CN110659384B - Video structured analysis method and device - Google Patents
Video structured analysis method and device Download PDFInfo
- Publication number
- CN110659384B CN110659384B CN201810609988.XA CN201810609988A CN110659384B CN 110659384 B CN110659384 B CN 110659384B CN 201810609988 A CN201810609988 A CN 201810609988A CN 110659384 B CN110659384 B CN 110659384B
- Authority
- CN
- China
- Prior art keywords
- video
- detected
- video frame
- target
- frame
- 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
Abstract
The embodiment of the invention provides a video structured analysis method and a video structured analysis device, which are applied to the technical field of video analysis, wherein the video structured analysis method comprises the following steps: acquiring video data to be analyzed; determining whether a video frame to be detected in video data contains a target subject to be paid attention or not through a preset algorithm aiming at the video frame to be detected in the video data; and if the video frame to be detected contains the target main body, performing video structural analysis on the video frame to be detected. By the video structured analysis method, the video structured analysis is carried out on the video frame to be detected containing the target main body, the speed of carrying out the video structured analysis on the video data can be increased, and the efficiency of the video structured analysis is improved.
Description
Technical Field
The invention relates to the technical field of video analysis, in particular to a video structured analysis method and a video structured analysis device.
Background
Video data is used as an important source of visual perception of the Internet of things, and plays an increasingly important role in the field of public safety. The video structured analysis is an in-depth application aiming at unstructured video data, so that the video data becomes sensible and describable intelligent data, and the application field is extremely wide. For public safety, video structural analysis almost permeates to the aspect of public safety, and corresponding data are extracted by performing video structural analysis on video data shot by monitoring in public places.
In a related video structured analysis method, it is a common practice to analyze each video frame in a video to be analyzed frame by frame, thereby completing video structured analysis of the video to be analyzed. The existing video analysis method needs a large amount of time, and the efficiency of video structured analysis is low.
Disclosure of Invention
The embodiment of the invention aims to provide a video structural analysis method and a video structural analysis device so as to improve the efficiency of video structural analysis. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a video structured analysis method, where the method includes:
acquiring video data to be analyzed;
determining whether a video frame to be detected in the video data contains a concerned target main body or not through a preset algorithm;
and if the video frame to be detected contains the target main body, performing video structuralization analysis on the video frame to be detected.
Optionally, the determining, by using a preset algorithm, whether the video frame to be detected in the video data includes a target subject of interest includes:
determining a current video frame to be detected of the video data according to a preset video frame to be detected selection rule;
detecting whether the current video frame to be detected contains the target main body or not through a preset algorithm;
after performing video structuralization analysis on the video frame to be detected if the video frame to be detected contains the target main body, the method further comprises:
and if the current video frame to be detected contains the target main body, starting or keeping video structural analysis on each video frame between the current video frame to be detected and the next video frame to be detected in the video data.
Optionally, after the detecting, by the preset algorithm, whether the current video frame to be detected includes the target subject, the method further includes:
and if the current video frame to be detected does not contain the target main body, closing or not opening video structural analysis of a current video segment in the video data, wherein the current video segment comprises the current video frame to be detected and video frames between the current video frame to be detected and the next video frame to be detected.
Optionally, the detecting, by using a preset algorithm, whether the current video frame to be detected includes the target subject includes:
extracting foreground information of the current video frame to be detected by a preset background modeling method;
if the foreground information accords with a preset foreground rule, judging that the current video frame to be detected comprises the target main body;
and if the foreground information does not accord with the preset foreground rule, judging that the current video frame to be detected does not contain the target main body.
Optionally, the detecting, by using a preset algorithm, whether the current video frame to be detected includes the target subject includes:
and detecting whether the current video frame to be detected of the video data contains the target main body or not by a preset target detection method.
In a second aspect, an embodiment of the present invention provides a video structural analysis method, where the method includes:
the method comprises the steps of obtaining video data to be analyzed, and splitting the video data into a plurality of video segments according to a preset splitting rule;
respectively selecting a video frame to be detected in each video segment according to a preset video frame selection rule;
respectively judging whether each video frame to be detected contains a target subject to be paid attention to or not through a preset algorithm;
taking a video segment containing the video frame to be detected of the target main body as a target video segment;
and performing video structural analysis on each target video segment.
Optionally, the determining, by using a preset algorithm, whether each of the video frames to be detected includes a target subject to be paid attention to or not includes:
extracting foreground information of each video frame to be detected by a preset background modeling method aiming at each frame of the video frame to be detected;
if the foreground information accords with a preset foreground rule, judging that the video frame to be detected comprises the target main body;
and if the foreground information does not accord with the preset foreground rule, judging that the video frame to be detected does not contain the target main body.
Optionally, the determining, by using a preset algorithm, whether each video frame to be detected includes a target subject to be paid attention to or not includes:
and respectively detecting whether each video frame to be detected contains a target subject to be paid attention or not by a preset target detection method.
In a third aspect, an embodiment of the present invention provides a video structural analysis apparatus, where the apparatus includes:
the video data acquisition module is used for acquiring video data to be analyzed;
the target main body judging module is used for determining whether the video frame to be detected contains a concerned target main body or not through a preset algorithm aiming at the video frame to be detected in the video data;
and the first structure analysis module is used for carrying out video structural analysis on the video frame to be detected if the video frame to be detected contains the target main body.
Optionally, the target subject determination module includes:
the video frame determination submodule is used for determining a current video frame to be detected of the video data according to a preset video frame selection rule to be detected;
the video frame detection submodule is used for detecting whether the current video frame to be detected contains the target main body or not through a preset algorithm;
the device further comprises:
and the second structure analysis module is used for starting or maintaining video structural analysis on each video frame between the current video frame to be detected and the next video frame to be detected in the video data if the current video frame to be detected contains the target main body.
Optionally, the video structural analysis apparatus according to the embodiment of the present invention further includes:
and the analysis closing module is used for closing or not opening the video structural analysis of the current video segment in the video data if the current video frame to be detected does not contain the target main body, wherein the current video segment comprises the current video frame to be detected and video frames from the current video frame to be detected to the next video frame to be detected.
Optionally, the video frame detection sub-module includes:
a foreground information extraction unit, configured to extract foreground information of the current video frame to be detected by using a preset background modeling method;
the first judging unit is used for judging that the current video frame to be detected comprises the target main body if the foreground information accords with a preset foreground rule;
and the second judging unit is used for judging that the current video frame to be detected does not contain the target main body if the foreground information does not accord with the preset foreground rule.
Optionally, the video frame detection sub-module is specifically configured to:
and detecting whether the current video frame to be detected of the video data contains the target main body or not by a preset target detection method.
In a fourth aspect, an embodiment of the present invention provides a video structural analysis device, where the device includes:
the video segment splitting module is used for acquiring video data to be analyzed and splitting the video data into a plurality of video segments according to a preset splitting rule;
the video frame selection module is used for respectively selecting the video frames to be detected in each video segment according to a preset video frame selection rule;
the target main body detection module is used for respectively judging whether each video frame to be detected contains a target main body to be concerned through a preset algorithm;
the target video segment determining module is used for taking a video segment containing the video frame to be detected of the target main body as a target video segment;
and the third structure analysis module is used for carrying out video structural analysis on each target video segment.
Optionally, the target subject detecting module includes:
the foreground information extraction submodule is used for extracting the foreground information of the video frame to be detected by a preset background modeling method aiming at each frame of the video frame to be detected;
the first judgment submodule is used for judging that the video frame to be detected contains the target main body if the foreground information accords with a preset foreground rule;
and the second judging submodule is used for judging that the video frame to be detected does not contain the target main body if the foreground information does not accord with the preset foreground rule.
Optionally, the target subject detection module is specifically configured to:
and respectively detecting whether each video frame to be detected contains a target subject to be paid attention or not by a preset target detection method.
In a fifth aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory, where the memory is used for storing a computer program; the processor is configured to implement the video structural analysis method according to any one of the first aspect or the video structural analysis method according to any one of the second aspect when executing the program stored in the memory.
In a sixth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the video structural analysis method according to any one of the above first aspects or the video structural analysis method according to any one of the above second aspects.
The video structured analysis method and the video structured analysis device provided by the embodiment of the invention are used for acquiring video data to be analyzed; determining whether a video frame to be detected in video data contains a concerned target main body or not through a preset algorithm; and if the video frame to be detected contains the target main body, performing video structural analysis on the video frame to be detected. The video structural analysis is carried out on the video frame to be detected containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved. Of course, it is not necessary for any product or method to achieve all of the above-described advantages at the same time for practicing the invention.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a video structural analysis method according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a video frame in video data according to an embodiment of the present invention;
FIG. 3 is a second flowchart of a video structuring analysis method according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart of a third method for structural analysis of a video according to an embodiment of the present invention;
FIG. 5 is a first schematic diagram of a video structuring analysis device according to an embodiment of the present invention;
FIG. 6 is a second schematic diagram of a video structuring analysis device according to an embodiment of the present invention;
FIG. 7 is a schematic view of the workflow of the video structuring analysis module according to an embodiment of the present invention;
FIG. 8 is a third schematic diagram of a video structuring analysis device according to an embodiment of the present invention;
fig. 9 is a schematic diagram of an electronic device according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The video structuralization is a technology for extracting video content information, and organizes the video content into text information which can be understood by a computer and people according to semantic relation by adopting processing means such as space-time segmentation, feature extraction, object identification and the like. In the prior art, each video frame in a video to be analyzed is subjected to video structural analysis frame by frame, and the efficiency of the video structural analysis is low.
In view of the above, an embodiment of the present invention provides a video structural analysis method, and referring to fig. 1, the method includes:
s101, video data to be analyzed are obtained.
The video structural analysis method in the embodiment of the invention can be realized by a rapid analysis system, and the rapid analysis system is any system capable of realizing the video structural analysis method in the embodiment of the invention. For example:
the rapid analysis system may be an apparatus comprising: a processor, a memory, a communication interface, and a bus; the processor, the memory and the communication interface are connected through a bus and complete mutual communication; the memory stores executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory for executing the video structured analysis method of the embodiment of the present invention.
The rapid analysis system may also be an application program for executing the video structural analysis method of the embodiment of the present invention at runtime.
The fast analysis system may also be a storage medium for storing executable code for performing the video structured analysis method of the embodiment of the present invention.
The rapid analysis system obtains video data to be analyzed, and can read the video data to be analyzed in the database so as to realize video structural analysis on historical video data. The rapid analysis system obtains video data to be analyzed and can also receive the video data sent by video acquisition equipment such as a camera in real time so as to realize video structured analysis on the video data acquired in real time. The method is suitable for various application scenes, so that various user requirements are met.
S102, aiming at a video frame to be detected in the video data, determining whether the video frame to be detected contains a target subject to be paid attention or not through a preset algorithm.
The target subject is a target that the user is interested in and wishes to acquire relevant data, such as a vehicle on a road, a passenger in an elevator, a visitor in a scenic spot, an animal in a protected area, or the like. The preset algorithm is any method capable of identifying whether the video frame contains the target subject, such as a background modeling method or a target detection method. The rapid analysis system detects whether the video frame to be detected contains the target main body or not through a preset algorithm.
And S103, if the video frame to be detected contains the target main body, performing video structuralization analysis on the video frame to be detected.
The rapid analysis system carries out video structural analysis on the video frame to be detected containing the target main body, and does not carry out video structural analysis on the video frame to be detected not containing the target main body.
Performing video structuring analysis on video frames of video data includes: extracting a background image of the video, extracting a target image and a mask image of each frame of the moving target, and generating a video structured target description file. The video structured object description file may include: the target track information comprises target track information, target attribute information and event information, wherein the target track information can comprise an ID number of each target track, the number of track points, a frame number of each track point, time information, space information, other track ID numbers which are overlapped with the target track information, and the like, and the target attribute information can comprise all attribute information of the target, such as target appearance time, target movement speed, target movement direction, license plate number, vehicle type, vehicle brand, vehicle color, wearing color of people, age, height, wearing glasses, carrying bags and the like.
In the embodiment of the invention, the video structural analysis is carried out on the video frame to be detected containing the target main body, and the video structural analysis is not carried out on the video frame to be detected not containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
Optionally, the determining, by using a preset algorithm, whether the video frame to be detected in the video data includes a target subject to be focused includes:
and determining the current video frame to be detected of the video data according to a preset video frame to be detected selection rule.
The preset video frame selection rule to be detected is any video frame selection rule, for example, one frame of video frame is selected as a video frame to be detected every N (N is a non-negative integer) frames of video frames according to a time sequence order.
Optionally, the preset selection rule is as follows: and respectively selecting each frame of video frame in the video data as a video frame to be detected according to the time sequence. Each frame of video frame in the video data is used as a video frame to be detected, so that the speed of performing video structural analysis on the video data can be increased under the condition that a target main body is not lost.
And detecting whether the current video frame to be detected contains the target main body or not through a preset algorithm.
The rapid analysis system detects whether the current video frame to be detected contains a target main body or not through a preset algorithm, such as a background modeling method or a target detection method.
Correspondingly, after performing video structuralization analysis on the video frame to be detected if the video frame to be detected contains the target main body, the method further includes:
and if the current video frame to be detected contains the target main body, starting or keeping video structural analysis on each video frame between the current video frame to be detected and the next video frame to be detected in the video data.
In order to accelerate the video structural analysis, the detection of the target subject and the video structural analysis can be carried out simultaneously. And when detecting that the current video frame to be detected contains the target main body, starting or maintaining video structured analysis on the corresponding video segment in the video data according to the time sequence. For example, according to the time sequence, when the current video frame to be detected is the first video frame to be detected, or the last video frame to be detected of the current video frame to be detected does not contain the target main body, starting video structured analysis on the video data; and when the last video frame to be detected of the current video frames to be detected contains the target main body, keeping the video structural analysis of the video data.
For example, as shown in fig. 2, the current video frame to be detected is an L-th frame video frame, and the next video frame to be detected is an S-th frame video frame, then the set of video frames in the corresponding video segment in the video data can be represented as [ L +1, S-1]. Because the rapid analysis system also performs video structural analysis on the video frame to be detected including the target subject, when the current video frame to be detected, i.e. the lth frame video frame, includes the target subject, the video frame set which needs to perform the video structural analysis can be represented as [ L, S-1].
And selecting the next video frame to be detected of the current video frame to be detected according to a preset video frame selection rule to be detected. For example, when the preset selection rule is that one video frame is selected as a video frame to be detected every N frames of video frames according to a time sequence order, if the current video frame to be detected is a kth frame of video, the next video frame to be detected of the current video frame to be detected is a kth + N +1 frame of video; correspondingly, if the current video frame to be detected is the Kth frame video frame, the last video frame to be detected of the current video frame to be detected is the Kth-N-1 frame video frame.
In the embodiment of the invention, whether the video frame to be detected contains the target main body or not can be detected, and meanwhile, the corresponding video segment can be subjected to video structural analysis, so that the speed of the video structural analysis can be further increased.
Optionally, after the detecting, by the preset algorithm, whether the current video frame to be detected includes the target subject, the method further includes:
and if the current video frame to be detected does not contain the target main body, closing or not opening video structural analysis of a current video segment in the video data, wherein the current video segment comprises the current video frame to be detected and video frames from the current video frame to be detected to the next video frame to be detected.
For example, when a previous video frame to be detected of the current video frame to be detected does not contain a target main body, the video structuralization analysis on the corresponding video segment in the video data is not started; or when the last video frame to be detected of the current video frame to be detected contains the target main body, closing the video structural analysis of the corresponding video segment in the video data.
In the embodiment of the invention, the video structuralization analysis can be carried out on the corresponding target video segment while detecting whether the target main body is contained in the video frame to be detected, so that the speed of the video structuralization analysis can be further increased.
Optionally, the detecting, by using a preset algorithm, whether the current video frame to be detected includes the target main body includes:
step one, extracting foreground information of the current video frame to be detected through a preset background modeling method.
The preset background modeling method is any background modeling method, such as a color background model method, an average background model method, a gaussian background model method, or a CodeBook background model method.
And step two, if the foreground information accords with a preset foreground rule, judging that the current video frame to be detected comprises the target main body.
The preset foreground rule is a rule for arbitrarily determining whether the target subject is included, for example, if the foreground information includes a stable foreground block (for example, if the shape or size of the foreground block meets a preset standard, the foreground block is a stable foreground block), it is determined that the target subject is included in the current video frame to be detected. In an actual application environment, for example, for video data acquired by a camera on a highway, a road is determined as a background by a preset background modeling method, and a stable foreground block is a running vehicle.
And step three, if the foreground information does not accord with the preset foreground rule, judging that the current video frame to be detected does not contain the target main body.
For example, if the foreground information does not include a stable foreground block, it is determined that the current video frame to be detected does not include the target subject.
In the embodiment of the invention, whether the current video frame to be detected contains the target main body is detected by the background modeling method, and the detection speed of the target main body is high by adopting the background modeling method, so that the speed of video structural analysis is increased.
Optionally, the detecting, by using a preset algorithm, whether the current video frame to be detected includes the target subject includes:
and detecting whether the current video frame to be detected of the video data contains the target main body or not by a preset target detection method.
Algorithms for arbitrarily detecting a target subject by a preset target detection method, for example, boosting, RCNN (Regions relational Neural Network), FRCNN (Fast regional Neural Network), fasternn (Fast regional Neural Network), SSD (Single Shot multi box Detector), and the like.
For example, the target subject is a pedestrian, whether the current video frame to be detected contains the pedestrian is detected according to a preset target detection method, if yes, the target subject is contained, and if not, the target subject is not contained.
In the embodiment of the invention, whether the video frame to be detected contains the target subject or not is detected by the target detection method, the target subject is accurately detected, the situation that the video frame not containing the target subject is judged to contain the target subject is reduced, and the speed of video structuralization analysis is increased.
Optionally, the determining, by using a preset algorithm, whether the video frame to be detected in the video data includes a target subject to be focused includes:
firstly, splitting the video data into a plurality of video segments according to a preset splitting rule.
The video data is divided into a plurality of video segments by a preset dividing method, for example, every N +1 frame of video frames of the video data is divided into one target video segment in time sequence order from the first frame of video frames of the video data.
And step two, respectively selecting the video frames to be detected in each video segment according to a preset video frame selection rule.
The preset video frame selection rule is a rule for randomly selecting a video frame to be detected from video segments, for example, a first frame video frame of each video segment is selected as the video frame to be detected, or M (M is a positive integer and M is not greater than N + 1) frame video frames are randomly selected as the video frame to be detected in each video segment.
And thirdly, respectively judging whether each frame of the video frame to be detected contains the target main body or not through the preset algorithm.
Correspondingly, the video structural analysis method of the embodiment of the invention further comprises the following steps:
taking a video segment containing the video frame to be detected of the target main body as a target video segment;
and performing video structural analysis on each target video segment.
The target video segment may be a frame of video frame or a collection of frames of video. And performing video structuralization analysis on the target video segment, and not performing video structuralization analysis on other video segments except the target video segment in the video data.
Optionally, in the embodiment of the present invention, each frame of the video data may be divided into one video segment. The method comprises the steps of detecting whether each frame of video data contains a target body, extracting all video frames containing the target body, and performing video structural analysis. The method can accelerate the speed of performing video structural analysis on the video data under the condition of not losing the target subject.
In the actual calculation process, the speed of detecting whether the video frame contains the target main body is higher than the speed of performing video structural analysis on the video frame.
Fig. 3 is another video structural analysis method according to an embodiment of the present invention, where the method includes:
s301, acquiring video data to be detected;
s302, selecting a current video frame to be detected of the video data according to a preset selection rule;
the preset selection rule is any video frame selection rule, for example, one frame of video frame is selected as the video frame to be detected every N (N is a non-negative integer) frames of video frames according to a time sequence order.
S303, detecting whether a current video frame to be detected of the video data contains the target main body or not through a preset algorithm, and if the current video frame to be detected contains the target main body, executing S304; if the video frame to be detected does not include the target subject, S302 is executed.
The preset algorithm is any method capable of identifying whether the video frame contains the target subject, such as a background modeling method or a target detection method.
Because whether the video frame to be detected currently contains the target main body or not is detected, the video frame to be detected currently does not contain the target main body, and when the step S302 is executed, the selected video frame is the video frame to be detected next to the detected video frame to be detected currently. For example, the current video frame to be detected is the ith (i is a positive integer) frame video frame to be detected, and when the current video frame to be detected does not contain the target subject, the current video frame to be detected is the (i + 1) th frame video frame to be detected selected when the step S302 is executed again.
S304, performing video structural analysis on the corresponding video segments in the video data, and outputting structural data, wherein the corresponding video segments in the video data comprise the current video frame to be detected and video frames from the current video frame to be detected to the next video frame to be detected.
In the embodiment of the invention, the video structural analysis is only carried out on the video segment containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
In a further embodiment of the present invention, there is provided a video structured analysis method, see fig. 4, the method comprising:
s401, video data to be analyzed are obtained, and the video data are split into a plurality of video segments according to a preset splitting rule.
The video structural analysis method in the embodiment of the invention can be realized by a rapid analysis system, and the rapid analysis system is any system capable of realizing the video structural analysis method in the embodiment of the invention. For example:
the rapid analysis system may be an apparatus comprising: a processor, a memory, a communication interface, and a bus; the processor, the memory and the communication interface are connected through a bus and complete mutual communication; the memory stores executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory for executing the video structured analysis method of the embodiment of the present invention.
The rapid analysis system may also be an application program for executing the video structural analysis method of the embodiment of the present invention at runtime.
The fast analysis system may also be a storage medium for storing executable code for performing the video structured analysis method of the embodiment of the present invention.
The preset splitting rule is a rule for splitting the video data into video segments at will, for example, according to a time sequence, starting from a first frame of video frame of the video data, dividing every N +1 frames of video frame into a target video segment; or dividing the video frame in each unit into a video segment and the like according to the shooting time and the preset unit time as a dividing unit. Each video segment may include multiple video frames or only one video frame, and the actual number of video frames in each video segment is determined according to a preset splitting rule.
Optionally, in the embodiment of the present invention, each frame of the video data may be divided into one video segment. The method comprises the steps of detecting whether each frame of video data contains a target body, extracting all video frames containing the target body, and performing video structural analysis. The method can accelerate the speed of performing video structural analysis on the video data under the condition of not losing the target subject.
S402, respectively selecting the video frame to be detected in each video segment according to a preset video frame selection rule.
The preset video frame selection rule is a rule for selecting a video frame to be detected from video segments, for example, a first frame video frame of each video segment is selected as a video frame to be detected, or M (M is a positive integer and M is not greater than N + 1) frames are randomly selected in each video segment as video frames to be detected.
And S403, respectively judging whether each video frame to be detected contains the concerned target main body through a preset algorithm.
The rapid analysis system respectively judges whether each video frame to be detected contains a target main body through a background modeling method, a target detection method and the like.
S404, using the video segment containing the video frame to be detected of the target subject as the target video segment.
The rapid analysis system takes the video segment of the video frame to be detected containing the target main body as the target video segment.
And S405, performing video structuralization analysis on each target video segment.
The rapid analysis system carries out video structural analysis on the target video segment and does not carry out video structural analysis on other video segments except the target video segment in the video data.
In the actual calculation process, the speed of detecting whether the video frame contains the target main body is higher than the speed of performing video structural analysis on the video frame.
Optionally, the determining, by using a preset algorithm, whether each of the video frames to be detected includes a target subject to be paid attention to includes:
step one, aiming at each frame of the video frame to be detected, extracting foreground information of the video frame to be detected through a preset background modeling method.
The preset background modeling method is any background modeling method, for example, a color background model method, an average background model method, a gaussian background model method, or a CodeBook background model method.
And step two, if the foreground information accords with a preset foreground rule, judging that the video frame to be detected comprises the target main body.
The preset foreground rule is a rule for arbitrarily determining whether the target subject is included, for example, if the foreground information includes a stable foreground block (for example, if the shape or size of the foreground block meets a preset standard, the foreground block is a stable foreground block), it is determined that the target subject is included in the current video frame to be detected. In an actual application environment, for example, for video data acquired by a camera on a highway, a road is determined as a background by a preset background modeling method, and a stable foreground block is a running vehicle.
And step three, if the foreground information does not accord with the preset foreground rule, judging that the video frame to be detected does not contain the target main body.
For example, if the foreground information does not include a stable foreground block, it is determined that the current video frame to be detected does not include the target subject.
In the embodiment of the invention, whether the current video frame to be detected contains the target main body is detected by the background modeling method, and the detection speed of the target main body is high by adopting the background modeling method, so that the speed of video structural analysis is increased.
Optionally, the determining, by using a preset algorithm, whether each of the video frames to be detected includes a target subject to be paid attention to includes:
and respectively detecting whether each video frame to be detected contains the concerned target main body or not by a preset target detection method.
The target detection method is an algorithm for arbitrarily detecting a target subject by a preset target detection method, for example, boosting, RCNN, FRCNN, fastercnnn, SSD, and the like.
For example, the target subject is a pedestrian, whether the current video frame to be detected contains the pedestrian is detected according to a preset target detection method, if yes, the target subject is contained, and if not, the target subject is not contained.
In the embodiment of the invention, whether the video frame to be detected contains the target main body or not is detected by the target detection method, so that the target main body is accurately detected, the condition that the video frame not containing the target main body is judged to contain the target main body is reduced, and the speed of video structural analysis is increased.
An embodiment of the present invention further provides a video structured analysis apparatus, as shown in fig. 5, including:
a video data obtaining module 501, configured to obtain video data to be analyzed;
a target subject determination module 502, configured to determine, by using a preset algorithm, whether a video frame to be detected in the video data includes a target subject to be focused;
the first structure analysis module 503 is configured to perform video structural analysis on the video frame to be detected if the video frame to be detected includes the target subject.
In the embodiment of the invention, the video structural analysis is carried out on the video frame to be detected containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
Optionally, in the video structural analysis apparatus according to the embodiment of the present invention, the target subject determination module 502 includes:
the video frame determination submodule is used for determining the current video frame to be detected of the video data according to a preset video frame selection rule to be detected;
a video frame detection submodule, configured to detect whether the current video frame to be detected includes the target main body through a preset algorithm;
the above-mentioned device still includes:
and the second structure analysis module is used for starting or maintaining video structural analysis on each video frame between the current video frame to be detected and the next video frame to be detected in the video data if the current video frame to be detected contains the target main body.
Correspondingly, the video structural analysis device of the embodiment of the invention further comprises:
and an analysis closing module, configured to close or not open video structuralization analysis on a current video segment in the video data if the current video frame to be detected does not include the target main body, where the current video segment includes the current video frame to be detected and video frames from the current video frame to be detected to the next video frame to be detected.
In the embodiment of the invention, the video structuralization analysis can be carried out on the corresponding target video segment while detecting whether the target main body is contained in the video frame to be detected, so that the speed of the video structuralization analysis can be further increased.
Optionally, in the video structural analysis device according to the embodiment of the present invention, the video frame detection sub-module includes:
a foreground information extraction unit, configured to extract foreground information of the current video frame to be detected by using a preset background modeling method;
the first judging unit is used for judging that the current video frame to be detected comprises the target main body if the foreground information accords with a preset foreground rule;
and the second judging unit is used for judging that the current video frame to be detected does not contain the target main body if the foreground information does not accord with the preset foreground rule.
In the embodiment of the invention, whether the current video frame to be detected contains the target main body is detected by the background modeling method, and the detection speed of the background modeling method is high, so that the speed of video structural analysis is increased.
Optionally, in the video structural analysis device according to the embodiment of the present invention, the video frame detection sub-module is specifically configured to:
and detecting whether the current video frame to be detected of the video data contains the target main body or not by a preset target detection method.
In the embodiment of the invention, whether the video frame to be detected contains the target main body or not is detected by the target detection method, so that the target main body is accurately detected, the condition that the video frame not containing the target main body is judged to contain the target main body is reduced, and the speed of video structural analysis is increased.
Optionally, in the video structural analysis apparatus according to the embodiment of the present invention, the target subject determination module 502 includes:
the video splitting submodule is used for splitting the video data into a plurality of video segments according to a preset splitting rule;
the video frame selection submodule is used for respectively selecting the video frame to be detected in each video segment according to a preset video frame selection rule;
the video frame judgment submodule is used for respectively judging whether each video frame to be detected contains the target main body or not through a preset algorithm;
correspondingly, the video structural analysis device of the embodiment of the invention further comprises:
the target video segment determining module is used for taking a video segment containing the video frame to be detected of the target main body as a target video segment;
and the third structural analysis module is used for carrying out video structural analysis on each target video segment.
In the embodiment of the invention, the video structural analysis is only carried out on the target video segment containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
Referring to fig. 6, fig. 6 is another schematic diagram of a video structural analysis apparatus according to an embodiment of the present invention, including:
the video analysis module 601 is configured to acquire a video frame to be detected in video data to be detected, and detect whether a current video to be detected contains a target main body through a preset algorithm; when the current video frame to be detected contains the target main body, starting the video structure analysis module 602 or keeping the video structure analysis module 602 working; when the video frame to be detected does not include the target subject, the video structure analysis module 602 is turned off or not turned on.
And the video structural analysis module 602 is configured to perform video structural analysis on the video data to be detected to obtain a target description file.
In order to accelerate the video structural analysis, the detection of the target subject and the video structural analysis can be performed simultaneously. The video analysis module 601 detects whether a video frame to be detected contains a target body, and simultaneously controls the working state of the video structured analysis module 602, if the video frame to be detected contains the target body, the video structured analysis module 602 is controlled to be in the working state, and if the video frame to be detected does not contain the target body, the video structured analysis module 602 is controlled to be in the closed state.
As shown in fig. 7, the video structured analysis module 602 is configured to determine an object description file of the object subject, where the object description file includes: a target track description file and a target attribute description file. Specifically, a target track description file of the target subject is determined through a preset target extraction technology, and a target attribute description file of the target subject is determined through a target attribute extraction technology. The target track description file may include an ID number of each target subject track, the number of track points, a frame number of each track point, time information, space information, and other track ID numbers overlapped therewith, and the target attribute description file may include all attribute information of the target subject, such as target subject appearance time, target subject movement speed, target subject movement direction, license plate number, vehicle type, vehicle brand, vehicle color, wearing color of a person, age, height, wearing glasses, wearing backpack or not, and the like.
In the embodiment of the invention, the video structural analysis is carried out on the target video segment containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
An embodiment of the present invention further provides a video structured analysis apparatus, as shown in fig. 8, including:
a video segment splitting module 801, configured to acquire video data to be analyzed, and split the video data into multiple video segments according to a preset splitting rule;
a video frame selection module 802, configured to select a video frame to be detected in each video segment according to a preset video frame selection rule;
a target subject detection module 803, configured to respectively determine whether each to-be-detected video frame includes a target subject to be paid attention to through a preset algorithm;
a target video segment determining module 804, configured to use a video segment in which a video frame to be detected including the target subject is located as a target video segment;
the third structure analysis module 805 is configured to perform video structural analysis on each of the target video segments.
In the actual calculation process, the speed of detecting whether the video frame contains the target main body is higher than the speed of performing video structural analysis on the video frame.
Optionally, the target subject detecting module 803 includes:
the foreground information extraction submodule is used for extracting the foreground information of the video frame to be detected by a preset background modeling method aiming at each frame of the video frame to be detected;
the first judgment submodule is used for judging that the video frame to be detected contains the target main body if the foreground information accords with a preset foreground rule;
and the second judging submodule is used for judging that the video frame to be detected does not contain the target main body if the foreground information does not accord with the preset foreground rule.
In the embodiment of the invention, whether the current video frame to be detected contains the target main body is detected by the background modeling method, and the detection speed of the target main body is high by adopting the background modeling method, so that the speed of video structural analysis is increased.
Optionally, the target subject detection module 803 is specifically configured to:
and respectively detecting whether each video frame to be detected contains the concerned target main body or not by a preset target detection method.
In the embodiment of the invention, whether the video frame to be detected contains the target main body or not is detected by the target detection method, so that the target main body is accurately detected, the condition that the video frame not containing the target main body is judged to contain the target main body is reduced, and the speed of video structural analysis is increased.
The embodiment of the invention also provides electronic equipment, which comprises a processor and a memory, wherein the memory is used for storing the computer program; the processor is configured to implement the following steps when executing the program stored in the memory:
acquiring video data to be analyzed;
determining whether the video frame to be detected contains a target subject to be paid attention or not through a preset algorithm aiming at the video frame to be detected in the video data;
and if the video frame to be detected contains the target main body, performing video structuralization analysis on the video frame to be detected.
Or the processor is configured to implement the following steps when executing the program stored in the memory:
acquiring video data to be analyzed, and splitting the video data into a plurality of video segments according to a preset splitting rule;
respectively selecting a video frame to be detected in each video segment according to a preset video frame selection rule;
respectively judging whether each video frame to be detected contains a concerned target main body or not through a preset algorithm;
taking a video segment containing the video frame to be detected of the target main body as a target video segment;
and performing video structural analysis on each target video segment.
In the embodiment of the invention, the video structural analysis is carried out on the target video segment containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
Optionally, the processor may further implement any of the video structured analysis methods when the processor is configured to execute the program stored in the memory.
Optionally, as shown in fig. 9, the electronic device according to the embodiment of the present invention includes a processor 901, a communication interface 902, a memory 903, and a communication bus 904, where the processor 901, the communication interface 902, and the memory 903 complete mutual communication through the communication bus 904,
a memory 903 for storing computer programs;
the processor 901 is configured to implement the following steps when executing the program stored in the memory 903:
acquiring video data to be analyzed;
determining whether the video frame to be detected contains a target subject to be paid attention or not through a preset algorithm aiming at the video frame to be detected in the video data;
and if the video frame to be detected contains the target main body, performing video structuralization analysis on the video frame to be detected.
In the embodiment of the invention, the video structural analysis is carried out on the video frame to be detected containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
Or the processor 901, configured to implement the following steps when executing the program stored in the memory 903:
acquiring video data to be analyzed, and splitting the video data into a plurality of video segments according to a preset splitting rule;
respectively selecting a video frame to be detected in each video segment according to a preset video frame selection rule;
respectively judging whether each video frame to be detected contains a concerned target main body or not through a preset algorithm;
taking a video segment containing the video frame to be detected of the target main body as a target video segment;
and performing video structural analysis on each target video segment.
The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Alternatively, the memory may be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements the following steps:
acquiring video data to be analyzed;
determining whether a video frame to be detected in the video data contains a concerned target main body or not through a preset algorithm;
and if the video frame to be detected contains the target main body, performing video structuralization analysis on the video frame to be detected.
Or the computer program when executed by a processor implements the steps of:
acquiring video data to be analyzed, and splitting the video data into a plurality of video segments according to a preset splitting rule;
respectively selecting a video frame to be detected in each video segment according to a preset video frame selection rule;
respectively judging whether each video frame to be detected contains a concerned target main body or not through a preset algorithm;
taking a video segment containing the video frame to be detected of the target main body as a target video segment;
and performing video structural analysis on each target video segment.
In the embodiment of the invention, the video structural analysis is carried out on the video frame to be detected containing the target main body, so that the speed of carrying out the video structural analysis on the video data can be increased, and the efficiency of the video structural analysis is improved.
Optionally, the computer program stored in the computer-readable storage medium may further implement any of the video structural analysis methods described above when the computer program is executed by a processor.
It should be noted that, in this document, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional like elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the video structural analysis apparatus, the electronic device and the storage medium, since they are substantially similar to the method embodiments, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiments.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (18)
1. A method for structured analysis of video, the method comprising:
acquiring video data to be analyzed;
determining whether a video frame to be detected in the video data contains a target subject to be concerned or not through a preset algorithm;
if the video frame to be detected comprises the target main body, performing video structured analysis on the video frame to be detected to generate a video structured target description file, wherein the video structured target description file comprises: target track information, target attribute information, and event information;
the video frame to be detected is obtained by selecting according to a preset video frame to be detected selection rule, wherein the video frame to be detected selection rule is as follows: and respectively selecting each frame of video frame in the video data as a video frame to be detected according to a time sequence.
2. The method according to claim 1, wherein the determining, by a preset algorithm, whether the video frame to be detected in the video data contains a target subject of interest comprises:
determining a current video frame to be detected of the video data according to a preset video frame to be detected selection rule;
detecting whether the current video frame to be detected contains the target main body or not through a preset algorithm;
if the video frame to be detected contains the target main body, after performing video structural analysis on the video frame to be detected, the method further comprises the following steps:
and if the current video frame to be detected contains the target main body, starting or keeping video structural analysis on each video frame between the current video frame to be detected and the next video frame to be detected in the video data.
3. The method according to claim 2, wherein after said detecting whether the current video frame to be detected contains the target subject through the preset algorithm, the method further comprises:
and if the current video frame to be detected does not contain the target main body, closing or not opening video structural analysis of a current video segment in the video data, wherein the current video segment comprises the current video frame to be detected and video frames between the current video frame to be detected and the next video frame to be detected.
4. The method according to claim 2, wherein the detecting whether the current video frame to be detected includes the target subject through a preset algorithm comprises:
extracting foreground information of the current video frame to be detected by a preset background modeling method;
if the foreground information accords with a preset foreground rule, judging that the current video frame to be detected comprises the target main body;
and if the foreground information does not accord with the preset foreground rule, judging that the current video frame to be detected does not contain the target main body.
5. The method according to claim 2, wherein the detecting whether the current video frame to be detected includes the target subject through a preset algorithm comprises:
and detecting whether the current video frame to be detected of the video data contains the target main body or not by a preset target detection method.
6. A method for structured analysis of video, the method comprising:
the method comprises the steps of obtaining video data to be analyzed, and splitting the video data into a plurality of video segments according to a preset splitting rule;
respectively selecting a video frame to be detected in each video segment according to a preset video frame selection rule;
respectively judging whether each video frame to be detected contains a target subject to be paid attention to or not through a preset algorithm;
taking a video segment containing the video frame to be detected of the target main body as a target video segment;
performing video structural analysis on each target video segment to generate a video structural target description file, wherein the video structural target description file comprises: target track information, target attribute information and event information;
the video frame selection rule is as follows: and respectively selecting each frame of video frame in the video data as a video frame to be detected according to a time sequence.
7. The method according to claim 6, wherein the step of respectively determining whether each video frame to be detected contains a target subject of interest by using a preset algorithm comprises:
extracting foreground information of each video frame to be detected by a preset background modeling method aiming at each frame of the video frame to be detected;
if the foreground information accords with a preset foreground rule, judging that the video frame to be detected comprises the target main body;
and if the foreground information does not accord with the preset foreground rule, judging that the video frame to be detected does not contain the target main body.
8. The method according to claim 6, wherein the step of respectively determining whether each video frame to be detected contains a target subject of interest by using a preset algorithm comprises:
and respectively detecting whether each video frame to be detected contains a target subject to be paid attention or not by a preset target detection method.
9. A video structured analysis apparatus, the apparatus comprising:
the video data acquisition module is used for acquiring video data to be analyzed;
the target main body judging module is used for determining whether the video frame to be detected contains a concerned target main body or not through a preset algorithm aiming at the video frame to be detected in the video data;
a first structure analysis module, configured to, if the video frame to be detected includes the target main body, perform video structural analysis on the video frame to be detected, and generate a video structural target description file, where the video structural target description file includes: target track information, target attribute information, and event information;
the video frame to be detected is obtained by selecting according to a preset video frame to be detected selection rule, wherein the video frame to be detected selection rule is as follows: and respectively selecting each frame of video frame in the video data as a video frame to be detected according to a time sequence.
10. The apparatus of claim 9, wherein the target subject determination module comprises:
the video frame determining submodule is used for determining a current video frame to be detected of the video data according to a preset video frame selection rule to be detected;
the video frame detection submodule is used for detecting whether the current video frame to be detected contains the target main body or not through a preset algorithm;
the device further comprises:
and the second structure analysis module is used for starting or maintaining video structural analysis on each video frame between the current video frame to be detected and the next video frame to be detected in the video data if the current video frame to be detected contains the target main body.
11. The apparatus of claim 10, further comprising:
and the analysis closing module is used for closing or not opening the video structural analysis of the current video segment in the video data if the current video frame to be detected does not contain the target main body, wherein the current video segment comprises the current video frame to be detected and video frames from the current video frame to be detected to the next video frame to be detected.
12. The apparatus of claim 10, wherein the video frame detection sub-module comprises:
the foreground information extraction unit is used for extracting foreground information of the current video frame to be detected through a preset background modeling method;
the first judging unit is used for judging that the current video frame to be detected comprises the target main body if the foreground information accords with a preset foreground rule;
and the second judging unit is used for judging that the current video frame to be detected does not contain the target main body if the foreground information does not accord with the preset foreground rule.
13. The apparatus of claim 10, wherein the video frame detection sub-module is specifically configured to:
and detecting whether the current video frame to be detected of the video data contains the target main body or not by a preset target detection method.
14. A video structured analysis apparatus, the apparatus comprising:
the video segment splitting module is used for acquiring video data to be analyzed and splitting the video data into a plurality of video segments according to a preset splitting rule;
the video frame selection module is used for respectively selecting the video frame to be detected in each video segment according to a preset video frame selection rule;
the target main body detection module is used for respectively judging whether each video frame to be detected contains a target main body to be concerned through a preset algorithm;
the target video segment determining module is used for taking a video segment containing the video frame to be detected of the target main body as a target video segment;
a third structural analysis module, configured to perform video structural analysis on each target video segment to generate a video structural target description file, where the video structural target description file includes: target track information, target attribute information and event information;
the video frame selection rule is as follows: and respectively selecting each frame of video frame in the video data as a video frame to be detected according to a time sequence.
15. The apparatus of claim 14, wherein the target subject detection module comprises:
the foreground information extraction submodule is used for extracting the foreground information of the video frame to be detected by a preset background modeling method aiming at each frame of the video frame to be detected;
the first judgment submodule is used for judging that the video frame to be detected contains the target main body if the foreground information accords with a preset foreground rule;
and the second judging submodule is used for judging that the video frame to be detected does not contain the target main body if the foreground information does not accord with the preset foreground rule.
16. The apparatus of claim 14, wherein the target subject detection module is specifically configured to:
and respectively detecting whether each video frame to be detected contains a target subject to be paid attention or not by a preset target detection method.
17. An electronic device comprising a processor and a memory, said memory for storing a computer program; the processor, when executing the program stored in the memory, is configured to implement the video structural analysis method according to any one of claims 1 to 5, or the video structural analysis method according to any one of claims 6 to 8.
18. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, implements the video structural analysis method according to any one of claims 1 to 5, or implements the video structural analysis method according to any one of claims 6 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810609988.XA CN110659384B (en) | 2018-06-13 | 2018-06-13 | Video structured analysis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810609988.XA CN110659384B (en) | 2018-06-13 | 2018-06-13 | Video structured analysis method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110659384A CN110659384A (en) | 2020-01-07 |
| CN110659384B true CN110659384B (en) | 2022-10-04 |
Family
ID=69026174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810609988.XA Active CN110659384B (en) | 2018-06-13 | 2018-06-13 | Video structured analysis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659384B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113989763B (en) * | 2021-12-30 | 2022-04-15 | 江西省云眼大视界科技有限公司 | Video structured analysis method and analysis system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107133563A (en) * | 2017-03-17 | 2017-09-05 | 深圳市能信安科技股份有限公司 | A kind of video analytic system and method based on police field |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060221181A1 (en) * | 2005-03-30 | 2006-10-05 | Cernium, Inc. | Video ghost detection by outline |
| CN103092930B (en) * | 2012-12-30 | 2017-02-08 | 贺江涛 | Method of generation of video abstract and device of generation of video abstract |
| CN104657712B (en) * | 2015-02-09 | 2017-11-14 | 惠州学院 | Masked man's detection method in a kind of monitor video |
| CN105893510A (en) * | 2016-03-30 | 2016-08-24 | 北京格灵深瞳信息技术有限公司 | Video structurization system and target search method thereof |
| CN106354816B (en) * | 2016-08-30 | 2019-12-13 | 东软集团股份有限公司 | video image processing method and device |
| CN108012156B (en) * | 2017-11-17 | 2020-09-25 | 深圳市华尊科技股份有限公司 | Video processing method and control platform |
-
2018
- 2018-06-13 CN CN201810609988.XA patent/CN110659384B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107133563A (en) * | 2017-03-17 | 2017-09-05 | 深圳市能信安科技股份有限公司 | A kind of video analytic system and method based on police field |
Non-Patent Citations (2)
| Title |
|---|
| "基于运动方向的异常行为检测";胡芝兰 等;《自动化学报》;20081115;第34卷(第11期);正文第1348-1353页 * |
| 胡芝兰 等."基于运动方向的异常行为检测".《自动化学报》.2008,第34卷(第11期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110659384A (en) | 2020-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10614310B2 (en) | Behavior recognition | |
| CN109145680B (en) | Method, device and equipment for acquiring obstacle information and computer storage medium | |
| CN111126258B (en) | Image recognition method and related device | |
| CN106952303B (en) | Vehicle distance detection method, device and system | |
| CN111857356B (en) | Method, device, equipment and storage medium for recognizing interaction gesture | |
| Hoang Ngan Le et al. | Robust hand detection and classification in vehicles and in the wild | |
| CN113095124A (en) | Face living body detection method and device and electronic equipment | |
| CN109800682B (en) | Driver attribute identification method and related product | |
| CN105512683A (en) | Target positioning method and device based on convolution neural network | |
| CN108805016B (en) | Head and shoulder area detection method and device | |
| CN110781768A (en) | Target object detection method and device, electronic device and medium | |
| CN107944382B (en) | Method for tracking target, device and electronic equipment | |
| CN109740415A (en) | Vehicle attribute recognition methods and Related product | |
| CN104615986A (en) | Method for utilizing multiple detectors to conduct pedestrian detection on video images of scene change | |
| CN110147731B (en) | Vehicle type identification method and related product | |
| CN113496208B (en) | Video scene classification method and device, storage medium and terminal | |
| CN110866428B (en) | Target tracking method, device, electronic equipment and storage medium | |
| CN111553321A (en) | Mobile vendor target detection model, detection method and management method thereof | |
| CN115830399B (en) | Classification model training method, device, equipment, storage medium and program product | |
| CN114445768A (en) | Target identification method and device, electronic equipment and storage medium | |
| CN112434566A (en) | Passenger flow statistical method and device, electronic equipment and storage medium | |
| Isa et al. | Real-time traffic sign detection and recognition using Raspberry Pi | |
| CN110414544B (en) | Target state classification method, device and system | |
| CN114743130A (en) | Multi-target pedestrian tracking method and system | |
| Kejriwal et al. | Vehicle detection and counting using deep learning basedYOLO and deep SORT algorithm for urban traffic management system |
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 |