified video from the first video packet, identifying the sampled first to-be-identified video, and obtaining a sampling evaluation result corresponding to the first video packet;
and obtaining a first trend distribution based on the sampling evaluation result corresponding to each first video packet.
7. The method of claim 1, wherein identifying anomalous video in each of the second video packets comprises:
for each of the second video packets, sampling a second video to be identified from the second video packet; and identifying the sampled second video to be identified.
8. A video processing apparatus, characterized in that the apparatus comprises:
an acquisition module configured to acquire a first feature of each of a plurality of videos;
a dividing module configured to divide the plurality of videos into a plurality of first video packets based on a first feature of each of the videos;
a first determining module configured to identify abnormal videos in each of the first video packets, and determine a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each of the first video packets;
a second determining module configured to determine a plurality of second video packets from each of the first video packets, identify abnormal videos in each of the second video packets, and determine a second distribution trend of abnormal videos in the plurality of second video packets based on a second identification result of each of the second video packets, the plurality of second video packets being different from the each of the first video packets;
a third determining module configured to determine that the second distribution trend is a video processing result if the trend of the second distribution trend and the trend of the first distribution trend meet a consistency condition, where the consistency condition is one of: the similarity of the trend of the second distribution trend and the trend of the first distribution trend is larger than a similarity threshold, and the difference degree between the trend of the second distribution trend and the trend of the first distribution trend is smaller than a difference degree threshold.
9. The apparatus of claim 8, further comprising:
a fourth determining module configured to determine, after determining that the second distribution trend is a video processing result, an abnormal proportion corresponding to the plurality of videos based on the second distribution trend, wherein the abnormal proportion corresponding to the plurality of videos is a proportion of the number of abnormal videos in the plurality of videos to the number of the plurality of videos.
10. The apparatus of claim 8, further comprising:
the re-execution module is configured to acquire a second feature of each of the plurality of videos under the condition that the trend of the second distribution trend and the trend of the first distribution trend do not meet a consistency condition, wherein the second feature is different from the first feature; dividing the plurality of videos into a plurality of first video packets based on the second characteristic of each video; identifying abnormal videos in each first video packet, and determining a first distribution trend of the abnormal videos in the plurality of first video packets based on a first identification result of each first video packet; determining a plurality of second video packets from the first video packets, identifying abnormal videos in each second video packet, and determining a second distribution trend of the abnormal videos in the plurality of second video packets based on a second identification result of each second video packet; and under the condition that the trend trends of the second distribution trend and the first distribution trend meet the consistency condition, determining that the second distribution trend is a video processing result.
11. The apparatus of claim 8, wherein the second determining module is further configured to determine a plurality of key points on the curve representing the first distribution trend, wherein each key point corresponds to one first video packet, the abscissa of the key point is a sequence number indicating the position of the first video packet in the plurality of first video packets, and the ordinate of the key point is a proportion of the first video packet corresponding to the key point, and the proportion is a proportion of the number of abnormal videos in the first video packet corresponding to the key point to the number of videos in the first video packet corresponding to the key point; and determining the first video packet corresponding to each key point as a second video packet.
12. The apparatus of claim 8, wherein the partitioning module is further configured to order the plurality of videos according to a first characteristic of the videos; dividing each preset number of videos into a first video packet.
13. The apparatus of claim 8, wherein the first determining module is further configured to, for each of the first video packets, sample a first to-be-identified video from the first video packet, identify the first to-be-identified video sampled, and obtain a sample evaluation result corresponding to the first video packet; and obtaining a first trend distribution based on the sampling evaluation result corresponding to each first video packet.
14. The apparatus of claim 8, wherein the second determining module is further configured to, for each of the second video packets, sample a second video to be identified from the second video packets; and identifying the sampled second video to be identified.
15. An electronic device, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the method of any one of claims 1 to 7.
16. A computer-readable storage medium whose instructions, when executed by a processor of an electronic device, enable the electronic device to perform the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110789911.7A CN113239236B (en) | 2021-07-13 | 2021-07-13 | Video processing method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110789911.7A CN113239236B (en) | 2021-07-13 | 2021-07-13 | Video processing method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113239236A CN113239236A (en) | 2021-08-10 |
| CN113239236B true CN113239236B (en) | 2021-10-01 |
Family
ID=77135459
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110789911.7A Active CN113239236B (en) | 2021-07-13 | 2021-07-13 | Video processing method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113239236B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101971209A (en) * | 2008-01-11 | 2011-02-09 | 卓然(法国) | Sparse geometry for super resolution video processing |
| CN102938078A (en) * | 2012-11-30 | 2013-02-20 | 北京理工大学 | Unbalanced timing sequence-based monitoring video abnormal event detection method |
| CN105513095A (en) * | 2015-12-30 | 2016-04-20 | 山东大学 | Behavior video non-supervision time-sequence partitioning method |
| CN105741321A (en) * | 2016-01-31 | 2016-07-06 | 华南理工大学 | Video target movement tendency analysis method based on trace point distribution |
| CN107786850A (en) * | 2017-11-02 | 2018-03-09 | 上海达实联欣科技发展有限公司 | A kind of master control room intelligent monitor system and method |
| CN112329517A (en) * | 2020-09-17 | 2021-02-05 | 中国南方电网有限责任公司超高压输电公司南宁监控中心 | Transformer substation disconnecting link confirmation video image analysis method and system |
| CN112528801A (en) * | 2020-12-02 | 2021-03-19 | 上海高德威智能交通系统有限公司 | Abnormal event detection method, model training method and device |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6937164B2 (en) * | 2003-02-17 | 2005-08-30 | The Boeing Company | Methods and apparatus for transportation vehicle security monitoring |
| CN102254428B (en) * | 2011-04-28 | 2013-05-01 | 崔志明 | Traffic jam detection method based on video processing |
| CN103413321A (en) * | 2013-07-16 | 2013-11-27 | 南京师范大学 | Crowd behavior model analysis and abnormal behavior detection method under geographical environment |
| SG10201510337RA (en) * | 2015-12-16 | 2017-07-28 | Vi Dimensions Pte Ltd | Video analysis methods and apparatus |
| CN110414335A (en) * | 2019-06-20 | 2019-11-05 | 北京奇艺世纪科技有限公司 | Video frequency identifying method, device and computer readable storage medium |
| CN111046084A (en) * | 2019-12-18 | 2020-04-21 | 重庆大学 | Association rule mining method for multivariate time series monitoring data |
| AU2021100365A4 (en) * | 2021-01-20 | 2021-04-15 | Three Gorges University | A multi-sensor-based intelligent monitoring and early warning system and method for dam safety |
-
2021
- 2021-07-13 CN CN202110789911.7A patent/CN113239236B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101971209A (en) * | 2008-01-11 | 2011-02-09 | 卓然(法国) | Sparse geometry for super resolution video processing |
| CN102938078A (en) * | 2012-11-30 | 2013-02-20 | 北京理工大学 | Unbalanced timing sequence-based monitoring video abnormal event detection method |
| CN105513095A (en) * | 2015-12-30 | 2016-04-20 | 山东大学 | Behavior video non-supervision time-sequence partitioning method |
| CN105741321A (en) * | 2016-01-31 | 2016-07-06 | 华南理工大学 | Video target movement tendency analysis method based on trace point distribution |
| CN107786850A (en) * | 2017-11-02 | 2018-03-09 | 上海达实联欣科技发展有限公司 | A kind of master control room intelligent monitor system and method |
| CN112329517A (en) * | 2020-09-17 | 2021-02-05 | 中国南方电网有限责任公司超高压输电公司南宁监控中心 | Transformer substation disconnecting link confirmation video image analysis method and system |
| CN112528801A (en) * | 2020-12-02 | 2021-03-19 | 上海高德威智能交通系统有限公司 | Abnormal event detection method, model training method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113239236A (en) | 2021-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106982196B (en) | Abnormal access detection method and equipment | |
| CN111178380B (en) | Data classification method and device and electronic equipment | |
| EP4198775A1 (en) | Abnormal user auditing method and apparatus, electronic device, and storage medium | |
| CN109165691B (en) | Training method and device for model for identifying cheating users and electronic equipment | |
| CN110634471B (en) | Voice quality inspection method and device, electronic equipment and storage medium | |
| CN112000808B (en) | Data processing method and device and readable storage medium | |
| CN110427375B (en) | Method and device for identifying field type | |
| CN112364014B (en) | Data query method, device, server and storage medium | |
| CN111061948B (en) | User tag recommendation method and device, computer equipment and storage medium | |
| CN110147493B (en) | Method, device, computer equipment and storage medium for determining active factors | |
| CN111177500A (en) | Data object classification method and device, computer equipment and storage medium | |
| CN111783812A (en) | Method and device for identifying forbidden images and computer readable storage medium | |
| CN111414528B (en) | Method and device for determining equipment identification, storage medium and electronic equipment | |
| CN112232206B (en) | Face recognition method and face recognition platform based on big data and artificial intelligence | |
| CN113723555A (en) | Abnormal data detection method and device, storage medium and terminal | |
| CN112434651A (en) | Information analysis method and device based on image recognition and computer equipment | |
| CN113239236B (en) | Video processing method and device, electronic equipment and storage medium | |
| CN112532645A (en) | Internet of things equipment operation data monitoring method and system and electronic equipment | |
| CN109145207B (en) | Information personalized recommendation method and device based on classification index prediction | |
| CN113535458B (en) | Abnormal false alarm processing method and device, storage medium and terminal | |
| CN111258788A (en) | Disk failure prediction method, device and computer readable storage medium | |
| CN111340349A (en) | Product reliability evaluation method and device, computer equipment and storage medium | |
| CN113905400B (en) | Network optimization processing method and device, electronic equipment and storage medium | |
| CN113473124B (en) | Information acquisition method, device, electronic equipment and storage medium | |
| CN109376306B (en) | Service recommendation method and system based on tag panorama |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |