CN103106250B - Video monitoring intellectual analysis search method and system thereof - Google Patents
Video monitoring intellectual analysis search method and system thereof Download PDFInfo
- Publication number
- CN103106250B CN103106250B CN201310012753.XA CN201310012753A CN103106250B CN 103106250 B CN103106250 B CN 103106250B CN 201310012753 A CN201310012753 A CN 201310012753A CN 103106250 B CN103106250 B CN 103106250B
- Authority
- CN
- China
- Prior art keywords
- video
- module
- restructuring
- moving target
- image
- 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.)
- Expired - Fee Related
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention belongs to technical field of security and protection, relate to a kind of video monitoring intellectual analysis search method and system thereof.It solve prior art and design the technical problems such as not reasonable.This method comprises the steps: A, video source modeling；B, video frequency searching: a, dynamically analyze；B, coding restructuring；C, result queue；D, behavior analysis.Dynamically analyzing module to be connected with coding recombination module, result queue module and behavior analysis module are both connected to encode on recombination module, and video source modeling module analyzes module with dynamic or coding recombination module is connected.Have an advantage in that: integrate image enhaucament, retrieval compression, intellectual analysis, summary are added up, embody advanced person, five efficient, reliable, convenient and practical big features, achieve the enhancing to video image, the retrieval compression of video file, the behavior intellectual analysis etc. of personage's vehicle, thus obtain required information quickly and accurately, the powerful effective criminal investigation of offer of solving a case can be provided for public security cadres and police and support.
Description
Technical field
The invention belongs to technical field of security and protection, relate to video monitoring, especially relate to one
Video monitoring intellectual analysis search method and system thereof.
Background technology
Along with socioeconomic further development, China puts into big for criminal investigation aspect
Amount man power and material, to science and technology want quality, want police strength, want fighting capacity have become as ought
The theme of modern public security department.Along with safe city system and society's monitoring Large scale construction
Go deep into, solve the problem that urban pavement video gathers reliably and preserves, have accumulated
A large amount of valuable first-hand on-site supervision image documents, for providing strong propping up in investigation
Support, utilizes monitor video to search the hands that clue to solve the case has become the most frequently used in criminal investigation and case detection
Section, has hit the criminal behavior of illegal one's share of expenses for a joint undertaking dramatically.But present stage safety city
City's monitoring system there is also problems:
1, in the face of adverse circumstances, monitoring system is disliked at the low illumination of night darkness, dense fog sand and dust
Bad weather or local light cross bright in the case of picture quality be substantially reduced, cause video figure
As camera acquisition of poor quality to poor image quality, upper giant-screen of poor quality, video recording money
Expecting of poor quality, policeman in charge of the case cannot therefrom obtain useful information, causes clear up a criminal case difficulty,
Inefficiency.
2, generally to there is camera position unreasonable for present stage video monitoring, and effective information is few
Etc. shortcoming, case evidence obtaining and generally require during analyzing the substantial amounts of manpower of cost, time
Between check magnanimity Video Document, seriously constrain its detection efficiency, the retrieval of Traditional Man
Evidence obtaining work there is also the most artificial erroneous judgement etc. of failing to judge, therefore public security is utilizing prison
In the actual combat of control video recording investigation, video is examined the problem of inefficiency and is needed solution badly.
3, present stage is in most cases to be applied for relevant department by criminal investigation group after the incident
Personnel dispatch associated monitoring data, then carry out artificial semi-automatic search judgement, by
In the reason of site environment, the image of shooting is unsatisfactory, it is impossible to see details clearly, inconvenience
Process in public security department or vehicle supervision department's supervision, in the case of video quality is low
Public security police is the most at a loss what to do, in the face of the video monitoring data of magnanimity also can only time-consuming be taken
The examination one by one of power, this is felt to be sorely distressed by criminal investigation people's police.
4, the video of long-term viewing low quality may increase the weight of the burden of people's eyes, holds
Being easily generated visual fatigue, even can have a dizzy spell, scene image is right in the presence of a harsh environment
All can be changed or degenerate than degree and color, the many features contained in image are the most capped
Or fuzzy, obtain is degraded image.
To this end, people have carried out long-term exploration, it is proposed that various solutions.
Such as, to disclose a kind of video based on video surveillance network the denseest for Chinese patent literature
Compression method [application number: 201110208090.X], it is to have taking the photograph of overlapping region from two
Camera obtains the first video source and the second video source and to moving Object Segmentation and tracking, carries
Take background image and video surveillance network movement locus；Generate based on regarding according to background image
Frequency monitors the panoramic picture of network, obtains two video cameras and respectively correspond tos panoramic picture
Projection matrix real-time update panoramic picture based on video surveillance network background；Utilize two
The projection matrix of individual camera pan image, projects to panoramic picture space track and obtains
Projected footprint；Use figure coupling, based on the random walk thought video source to two video cameras
Projected footprint coupling, obtain mate track pair, the coupling track of overlapping region is selected
Select and merge, obtaining complete large scene track；Large scene track is temporally rearranged,
Complete large scene track is represented by panoramic picture, obtains based on panoramic picture
Video concentrate." computer engineering and application " the 25th phase in 2008 discloses a kind of base
Video retrieval method (author: Xue Liqin in motion analysis；Zhang Xiujuan), people is described
From different application purposes, it is proposed that motion analysis in many different videos
Method.Summing up, these analysis methods are segmented into four classes: light stream estimation algorithm, motion
Vector estimation algorithm, divide based on histogrammic statistical analysis technique, parameterized global motion
Analysis method.
Although such scheme improves video analysis efficiency to a certain extent, but still
Also exist and be difficult to obtain information required in video easily, only shorten whole
The length of video, and to the association between the effect of video, details, video etc. not
Make any process.
Summary of the invention
It is an object of the invention to for the problems referred to above, it is provided that video time is significantly contracted by one
Short, it is possible to obtain information required in video easily, improve the video of work efficiency
Monitoring intelligent analyzes search method.
It is a further object of the present invention to provide one and can obtain in video required easily
Information, improve work efficiency video monitoring intellectual analysis searching system.
For reaching above-mentioned purpose, present invention employs following technical proposal: this video monitoring
Intellectual analysis search method, it is characterised in that this method comprises the steps:
A, video source modeling: video image camera collection arrived by video enhancement techniques
Present in image is dark, fuzzy, cross the brightest in any one or more situation eliminate,
Obtain the high-quality video file of video source modeling；
B, video frequency searching:
A, dynamically analysis: utilize video retrieval technology based on motion analysis, record monitoring
As the static fragment in video is skipped or F.F. automatically, moving target is provided prompting, and
Obtain and press the sport video fragment that frame intercepts；
B, coding restructuring: the sport video fragment intercepted by frame is carried out video restructuring, obtains
The restructuring sport video for selective analysis after must recombinating；
C, result queue: play original image and restructuring sport video picture simultaneously,
And by play restructuring sport video according to different moving targets on reproduction time axle
Distinguish labelling；
D, behavior analysis: the moving situation of different motion target in restructuring sport video is entered
Row is analyzed, if the movement velocity of moving target exceedes the threshold speed that this moving target sets,
Then send alarm signal.
In above-mentioned video monitoring intellectual analysis search method, in above-mentioned step c,
Set the time period separator only having personage's motion play and show the first color,
The time period separator only having vehicle motion play shows the second color, has play
The time period separator simultaneously having personage and vehicle motion shows the 3rd color.
In above-mentioned video monitoring intellectual analysis search method, in above-mentioned step d,
The speed setting personage's normal ambulation is not more than 2m/s, if meeting or exceeding 2m/s, for running quickly
Run, and signal an alert when personage is run；The speed setting normal vehicle operation is little
In 22m/s, if meeting or exceeding 22m/s, being overspeed of vehicle, and sending out during overspeed of vehicle
Go out alarm signal.
In above-mentioned video monitoring intellectual analysis search method, in above-mentioned step B,
Also include:
E, video frequency abstract: same moving target different time occur carries out sectional drawing respectively
Backup, forms summary picture, and shows the concrete time on summary picture, elected in certain
During one summary picture, if automatically playing in this summary picture place video segment coherent front
Dry second and rear some seconds.
In above-mentioned video monitoring intellectual analysis search method, in above-mentioned step B,
Also include:
F, trace labelling: the moving target in selected restructuring sport video, and to this motion
Target makes a mark, and when selected moving target continues motion, described labelling with
It is synchronized with the movement with moving target.
In above-mentioned video monitoring intellectual analysis search method, in above-mentioned step B,
Also include:
G, GIS link: utilize geography information binding time to extract relative motion piece of video
Section the relative motion video segment that automatically arrives each camera collection are according to time and sky
Between recombinate, form moving target and move the continuous videos fragment in moment.
In above-mentioned video monitoring intellectual analysis search method, in above-mentioned step B,
Also include:
H, render compensation: after a certain two field picture is achieved, under the conditions of good visual
The image at the same visual angle of the same photographic head of analog capture again, then according to the figure captured
As the image achieved is carried out later rendering compensation, it is provided that contrast.
In above-mentioned video monitoring intellectual analysis search method, in above-mentioned step A,
Described video source modeling includes that front-end collection strengthens, screen shows strengthens and video segment increases
Any one or more of persistent erection of the penis；Described front-end collection strengthens by photographic head with deposit
The front-end collection being connected to video source modeling between storage device strengthens module realization；Described
Screen shows strengthens by being connected to video source modeling between storage device and display screen
Play and strengthen module realization；Described video segment strengthens by selecting in storage device
Video image through for video source modeling video source modeling work station process after preserve to storage
In device standby.
Above-mentioned video monitoring intellectual analysis search method is to divide in following video monitoring intelligence
Implementing in analysis searching system, native system includes:
Video source modeling module, for by video enhancement techniques by camera collection to regard
Frequently any one or more situation during image is dark, fuzzy, the brightest present in image
Eliminate, it is thus achieved that the high-quality video file of video source modeling；
Dynamically analyze module, for utilizing video retrieval technology based on motion analysis, will
Static fragment in monitoring video video is skipped or F.F. automatically, provides moving target and carries
Show, and obtain the sport video fragment pressing frame intercepting；
Coding recombination module, for carrying out video weight by the sport video fragment intercepted by frame
Group, it is thus achieved that the restructuring sport video for selective analysis after restructuring；
Result queue module, plays original image and restructuring sport video for simultaneously
Picture, and by play restructuring sport video according to different moving targets play time
Labelling is distinguished on countershaft；
Behavior analysis module, for the motion of different motion target in restructuring sport video
Situation is analyzed, if the movement velocity of moving target exceedes the speed that this moving target sets
Degree threshold value, then send alarm signal；
Described module of dynamically analyzing is connected with coding recombination module, described result queue
Module and behavior analysis module are both connected to encode on recombination module, described video source modeling
Module is connected with dynamically analysis module or coding recombination module.
In above-mentioned video monitoring intellectual analysis searching system, native system also includes:
Video frequency abstract module, enters respectively for same moving target different time occur
Row sectional drawing backs up, and forms summary picture, and shows the concrete time on summary picture, when
When choosing a certain summary picture, automatically play in this summary picture place video segment coherent
Front some seconds and rear some seconds；
Trace labelling module, the moving target in selected restructuring sport video, and right
This moving target makes a mark, and when selected moving target continues motion, described
Labelling pursuit movement object synchronization is moved；
GIS interlocking module, is used for utilizing geography information binding time to extract relative motion
Video segment the relative motion video segment that automatically each camera collection arrived according to time
Between and space recombinate, form moving target and move the continuous videos fragment in moment；
Render compensating module, after a certain two field picture is achieved, in good visual
Under the conditions of the image at the same visual angle of the same photographic head of analog capture again, then according to seizure
To image to achieve image carry out later rendering compensation, it is provided that contrast；
Described video frequency abstract module, trace labelling module, GIS interlocking module and rendering
Compensating module is connected with coding recombination module respectively.
Compared with prior art, this video monitoring intellectual analysis search method and system thereof
Advantage be: skip or static fragment in F.F. video, video time significantly shortened,
Information required in video can be obtained easily, significantly save the time, improve work
Efficiency, and can easily required information be retrieved and labelling, the most not
Contact is set up, it is easy to operation is implemented between continuous print information.Collection image enhaucament, retrieval compression,
Intellectual analysis, summary are added up in one, embody advanced person, efficiently, reliable, convenient and
Five practical big features, it is achieved that enhancing, the retrieval pressure of video file to video image
Contracting, the behavior intellectual analysis etc. of personage's vehicle, thus the letter needed for obtaining quickly and accurately
Breath, can scout the powerful effective criminal investigation of offer of solving a case and support for public security cadres and police.
Accompanying drawing explanation
Fig. 1 is the structural representation that the present invention provides.
In figure, photographic head 1, storage device 2, front-end collection strengthen module 3, display screen
4, play and strengthen module 5, video source modeling work station 6, video source modeling module 7, dynamically
Analyze module 8, coding recombination module 9, result queue module 10, behavior analysis module
11, video frequency abstract module 12, trace labelling module 13, GIS interlocking module 14, render
Compensating module 15.
Detailed description of the invention
This video monitoring intellectual analysis search method comprises the steps:
A, video source modeling: video image camera collection arrived by video enhancement techniques
Present in image is dark, fuzzy, cross the brightest in any one or more situation eliminate,
Obtain the high-quality video file of video source modeling；
B, video frequency searching:
A, dynamically analysis: utilize video retrieval technology based on motion analysis, record monitoring
As the static fragment in video is skipped or F.F. automatically, moving target is provided prompting, and
Obtain and press the sport video fragment that frame intercepts；
B, coding restructuring: the sport video fragment intercepted by frame is carried out video restructuring, obtains
The restructuring sport video for selective analysis after must recombinating；
C, result queue: play original image and restructuring sport video picture simultaneously,
And by play restructuring sport video according to different moving targets on reproduction time axle
Distinguish labelling；
D, behavior analysis: the moving situation of different motion target in restructuring sport video is entered
Row is analyzed, if the movement velocity of moving target exceedes the threshold speed that this moving target sets,
Then send alarm signal.
In the present invention, such as Fig. 2, shown in 3 and 4, video source modeling include front-end collection strengthen,
Screen shows strengthens any one or more in strengthening with video segment.Front-end collection increases
The strong front end passing through to be connected to video source modeling between photographic head 1 and storage device 2 is adopted
Collection strengthens module 3 and realizes；Screen shows strengthens by storage device 2 and display screen 4
Between be connected to video source modeling broadcasting strengthen module 5 realize；Video segment strengthens logical
Cross video image selected in storage device 2 through being used for the video source modeling work of video source modeling
Preserve to storage device 2 standby after making station 6 process.
The safe city video monitoring system that present stage has built, front-end collection environment division
The most backward, it is impossible to well to capture the real scene under various adverse circumstances, at night
Between dark low illumination, dense fog sand and dust vile weather or local light cross bright in the case of image
Quality is substantially reduced, it is impossible to therefrom obtains useful information, and watches low quality for a long time
Video may increase the weight of the burden of people's eyes, easily produces visual fatigue, even can head
Dizzy dizzy.Contrast and the color of scene image all can be changed or move back in the presence of a harsh environment
Changing, the many features contained in image are all capped or fuzzy, and obtain is degraded image.
To this end, the application of image enhancement technique seem particularly in the urgent need to, can be for scouting etc. of handling a case
Effective reconnaissane tool is provided, is greatly improved work efficiency.As in figure 2 it is shown, front end is adopted
Integrate enhancing module 3 for FPGA plate (support BNC HDMI), so that rear end
High-quality video file is obtained during storage.Play and strengthen module 5 for being attached to video biography
FPAG plate on defeated final node, can render to key video sequence image supply on giant-screen
Exploration and analysis.Video source modeling work station 6 for by web camera, DVR,
Key video sequence image in the storage device 2 such as video server carries out strengthening to be analyzed, reduction
Real scene and personage.Wherein, video source modeling work station 6 is existing in prior art
Equipment.Wherein, the algorithm of above-mentioned various video source modeling includes: image dark strengthens；Figure
As mist elimination strengthens；Regional display strengthens；Wide dynamically process；Vedio noise reduction；Video debounce
Dynamic；Image removes motion blur；Regional area super-resolution strengthens.Above-mentioned algorithm all can be adopted
With algorithm existing in prior art, do not repeat.
In above-mentioned step c, set the time period district only having personage's motion play
Minute mark note display the first color, the time period separator only having vehicle motion play
Show the second color, while playing, have the time period of personage and vehicle motion to distinguish mark
Note display the 3rd color.The such as blue time period represents that personage, red time period represent car
Etc., personage and vehicle time of occurrence section simultaneously is green, and other times section is shown as black
Color.In above-mentioned step d, the speed setting personage's normal ambulation is not more than 2m/s,
If meeting or exceeding 2m/s, for running, and signal an alert when personage is run；Set
The speed of normal vehicle operation is not more than 22m/s, if meeting or exceeding 22m/s, is car
Hypervelocity, and signal an alert during overspeed of vehicle.Retrieval analysis for vehicle can be right
The features such as vehicle color, vehicle volume, the number-plate number (under high-resolution) carry out classification and pluck
Want, same car was occurred 2 times within the time of association and above emphasis ejects and reports to the police.
The condition of retrieval can with custom parameter (sensitivity), including movement velocity, article size,
Color, number etc., different parameters exports different configuration files, Ke Yizuo automatically
Preserve for retrieval prediction scheme, conveniently same type of case is compared.
In above-mentioned step B, also include:
E, video frequency abstract: same moving target different time occur carries out sectional drawing respectively
Backup, forms summary picture, and shows the concrete time on summary picture, elected in certain
During one summary picture, if automatically playing in this summary picture place video segment coherent front
Dry second and rear some seconds.In the present embodiment, automatically play this summary picture place piece of video
Front ten seconds coherent in Duan and rear ten seconds.
In above-mentioned step B, also include:
F, trace labelling: the moving target in selected restructuring sport video, and to this motion
Target makes a mark, and when selected moving target continues motion, described labelling with
It is synchronized with the movement with moving target.
In above-mentioned step B, also include:
G, GIS link: utilize geography information binding time to extract relative motion piece of video
Section the relative motion video segment that automatically arrives each camera collection are according to time and sky
Between recombinate, form moving target and move the continuous videos fragment in moment.
In above-mentioned step B, also include:
H, render compensation: after a certain two field picture is achieved, under the conditions of good visual
The image at the same visual angle of the same photographic head of analog capture again, then according to the figure captured
As the image achieved is carried out later rendering compensation, it is provided that contrast.
As it is shown in figure 1, this video monitoring intellectual analysis searching system includes:
Video source modeling module 7, for arrive camera collection by video enhancement techniques
Any one or more feelings during image is dark, fuzzy, the brightest present in video image
Condition eliminates, it is thus achieved that the high-quality video file of video source modeling；
Dynamically analyze module 8, for utilizing video retrieval technology based on motion analysis,
Static fragment in monitoring video video is skipped or F.F. automatically, moving target is given
Prompting, and obtain the sport video fragment pressing frame intercepting；
Coding recombination module 9, for carrying out video by the sport video fragment intercepted by frame
Restructuring, it is thus achieved that the restructuring sport video for selective analysis after restructuring；
Result queue module 10, regards with restructuring motion for playing original image simultaneously
Frequently picture, and the restructuring sport video play is being play according to different moving targets
Labelling is distinguished on time shaft；
Behavior analysis module 11, for the fortune of different motion target in restructuring sport video
Dynamic situation is analyzed, if the movement velocity of moving target exceedes what this moving target set
Threshold speed, then send alarm signal；
Dynamically analyze module 8 to be connected with coding recombination module 9, described result queue mould
Block 10 and behavior analysis module 11 are both connected to encode on recombination module 9, and described regards
Frequency strengthens module 7 and is connected with dynamically analyzing module 8 or encoding recombination module 9.
Native system also includes:
Video frequency abstract module 12, for same moving target difference different time occur
Carry out sectional drawing backup, form summary picture, and show the concrete time on summary picture,
In Dang Xuan during a certain summary picture, automatically play in this summary picture place video segment and connect
The front some seconds passed through and rear some seconds；
Trace labelling module 13, the moving target in selected restructuring sport video, and
This moving target is made a mark, and when selected moving target continues motion, described
Labelling pursuit movement object synchronization motion；
GIS interlocking module 14, is used for utilizing geography information binding time to extract relevant fortune
Dynamic video segment the relative motion video segment that automatically each camera collection arrived according to
Recombinating in time and space, forms moving target and move the continuous videos fragment in moment；
Render compensating module 15, after a certain two field picture is achieved, well regarding
The image at the same visual angle of the same photographic head of analog capture again under the conditions of feel, then according to catching
The image grasped carries out later rendering compensation to the image achieved, it is provided that contrast；
Video frequency abstract module 12, trace labelling module 13, GIS interlocking module 14 and wash with watercolours
Dye compensating module 15 is connected with coding recombination module 9 respectively.
Specific embodiment described herein is only to present invention spirit theory for example
Bright.Those skilled in the art can be to described specific embodiment
Make various amendment or supplement or use similar mode to substitute, but without departing from
The spirit of the present invention or surmount scope defined in appended claims.
Although the most more employing photographic head 1, storage device 2, front-end collection enhancing
Module 3, display screen 4, broadcasting strengthen module 5, video source modeling work station 6, video increasing
Strong module 7, dynamically analyze module 8, coding recombination module 9, result queue module 10,
Behavior analysis module 11, video frequency abstract module 12, trace labelling module 13, GIS link
Module 14, render the terms such as compensating module 15, but be not precluded from using other term
Probability.Use these terms to be only used to more easily and describe and explain the present invention's
Essence；Being construed as any additional restriction is all to disagree with spirit of the present invention
The back of the body.
Claims (3)
1. a video monitoring intellectual analysis search method, it is characterised in that this method bag
Include following step:
A, video source modeling: video image camera collection arrived by video enhancement techniques
Present in image is dark, fuzzy, cross the brightest in any one or more situation eliminate,
Obtain the high-quality video file of video source modeling；
B, video frequency searching:
A, dynamically analysis: utilize video retrieval technology based on motion analysis, record monitoring
As the static fragment in video is skipped or F.F. automatically, moving target is provided prompting, and
Obtain and press the sport video fragment that frame intercepts；
B, coding restructuring: the sport video fragment intercepted by frame is carried out video restructuring, obtains
The restructuring sport video for selective analysis after must recombinating；
C, result queue: play original image and restructuring sport video picture simultaneously,
And by play restructuring sport video according to different moving targets on reproduction time axle
Distinguish labelling；
D, behavior analysis: the moving situation of different motion target in restructuring sport video is entered
Row is analyzed, if the movement velocity of moving target exceedes the threshold speed that this moving target sets,
Then send alarm signal；
In above-mentioned step c, set the time period district only having personage's motion play
Minute mark note display the first color, the time period separator only having vehicle motion play
Show the second color, while playing, have the time period of personage and vehicle motion to distinguish mark
Note display the 3rd color；
In above-mentioned step d, the speed setting personage's normal ambulation is not more than 2m/s,
If meeting or exceeding 2m/s, for running, and signal an alert when personage is run；Set
The speed of normal vehicle operation is not more than 22m/s, if meeting or exceeding 22m/s, is car
Hypervelocity, and signal an alert during overspeed of vehicle；
In above-mentioned step B, also include:
E, video frequency abstract: same moving target different time occur carries out sectional drawing respectively
Backup, forms summary picture, and shows the concrete time on summary picture, elected in certain
During one summary picture, if automatically playing in this summary picture place video segment coherent front
Dry second and rear some seconds；
In above-mentioned step B, also include:
F, trace labelling: the moving target in selected restructuring sport video, and to this motion
Target makes a mark, and when selected moving target continues motion, described labelling with
It is synchronized with the movement with moving target；
In above-mentioned step B, also include:
G, GIS link: utilize geography information binding time to extract relative motion piece of video
Section the relative motion video segment that automatically arrives each camera collection are according to time and sky
Between recombinate, form moving target and move the continuous videos fragment in moment；
In above-mentioned step B, also include:
H, render compensation: after a certain two field picture is achieved, under the conditions of good visual
The image at the same visual angle of the same photographic head of analog capture again, then according to the figure captured
As the image achieved is carried out later rendering compensation, it is provided that contrast.
Video monitoring intellectual analysis search method the most according to claim 1, it is special
Levy and be, in above-mentioned step A, described video source modeling include front-end collection strengthen,
Screen shows strengthens any one or more in strengthening with video segment；Described front end
Gather and strengthen by being connected to video between photographic head (1) and storage device (2)
The front-end collection strengthened strengthens module (3) and realizes；Described screen shows strengthen by
The broadcasting being connected to video source modeling between storage device (2) and display screen (4) strengthens
Module (5) realizes；Described video segment strengthens by selecting in storage device (2)
Fixed video image preserves after processing for the video source modeling work station (6) of video source modeling
Standby to storing in device (2).
3. a video monitoring intellectual analysis searching system, it is characterised in that native system bag
Include:
Video source modeling module (7), for arriving camera collection by video enhancement techniques
Video image present in image is dark, fuzzy, cross the brightest in any one or more
Situation eliminates, it is thus achieved that the high-quality video file of video source modeling；
Dynamically analyze module (8), for utilizing video retrieval technology based on motion analysis,
Static fragment in monitoring video video is skipped or F.F. automatically, moving target is given
Prompting, and obtain the sport video fragment pressing frame intercepting；
Coding recombination module (9), for regarding the sport video fragment intercepted by frame
Frequency restructuring, it is thus achieved that the restructuring sport video for selective analysis after restructuring；
Result queue module (10), plays original image and restructuring motion for simultaneously
Video pictures, and the restructuring sport video play is being broadcast according to different moving targets
Put and on time shaft, distinguish labelling；
Behavior analysis module (11), for different motion target in restructuring sport video
Moving situation is analyzed, if the movement velocity of moving target exceedes this moving target and sets
Threshold speed, then send alarm signal；
Described module (8) of dynamically analyzing is connected with coding recombination module (9), described
Result queue module (10) and behavior analysis module (11) be both connected to coding restructuring
In module (9), described video source modeling module (7) with dynamically analyze module (8) or
Coding recombination module (9) is connected；
Native system also includes:
Video frequency abstract module (12), divides for same moving target different time occur
Do not carry out sectional drawing backup, form summary picture, and show the concrete time on summary picture,
In Dang Xuan during a certain summary picture, automatically play in this summary picture place video segment and connect
The front some seconds passed through and rear some seconds；
Trace labelling module (13), the moving target in selected restructuring sport video,
And this moving target is made a mark, and when selected moving target continues motion, institute
The labelling pursuit movement object synchronization motion stated；
GIS interlocking module (14), is used for utilizing geography information binding time to extract relevant
Sport video fragment and automatically to each camera collection to relative motion video segment press
Recombinate according to time and space, form moving target and move the continuous videos fragment in moment；
Render compensating module (15), after a certain two field picture is achieved, well
The image at the same visual angle of the same photographic head of analog capture again, then basis under visual condition
The image captured carries out later rendering compensation to the image achieved, it is provided that contrast；
Described video frequency abstract module (12), trace labelling module (13), GIS linkage
Module (14) is connected with coding recombination module (9) respectively with rendering compensating module (15).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310012753.XA CN103106250B (en) | 2013-01-14 | 2013-01-14 | Video monitoring intellectual analysis search method and system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310012753.XA CN103106250B (en) | 2013-01-14 | 2013-01-14 | Video monitoring intellectual analysis search method and system thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103106250A CN103106250A (en) | 2013-05-15 |
| CN103106250B true CN103106250B (en) | 2016-11-23 |
Family
ID=48314105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310012753.XA Expired - Fee Related CN103106250B (en) | 2013-01-14 | 2013-01-14 | Video monitoring intellectual analysis search method and system thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103106250B (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102184987B1 (en) * | 2013-11-15 | 2020-12-01 | 엘지전자 주식회사 | Picture display device and operating method thereof |
| CN105323501A (en) * | 2014-07-28 | 2016-02-10 | 中兴通讯股份有限公司 | Concentrated video moving object marking method, playing method and apparatus thereof |
| CN104539897A (en) * | 2014-12-27 | 2015-04-22 | 任文华 | Monitoring camera |
| CN105323547B (en) * | 2015-03-17 | 2018-05-15 | 四川浩特通信有限公司 | A kind of video concentration systems |
| CN105306880A (en) * | 2015-03-17 | 2016-02-03 | 四川浩特通信有限公司 | Video concentration method |
| CN105323548B (en) * | 2015-03-17 | 2018-05-15 | 四川浩特通信有限公司 | A kind of intelligence actual combat system |
| CN104936021B (en) * | 2015-06-02 | 2018-03-09 | 无锡天脉聚源传媒科技有限公司 | The monitoring method and device of a kind of video playback |
| CN105007442A (en) * | 2015-07-22 | 2015-10-28 | 深圳市万姓宗祠网络科技股份有限公司 | Marking method for continuously recording video and image |
| CN105828030A (en) * | 2016-03-14 | 2016-08-03 | 珠海经济特区远宏科技有限公司 | Video investigation mobile terminal system |
| CN105871872A (en) * | 2016-04-28 | 2016-08-17 | 珠海经济特区远宏科技有限公司 | Movable video investigation method |
| CN106021621B (en) * | 2016-07-16 | 2019-11-01 | 大箴(杭州)科技有限公司 | A kind of satellite based on big data processing is solved a case system |
| CN109803112A (en) * | 2017-11-16 | 2019-05-24 | 中兴通讯股份有限公司 | Video analysis management method based on big data, apparatus and system, storage medium |
| CN108289194A (en) * | 2017-12-28 | 2018-07-17 | 天津天地伟业信息系统集成有限公司 | Video quick playback method and device |
| CN108733821A (en) * | 2018-05-22 | 2018-11-02 | 武汉微创光电股份有限公司 | A kind of distribution of monitor video sectional drawing and methods of exhibiting and system |
| CN108765943A (en) * | 2018-05-30 | 2018-11-06 | 深圳市城市公共安全技术研究院有限公司 | Intelligent vehicle monitoring method, monitoring system and server |
| CN111866428B (en) * | 2019-04-29 | 2023-03-14 | 杭州海康威视数字技术股份有限公司 | Historical video data processing method and device |
| CN111147778B (en) * | 2019-12-23 | 2022-06-07 | 浙江大华技术股份有限公司 | Track recording method and device based on digital video recorder and storage medium |
| CN115455275B (en) * | 2022-11-08 | 2023-02-03 | 广东卓维网络有限公司 | Video processing system integrated with inspection equipment |
| CN117714814B (en) * | 2023-12-16 | 2024-05-17 | 浙江鼎世科技有限公司 | Video storage access system based on intelligent strategy |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1863295A (en) * | 2004-05-13 | 2006-11-15 | 索尼公司 | Image data processing apparatus, image data processing method, program and recording medium |
| CN102256065A (en) * | 2011-07-25 | 2011-11-23 | 中国科学院自动化研究所 | Automatic video condensing method based on video monitoring network |
| CN102799862A (en) * | 2012-06-29 | 2012-11-28 | 陕西省交通规划设计研究院 | System and method for pedestrian rapid positioning and event detection based on high definition video monitor image |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7751632B2 (en) * | 2005-02-15 | 2010-07-06 | Panasonic Corporation | Intelligent, dynamic, long-term digital surveilance media storage system |
-
2013
- 2013-01-14 CN CN201310012753.XA patent/CN103106250B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1863295A (en) * | 2004-05-13 | 2006-11-15 | 索尼公司 | Image data processing apparatus, image data processing method, program and recording medium |
| CN102256065A (en) * | 2011-07-25 | 2011-11-23 | 中国科学院自动化研究所 | Automatic video condensing method based on video monitoring network |
| CN102799862A (en) * | 2012-06-29 | 2012-11-28 | 陕西省交通规划设计研究院 | System and method for pedestrian rapid positioning and event detection based on high definition video monitor image |
Non-Patent Citations (1)
| Title |
|---|
| 基于视频处理的交通事件识别方法研究;胡宏宇;《中国博士论文全文数据库 信息科技辑》;20100815(第8期);第69,73,79-80页，图4-10 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103106250A (en) | 2013-05-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103106250B (en) | Video monitoring intellectual analysis search method and system thereof | |
| CN109241933A (en) | Video linkage monitoring method, monitoring server, video linkage monitoring system | |
| CN107004271B (en) | Display method, display apparatus, electronic device, computer program product, and storage medium | |
| US10909424B2 (en) | Method and system for object tracking and recognition using low power compressive sensing camera in real-time applications | |
| CN109919058A (en) | A kind of multisource video image highest priority rapid detection method based on Yolo V3 | |
| CN109711318A (en) | A kind of plurality of human faces detection and tracking based on video flowing | |
| CN106056624A (en) | Unmanned aerial vehicle high-definition image small target detecting and tracking system and detecting and tracking method thereof | |
| CN109948553A (en) | A kind of multiple dimensioned dense population method of counting | |
| CN102892007B (en) | Promote the synchronous method and system with obtaining tracking of color balance between multiple video camera | |
| CN108629299B (en) | Long-time multi-target tracking method and system combining face matching | |
| CN106339657B (en) | Crop straw burning monitoring method based on monitor video, device | |
| CN102665071A (en) | Intelligent processing and search method for social security video monitoring images | |
| CN107358141B (en) | Data identification method and device | |
| Zhou et al. | A UAV patrol system using panoramic stitching and object detection | |
| CN109766743A (en) | A kind of intelligent bionic policing system | |
| CN103729620B (en) | A kind of multi-view pedestrian detection method based on multi-view Bayesian network | |
| CN109359593A (en) | A kind of sleet environment fuzzy pictures monitoring and pre-alarming method based on image local grid | |
| CN107644431A (en) | Effective method for concentration of video monitoring in the case of a kind of low-light (level) | |
| CN110516602A (en) | A kind of public traffice passenger flow statistical method based on monocular camera and depth learning technology | |
| CN111860457A (en) | Fighting behavior recognition early warning method and recognition early warning system thereof | |
| EP3975133A1 (en) | Processing of images captured by vehicle mounted cameras | |
| CN103914850B (en) | Automatic video labeling method and system based on motion matching | |
| CN103077227B (en) | Video concentrates retrieval analysis method and system thereof | |
| CN103106633A (en) | Method and system of visual front view article screenshot based on gaussian mixture model | |
| CN111369578A (en) | Intelligent tracking method and system for holder transaction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20161123 Termination date: 20190114 |