CN106022311A - City monitoring video identification-based emergency event discovery method and system - Google Patents
City monitoring video identification-based emergency event discovery method and system Download PDFInfo
- Publication number
- CN106022311A CN106022311A CN201610420594.0A CN201610420594A CN106022311A CN 106022311 A CN106022311 A CN 106022311A CN 201610420594 A CN201610420594 A CN 201610420594A CN 106022311 A CN106022311 A CN 106022311A
- Authority
- CN
- China
- Prior art keywords
- emergency event
- supervision
- monitored object
- video
- cities
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/44—Event detection
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a city monitoring video identification-based emergency event discovery method and a city monitoring video identification-based emergency event discovery system. The city monitoring video identification-based emergency event discovery method of the invention includes the following steps that: first step, distributed monitoring cameras are utilized to obtain monitoring video signals in real time; second step, according to the number and movement modes of identified monitoring objects in the monitoring video signals, an image identification algorithm is utilized to analyze the monitoring video signals, so that the analytical information of the monitoring objects can be obtained; third step, potential city emergency events in the monitoring video signals are identified by using the analytical information of the monitoring objects and the recording time and recording geographic information of monitoring videos; and fourth step, identified potential city emergency events are outputted.
Description
Technical field
The present invention relates to video identification field, particularly relate to a kind of can find emergent sexual behavior possible in city in time
The emergency event based on supervision of the cities video identification of part finds method and system.
Background technology
The raising day by day to public place safety requirements along with government and the public, increasing monitoring camera is by portion
Affix one's name in the crowd such as such as traffic intersection, square, garden, park, mega-store, key area, city that vehicle is intensive.
The social public administration problem that monitored object aggregation phenomenon reflects also becomes increasingly conspicuous, have very important research and
Practice significance.
Nowadays, there is the monitor video of magnanimity to be produced and store all the time, and the understanding of monitor video and analysis
Depend on is accomplished manually more.The frequency occurred due to city emergency sexual behavior part is relatively low, the biggest in supervision of the cities video
Partial content break-up value is relatively low, it is not necessary to configuration monitoring personnel process.And when city emergency event occurs,
Owing to monitoring personnel's experience differs, also tend to the city that can not be recognized accurately in supervision of the cities video at the event initial stage
Emergency event.Therefore, manual analysis supervision of the cities video time and effort consuming and poor effect are utilized.When city monitors
Area distribution area increases, when monitoring region kind increases, manual analysis be more difficult to numerous from enormous amount, place,
Time compole length monitor video in identify that wherein occur answers acute events accurately, in time, efficiently.
Therefore, those skilled in the art be devoted to develop one can utilize emphasis district, computer intelligence identification city
The method of emergency event in the monitor video in territory.
Summary of the invention
Because the drawbacks described above of prior art, the technical problem to be solved is to provide a kind of utilization and calculates
The method of emergency event in the monitor video of key area, machine Intelligent Recognition city, and provide a set of corresponding intelligence to regard
Frequently resolution system, it is possible to the most timely and accurately find to answer acute events in supervision of the cities video.
For achieving the above object, the invention provides a kind of emergency event discovery side based on supervision of the cities video identification
Method, including:
First step: utilize the monitoring camera of distributed arrangement, obtains monitor video signal in real time;
Second step: for identifying quantity and the motor pattern of the monitored object obtained in monitor video signal, utilize
Monitor video signal is resolved by image recognition algorithm, to obtain the parsing information of monitored object;
Third step: according to shooting time and the shooting geography information of monitor video, utilize the parsing of monitored object
Information, is identified the potential city emergency event in monitor video;
4th step: the potential city emergency event that output identifies.
Preferably, in the second step, difference based on foreground area and background area under visual angle, the angle of depression is used
Monitored object recognizer, the moving object to occurring in monitor video is identified, and moving object is determined
For monitored object.
Preferably, in the second step, the adjacent two frame pictures in monitor video are corresponding by comparing in picture
Region, carries out quantity identification and position mark, and notable difference part is examined in video notable difference part
The moving object recorded;Diversity during wherein notable difference part is adjacent two frame pictures exceedes the portion of threshold value
Point.
Preferably, in the second step, use the neural network algorithm crossed through the artificial sample training marked, utilize
Quantitative relation between region area and the actual number of monitored object of the monitored object in monitor video signal is right
The kind of monitored object and quantity carry out Computer Automatic Recognition the most accurately.
Preferably, in third step, shooting time and shooting geography information according to monitor video form monitoring
The movement locus line of object, and the movement locus line of monitored object is contrasted with predetermined Motion mask, and
According to comparing result, the potential city emergency event in monitor video is identified.
Preferably, in the 4th step, the potential city emergency event that will identify that is with the semantic model table of specification
State and export as system.
Present invention also offers a kind of emergency event based on supervision of the cities video identification and find system, for realizing
State emergency event based on supervision of the cities video identification and find method.
Below with reference to accompanying drawing, the technique effect of design, concrete structure and the generation of the present invention is described further,
To be fully understood from the purpose of the present invention, feature and effect.
Accompanying drawing explanation
Fig. 1 is that based on supervision of the cities video identification according to the preferred embodiment of the invention emergency event finds method
Flow chart.
It should be noted that accompanying drawing is used for illustrating the present invention, and the unrestricted present invention.Note, represent the attached of structure
Figure may be not necessarily drawn to scale.Further, in accompanying drawing, same or like element indicates same or like
Label.
Detailed description of the invention
The cost run in being accomplished manually for overcoming existing Analysis of Video Monitoring for Cities to rely on is high, efficiency is low, weak effect
Etc. all various problems, the present invention provides a kind of method utilizing computer to carry out intelligently parsing.To specifically retouch below
State the preferred embodiments of the present invention.
Fig. 1 is that based on supervision of the cities video identification according to the preferred embodiment of the invention emergency event finds method
Flow chart.
As it is shown in figure 1, based on supervision of the cities video identification according to the preferred embodiment of the invention emergency event finds
Method includes:
First step S1: utilize the monitoring camera of distributed arrangement, obtains monitor video signal in real time as being
The input of system;
Such as, described monitoring camera is around the key monitoring place (such as large-scale square, sports ground etc.) in city
Arrange.
Second step S2: for identifying quantity and the motor pattern of the monitored object obtained in monitor video signal,
Utilize image recognition algorithm that monitor video signal is resolved, to obtain the parsing information of monitored object;
Specifically, in second step S2, use the difference based on foreground area Yu background area under visual angle, the angle of depression
Other monitored object recognizer, the moving object to occurring in monitor video is identified, and by moving object
It is defined as monitored object.
More specifically, in second step S2, the adjacent two frame pictures in monitor video, by comparing picture
Middle corresponding region, carries out quantity identification and position to notable difference part (such as, diversity exceedes the part of threshold value)
Put mark, and notable difference part is detected, in video, the moving object obtained.
For example, it is contemplated that be pedestrian and vehicle to monitoring moving object main in supervision of the cities video, in recognizer
In be characterized with counter-profile area, classifying rules can be set in order to distinguish two kinds of main monitored object.
For improving the accuracy identified further, it is further proposed that in second step S2, use through artificial
The neural network algorithm crossed of sample training of mark, utilize monitored object in monitor video signal region area and
Quantitative relation between the actual number of monitored object, kind and quantity to monitored object are counted the most accurately
Calculation machine identifies automatically.
Third step S3: according to shooting time and the shooting geography information of monitor video, utilize monitored object
Parsing information, is identified the potential city emergency event in monitor video;
Specifically, in third step S3, shooting time and shooting geography information according to monitor video are formed
The movement locus line of monitored object, and the movement locus line of monitored object is contrasted with predetermined Motion mask,
And according to comparing result, the potential city emergency event in monitor video is identified.
Such as, timing acquiring monitored object location point in monitored picture, and generate the movement locus line of monitored object,
And it is mated with the movement locus template pre-set.For the motion of monitored object under city emergency event
State, in the concrete example of the present invention, such as, can design 6 kinds of special Motion masks: (a) four-way is assembled
Pattern (b) four-way is scattered the unidirectional pattern of scattering of the unidirectional accumulation mode (e) of pattern (c) single-way moving pattern (d)
(f) annular movement pattern.By calculating video segment movement locus line and the matching degree of template, find current
Abnormality possible in monitor video.
4th step S4: the potential city emergency event that output identifies.
Such as, in the fourth step s 4, the potential city emergency event that will identify that is with the semantic model of specification
State and export as system.
For example, with the space time information that can directly obtain in monitor video, and by aforementioned video solution of the present invention
The monitored object real time information that analysis technical Analysis obtains, builds the semantic model of emergency event in supervision of the cities video,
Such as: emergency event situation=< event identifier code, event time, location of incident, monitored object density, monitoring
Object activity pattern >.The method that thus can use case-based reasioning, the incidence relation to emergency event situation
It is identified and evaluates such that it is able to identifying the most contingent in time from one section of real-time supervision of the cities video
City emergency event.
In the specific implementation, such as first with this monitored object automatic identification technology, to monitor video data signal
Carry out resolving and identifying quantity and the position thereof of monitored object in video each frame picture.Then monitored object motion mould
Formula matching technique, generates the movement locus line of monitored object, and coupling obtains corresponding motor pattern.Finally, profit
With emergency event modeling and inference method, the event in one section of video segment is carried out semantic modeling, and judges that it is
No for answering acute events.If identifying and obtaining answering acute events, then emergency management system sends alarm, reminds superintendent
Member takes Supervision Measures.
According to other embodiments of the invention, additionally provide a kind of emergency event based on supervision of the cities video identification and send out
Existing system, finds method for realizing above-mentioned emergency event based on supervision of the cities video identification.For instance, it is possible to
Monitoring camera carries out support network and the computer system of database function of video data transmission, it is possible to achieve on
State emergency event based on supervision of the cities video identification and find method.
Generally, the present invention at least has the advantage that
(1) time and effort consuming and the problem of poor effect are participated in for artificial in supervision of the cities video, it is proposed that a kind of
Utilize computer automatically, the method for intellectual analysis build corresponding management system, it is possible to fully meet multiple types,
The actual demand of large scale scene real-time monitoring.
(2) problem resolving difficulty for complex scene monitor video in supervision of the cities video, it is proposed that a kind of with
The monitored object that pedestrian and vehicle are main identifies the intelligently parsing technology with motor pattern automatically, it is possible to good conformity people
Group, the monitor video situation of the intensive public place, city of wagon flow activity.
(3) it is difficult to identify in time and the problem accurately expressed for emergency event in supervision of the cities video, it is proposed that
A kind of emergency event semantic modeling and inference method, it is possible to effectively understand and possible in monitoring scene answer acute events also
Make and timely responding to.
Described above illustrate and describes the preferred embodiments of the present invention, as previously mentioned, it should be understood that the present invention is not
It is confined to form disclosed herein, is not to be taken as the eliminating to other embodiments, and can be used for other groups various
Close, amendment and environment, and can be in invention contemplated scope described herein, by above-mentioned teaching or association area
Technology or knowledge are modified.And the change that those skilled in the art are carried out and change are without departing from the spirit of the present invention and model
Enclose, the most all should be in the protection domain of claims of the present invention.
Claims (7)
1. an emergency event based on supervision of the cities video identification finds method, it is characterised in that including:
First step: utilize the monitoring camera of distributed arrangement, obtains monitor video signal in real time;
Second step: for identifying quantity and the motor pattern of the monitored object obtained in monitor video signal, profit
With image recognition algorithm, monitor video signal is resolved, to obtain the parsing information of monitored object;
Third step: according to shooting time and the shooting geography information of monitor video, utilize the solution of monitored object
Analysis information, is identified the potential city emergency event in monitor video;
4th step: the potential city emergency event that output identifies.
2. emergency event based on supervision of the cities video identification as claimed in claim 1 finds method, and its feature exists
In, in the second step, use the prison based on foreground area Yu the difference of background area under visual angle, the angle of depression
Control object recognition algorithm, the moving object to occurring in monitor video is identified, and by moving object
It is defined as monitored object.
3. emergency event based on supervision of the cities video identification as claimed in claim 2 finds method, and its feature exists
In, in the second step, the adjacent two frame pictures in monitor video, by comparing corresponding district in picture
Territory, carries out quantity identification and position mark to notable difference part, and using notable difference part as video
The moving object that middle detection obtains;Diversity during wherein notable difference part is adjacent two frame pictures exceedes
The part of threshold value.
4. the emergency event based on supervision of the cities video identification as described in one of claims 1 to 3 finds method,
It is characterized in that, in the second step, use the neural network algorithm crossed through the artificial sample training marked,
Utilize between region area and the actual number of monitored object of the monitored object in monitor video signal
Quantitative relation, kind and quantity to monitored object carry out Computer Automatic Recognition the most accurately.
5. the emergency event based on supervision of the cities video identification as described in one of claims 1 to 3 finds method,
It is characterized in that, in third step, according to shooting time and the shooting geography information shape of monitor video
Become the movement locus line of monitored object, and the movement locus line of monitored object is entered with predetermined Motion mask
Row contrast, and according to comparing result, the potential city emergency event in monitor video is identified.
6. the emergency event based on supervision of the cities video identification as described in one of claims 1 to 3 finds method,
It is characterized in that, in the 4th step, the potential city emergency event that will identify that is with the semanteme of specification
Model formulation exports as system.
7. emergency event based on supervision of the cities video identification finds a system, for realize such as claim 1 to
One of 6 described emergency events based on supervision of the cities video identification find method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610420594.0A CN106022311A (en) | 2016-06-13 | 2016-06-13 | City monitoring video identification-based emergency event discovery method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610420594.0A CN106022311A (en) | 2016-06-13 | 2016-06-13 | City monitoring video identification-based emergency event discovery method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106022311A true CN106022311A (en) | 2016-10-12 |
Family
ID=57088464
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610420594.0A Pending CN106022311A (en) | 2016-06-13 | 2016-06-13 | City monitoring video identification-based emergency event discovery method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106022311A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107730530A (en) * | 2017-10-31 | 2018-02-23 | 西华大学 | A kind of remote emergency management control method based on smart city |
CN108256447A (en) * | 2017-12-29 | 2018-07-06 | 广州海昇计算机科技有限公司 | A kind of unmanned plane video analysis method based on deep neural network |
CN108921012A (en) * | 2018-05-16 | 2018-11-30 | 中国科学院计算技术研究所 | A method of utilizing artificial intelligence chip processing image/video frame |
CN109994059A (en) * | 2018-01-03 | 2019-07-09 | 李智萌 | A kind of electronic information presentation device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101017573A (en) * | 2007-02-09 | 2007-08-15 | 南京大学 | Method for detecting and identifying moving target based on video monitoring |
CN102163290A (en) * | 2011-05-16 | 2011-08-24 | 天津大学 | Method for modeling abnormal events in multi-visual angle video monitoring based on temporal-spatial correlation information |
US20130243343A1 (en) * | 2012-03-16 | 2013-09-19 | Nec (China) Co., Ltd. | Method and device for people group detection |
-
2016
- 2016-06-13 CN CN201610420594.0A patent/CN106022311A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101017573A (en) * | 2007-02-09 | 2007-08-15 | 南京大学 | Method for detecting and identifying moving target based on video monitoring |
CN102163290A (en) * | 2011-05-16 | 2011-08-24 | 天津大学 | Method for modeling abnormal events in multi-visual angle video monitoring based on temporal-spatial correlation information |
US20130243343A1 (en) * | 2012-03-16 | 2013-09-19 | Nec (China) Co., Ltd. | Method and device for people group detection |
Non-Patent Citations (1)
Title |
---|
傅慧源: ""面向人群监管的智能视频场景理解技术研究"", 《中国博士学位论文全文数据库信息科技辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107730530A (en) * | 2017-10-31 | 2018-02-23 | 西华大学 | A kind of remote emergency management control method based on smart city |
CN108256447A (en) * | 2017-12-29 | 2018-07-06 | 广州海昇计算机科技有限公司 | A kind of unmanned plane video analysis method based on deep neural network |
CN109994059A (en) * | 2018-01-03 | 2019-07-09 | 李智萌 | A kind of electronic information presentation device |
CN108921012A (en) * | 2018-05-16 | 2018-11-30 | 中国科学院计算技术研究所 | A method of utilizing artificial intelligence chip processing image/video frame |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105447458B (en) | A kind of large-scale crowd video analytic system and method | |
Qin et al. | Applying the convolutional neural network deep learning technology to behavioural recognition in intelligent video | |
CN102880692B (en) | A kind of monitor video semantic description towards retrieval and detection modeling method | |
CN100585656C (en) | An all-weather intelligent video analysis monitoring method based on a rule | |
CN101799876B (en) | Video/audio intelligent analysis management control system | |
Saxena et al. | Crowd behavior recognition for video surveillance | |
CN110738127A (en) | Helmet identification method based on unsupervised deep learning neural network algorithm | |
CN106022311A (en) | City monitoring video identification-based emergency event discovery method and system | |
CN102902960B (en) | Leave-behind object detection method based on Gaussian modelling and target contour | |
CN104700434A (en) | Crowd movement track anomaly detection method for complicated structuring scenes | |
CN113642403B (en) | Crowd abnormal intelligent safety detection system based on edge calculation | |
CN112232316B (en) | Crowd gathering detection method and device, electronic equipment and storage medium | |
Meng et al. | Video‐Based Vehicle Counting for Expressway: A Novel Approach Based on Vehicle Detection and Correlation‐Matched Tracking Using Image Data from PTZ Cameras | |
CN108256447A (en) | A kind of unmanned plane video analysis method based on deep neural network | |
CN116986246A (en) | Intelligent inspection system and method for coal conveying belt | |
CN115311735A (en) | Intelligent recognition early warning method for abnormal behaviors | |
CN116580350A (en) | Laboratory safety monitoring and early warning method and system | |
CN116993681A (en) | Substation inspection defect detection method and system | |
CN101883257A (en) | Guard detection system and detection method thereof | |
Pratama et al. | Smart video surveillance system for level crossing: a systematic literature review | |
CN115859689B (en) | Panoramic visualization digital twin application method | |
CN111931650A (en) | Target detection model construction and red light running responsibility tracing method, system, terminal and medium | |
CN115294519A (en) | Abnormal event detection and early warning method based on lightweight network | |
CN115798133A (en) | Flame alarm method, device, equipment and storage medium | |
Saluky et al. | Abandoned Object Detection Method Using Convolutional Neural Network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161012 |