CN109302562A - Target acquisition method for tracing, system, medium and equipment based on video - Google Patents
Target acquisition method for tracing, system, medium and equipment based on video Download PDFInfo
- Publication number
- CN109302562A CN109302562A CN201811191956.9A CN201811191956A CN109302562A CN 109302562 A CN109302562 A CN 109302562A CN 201811191956 A CN201811191956 A CN 201811191956A CN 109302562 A CN109302562 A CN 109302562A
- Authority
- CN
- China
- Prior art keywords
- target
- monitor area
- detection
- turntable
- tracing
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
-
- 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
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/025—Services making use of location information using location based information parameters
- H04W4/027—Services making use of location information using location based information parameters using movement velocity, acceleration information
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Studio Devices (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The present invention provides a kind of target acquisition method for tracing based on video, comprising: adjustment supervision equipment turns to monitor area；Initialization tracking is carried out to the monitor area；In initialization tracing process, judge whether target enters the monitor area；If it is not, then continuing to carry out initialization tracking to the monitor area；If so, keeping track to the target；During keeping track, judge whether the target loses；If it is not, then continuing to keep track the target；If so, predicting and finding the target；If within a preset period of time, giving the target for change, then the target is continued to keep track；If within a preset period of time, not giving the target for change, then terminate to track.In this way, infrared image is difficult to accurately to detect, the strong target of movement maneuverability has better track lock ability, can reliablely and stablely track target.
Description
Technical field
The present invention relates to target tracking technologies, and in particular to a kind of target acquisition method for tracing based on video, system, Jie
Matter and equipment.
Background technique
Video object track lock is common in national defense system in the tracking of the targets such as aircraft, guided missile or security system
To in the tracking of the targets such as people, vehicle.The present invention is mainly reflected in two for difference existing for problem for problem and these applications
Aspect.First is that target has high maneuverability.Such as unmanned plane and birds, it can promptly be accelerated, be slowed down, the movement such as deflecting,
Its acceleration change is big.For aircraft, the big still velocity variations of guided missile uniform velocity and little moving target, motion profile
It is more difficult to predict, the requirement to tracking is also higher.Second is that using the feature based on visible images.More compared to Infrared Image Information
It is abundant, there is good detectivity to cryogenic object.
The problems such as object variations are big, background is complicated, imaging circumstances are changeable, classical way and depth are faced on method for tracing
There is limitation in learning method.Therefore, the present invention combines the two, respectively takes the chief, to realize reliable and stable tracking target.
Summary of the invention
For the defects in the prior art, the present invention provides a kind of target acquisition method for tracing based on video, system, Jie
Matter and equipment, are difficult to accurately to detect, the strong target of movement maneuverability has better track lock ability to infrared image, can
Reliablely and stablely track target.
In a first aspect, the present invention provides a kind of target acquisition method for tracing based on video, comprising:
It adjusts supervision equipment and turns to monitor area；
Initialization tracking is carried out to the monitor area；In initialization tracing process, it is described to judge whether target enters
Monitor area；If it is not, then continuing to carry out initialization tracking to the monitor area；If so, persistently being chased after to the target
Track；
During keeping track, judge whether the target loses；If it is not, then continuing persistently to chase after the target
Track；If so, predicting and finding the target；
If within a preset period of time, giving the target for change, then the target is continued to keep track；If when default
Between in section, do not give the target for change, then terminate to track.
Optionally, the adjustment supervision equipment turns to monitor area, comprising:
The turntable angle and lens focus for adjusting supervision equipment make supervision equipment turn to monitor area.
It is optionally, described that initialization tracking is carried out to the monitor area, comprising:
Method for tracing is initialized using two frames, initialization tracking is carried out to the monitor area, the specific steps are as follows:
Step 11): carrying out target detection to current frame image, obtains a kind of first detection result；
Step 12): the second detection result of one kind of target detection is once carried out before reading to picture frame；
Step 13): bond area matching algorithm, to a kind of first detection result and a kind of second detection result
It is matched, obtains the first pairing score；
Step 14): judge whether the highest pairing score in the first pairing score is more than target gate；
If it is not, then the target does not enter into the monitor area；
A kind of first detection result is covered into a kind of second detection result, newest video frame is read, continues to hold
Row step 11) is to step 14)；
If so, determining that the target enters the monitor area；
According in a kind of first detection result and a kind of second detection result center range difference and when
Between it is poor, calculate first object movement velocity；According to the first object movement velocity, the target is kept track.
It is optionally, described that initialization tracking is carried out to the monitor area, comprising:
Method for tracing is initialized using three frames, initialization tracking is carried out to the monitor area, the specific steps are as follows:
Step 21): target detection is carried out to current frame image, obtains two the first detection results of class；
Step 22): two the second detection results of class of target detection are carried out to picture frame twice before reading and two class thirds are visited
Survey result；
Step 23): two classes, first detection result, two the second detection results of class and two class third detection results are carried out
Pairing obtains the second pairing score；
Step 24): judge whether the highest pairing score in the second pairing score is more than target gate；
If it is not, then the target does not enter into the monitor area；
Two classes, first detection result is covered into previous two classes, second detection result, by two class second
Detection result covers the two classes third detection result, and reads newest video frame, continues to execute step 21) to step 24)；
If so, determining that the target enters the monitor area；
According in target acquisition result three times center range difference and the time difference, calculate the second target speed,
According to second target speed, the target is kept track.
It is optionally, described that the target is kept track, comprising:
According to target speed, the turntable revolving speed in the supervision equipment is adjusted, the target is made to be maintained at the prison
The central area of viewed area, the specific steps are as follows:
Step 31): target speed is converted into the horizontal rotation speed and pitching rotation speed of turntable；
Step 32): according to the horizontal rotation speed and pitching rotation speed, the turntable revolving speed is adjusted；
Step 33): by initialization tracing process, highest matches the target acquisition knot of the corresponding last frame image of score
Fruit is recorded as historical information；
Step 34): target detection is carried out to current frame image, obtains the 4th target acquisition result；
Step 35): reading the historical information, and the 4th target acquisition result is matched with the historical information
It is right, it calculates third and matches score；
Step 36): judge whether the highest pairing score in the third pairing score is lower than target gate；
If so, thinking that the target is temporarily lost, predicting and finding the target；
If not, then it is assumed that the target is not lost；
Step 37): if target is not lost, the third is matched to the corresponding detection of highest pairing score in score
As a result it updates and arrives the historical information；
Step 38): according to the horizontal rotation speed of current goal imaging position and horizontal relative velocity adjustment turntable；According to
The pitching rotation speed of current goal imaging position and pitching relative velocity adjustment turntable, repeats step 32) to step 38).
Optionally, step 36) can be with are as follows:
If the 4th target acquisition result is 0, alternatively, third pairing score is lower than target gate, then it is assumed that institute
It states target temporarily to lose, predict and finds the target；
If the 4th target acquisition result is not 0, and third pairing score is not less than target gate, then it is assumed that institute
Target is stated not lose.
It is optionally, described to predict and find the target, comprising:
According to the last imaging position of the target and relative velocity, turntable revolving speed is adjusted, last along the target flies
Row transitional search target.
Second aspect, the present invention provide a kind of target acquisition tracing system based on video, comprising: processing chip, turntable,
Camera lens and camera；
The camera lens is arranged on the camera；The camera is fixed on the turntable；The turntable, camera lens and camera
Form supervision equipment；
The processing chip is all connected with the turntable, camera lens and camera；
The processing chip is used to control the working condition of the supervision equipment, and perform claim requires 1 to 7 any one institute
The target acquisition method for tracing based on video stated.
The third aspect, the present invention provide a kind of computer readable storage medium, are stored thereon with computer program, the program
The target acquisition method for tracing based on video in first aspect is realized when being executed by processor.
Fourth aspect, the present invention provide a kind of target acquisition tracing equipment based on video, comprising: memory, processor
And storage is on a memory and the computer program that can run on a processor, when processor execution described program, realizes the
The target acquisition method for tracing based on video in one side.
The present invention turns to monitor area by adjusting supervision equipment, then carries out initialization tracking to monitor area, is chasing after
Once confirmation target enters monitor area, then starting keeps track program during track, target is kept track, otherwise,
It keeps current initialization to track state, during keeping track, if target is not lost, remains operational；If target is lost, open
Dynamic prediction and finder, find target；If giving target for change, it is switched to and keeps track program, if not giving for change for a long time, terminates
Tracking.In this way, being difficult to accurately to detect to infrared image, the strong target of movement maneuverability has better track lock energy
Power can reliablely and stablely track target.
A kind of target acquisition tracing system based on video, a kind of computer readable storage medium and base provided by the invention
In the target acquisition tracing equipment of video, with the above-mentioned target acquisition method for tracing based on video for identical inventive concept,
Beneficial effect having the same.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described.In all the appended drawings, similar element
Or part is generally identified by similar appended drawing reference.In attached drawing, each element or part might not be drawn according to actual ratio.
Fig. 1 is a kind of flow chart of the target acquisition method for tracing based on video provided by the invention.
Specific embodiment
It is described in detail below in conjunction with embodiment of the attached drawing to technical solution of the present invention.Following embodiment is only used for
Clearly illustrate technical solution of the present invention, therefore be intended only as example, and cannot be used as a limitation and limit protection of the invention
Range.
It should be noted that unless otherwise indicated, technical term or scientific term used in this application should be this hair
The ordinary meaning that bright one of ordinary skill in the art are understood.
The present invention provides a kind of target acquisition method for tracing, system, medium and equipment based on video.Below with reference to attached
Figure is illustrated the embodiment of the present invention.
First embodiment:
Referring to FIG. 1, Fig. 1 is a kind of target acquisition method for tracing based on video that the specific embodiment of the invention provides
Schematic diagram, a kind of target acquisition method for tracing based on video provided in this embodiment, comprising: adjustment supervision equipment turns to monitoring
Region；Initialization tracking is carried out to the monitor area；In initialization tracing process, judge whether target enters the monitoring
Region；If it is not, then continuing to carry out initialization tracking to the monitor area；If so, keeping track to the target；?
During keeping track, judge whether the target loses；If it is not, then continuing to keep track the target；If so,
It predicts and finds the target；If within a preset period of time, giving the target for change, then the target is continued persistently to chase after
Track；If within a preset period of time, not giving the target for change, then terminate to track.
Wherein, supervision equipment may include: turntable, (2) height that (1) can be rotated freely with pitch orientation in the horizontal direction
Clear focal length Varifocal zoom lens, four (3) high-definition digital camera, (4) Embedded real-time signal processing chip main components.Turning table control is seen
Direction, lens control observation visual angle are examined, camera generates digital video, and chip carries out signal processing to video, and controls turntable, mirror
The work of head, camera.
When adjusting supervision equipment steering monitor area, the turntable angle and lens focus of adjustable supervision equipment make
Supervision equipment turns to monitor area, then carries out initialization tracking to monitor area.In tracing process once confirmation target into
Enter monitor area, then starting keeps track program, keeps track to target, otherwise, current initialization is kept to track state.
During keeping track, if target is not lost, remain operational；If target is lost, start prediction and finder, finds mesh
Mark；If giving target for change, it is switched to and keeps track program, if not giving for change for a long time, terminates to track.The present invention in this way,
Infrared image is difficult to accurately to detect, the strong target of movement maneuverability has better track lock ability, can be reliable and stable
Track target in ground.
In a specific embodiment provided by the invention, the process for carrying out initialization tracking to monitor area can be used
The method of two frames initialization tracking or the initialization tracking of three frames.
The first, two frames initialize method for tracing, the specific steps are as follows:
Step 11): carrying out target detection to current frame image, obtains a kind of first detection result；
Wherein, the result of target detection may include the testing result of multiple targets, and the testing result of each target can be with
Include: target's center position (X, Y), shared region is wide high (W, H), and identification score S etc..
Step 12): previous a kind of second detection result is read.
When being detected to image, can be every the detection of several frames once, not necessarily each frame is all detected.
Step 13): bond area matching algorithm, to a kind of first detection result and a kind of second detection result
It is matched, obtains the first pairing score；
A kind of first detection result and a kind of second detection result need with clock synchronization by one kind the first detection knot
Each result of fruit and each result of a kind of second detection result are matched one by one, wherein each result refers to each mesh
Mark corresponding as a result, calculating the first pairing score according to following formula, formula is as follows:
Wherein, the footnote of a kind of first detection result is denoted as i, and the footnote of a kind of second detection result is denoted as j, ω1、ω2、
ω3It is weight coefficient, Mi-jIt is the score of " Region Matching Algorithm "；It skips if this or if previous testing result quantity is 0 and matches
It is right.
If there are three the detection results of target for a kind of first detection result, there are four the spies of target for a kind of second detection result
It surveys as a result, can then form 3*4=12 to pairing as a result, obtaining 12 pairing scores.
Step 14): judge whether the highest pairing score in the first pairing score is more than target gate；
If it is not, then the target does not enter into the monitor area；By a kind of first detection result covering described one
The second detection result of class, reads newest video frame, continues to execute step 11) to step 14)；
If so, determining that the target enters the monitor area；According to a kind of first detection result and described one
Center range difference and time difference in the second detection result of class calculate first object movement velocity；According to first mesh
Movement velocity is marked, the target is kept track.
If there are multiple pairing scores in the first pairing score, need to judge whether highest pairing score is more than target door
Limit, with this to determine whether there is target to enter monitor area.
When calculating first object movement velocity, the corresponding result of target acquisition twice of score can be matched according to highest
Center range difference and time difference calculate target speed, and keep track program according to target speed starting.Its
In, center range difference refer between the target's center position of the corresponding result of target acquisition twice of highest pairing score away from
Deviation.Time difference refers to the time difference of the corresponding target acquisition twice of highest pairing score.
Wherein, Region Matching Algorithm are as follows:
Step 1: being extracted respectively in the region testing result i and the region testing result j using feature detection algorithm
Characteristic point, and feature vector is calculated to each characteristic point；
Step 2: to each feature vector generated in the j of region, matching score is found in the feature vector that region i is generated
The pairing of the maximum constitutive characteristic vector；Match score calculation formula are as follows:
Wherein, FiFor each feature vector generated in the i of region；FjFor each feature vector generated in the j of region.
Step 3；Feature vector pairing is ranked up from big to small according to the matching score of feature vector pairing；
Step 4: taking the matching score of preceding 5 pairs of feature vectors pairing, average value is calculated, as Region Matching Algorithm score.
Second, three frames initialize method for tracing, the specific steps are as follows:
Step 21): target detection is carried out to current frame image, obtains two the first detection results of class；
To current frame image carry out target detection, the target acquisition of multiple targets can be obtained as a result, each target inspection
Surveying result includes: target's center position (X, Y), shared region is wide high (W, H), and the data such as identification score S.
Step 22): two the second detection results of class of target detection are carried out to picture frame twice before reading and two class thirds are visited
Survey result.
Step 23): two classes, first detection result, two the second detection results of class and two class third detection results are carried out
Pairing obtains the second pairing score.
For example, two the first detection results of class, there are three the detection result of target, there are four targets for two the second detection results of class
Detection result, there are five the detection results of target for two class third detection results, then are capable of forming 3*4*5=60 pairing and tie
Fruit obtains 60 pairing scores.
If having a detection result in two the first detection results of class, two the second detection results of class and two class third detection results
It is 0, then cannot matches.Continue to execute step 21)-step 23).
After reading preceding target acquisition result twice, the testing result that 1 target is respectively chosen from testing result three times is carried out
Pairing calculates the second pairing score according to following formula:
Wherein, Ai-j-k=exp (- | Wi-Wj|-|Hi-Hj|)+exp(-|Wk-Wj|-|Hk-Hj|)
Vi-j-k=exp (- | Xi+Xk-2Xj|)+exp(-|Yi+Yk-2Yj|)
Wherein, Ai-i-kIndicate the parameter of measurement target sizes consistency；Vi-i-kIt indicates to measure target speed consistency
Parameter.Indicate the parameter of measurement resemblance consistency.
Testing result footnote is denoted as i, j, k, ω ' respectively three times1、ω'2、ω'3It is weight coefficient.
Step 24): judge whether the highest pairing score in the second pairing score is more than target gate；If it is not, then
The target does not enter into the monitor area；Two classes, first detection result is covered previous two class second to visit
Survey as a result, will two classes, second detection result covering, the two classes third detection result, and read newest video frame, continue to hold
Row step 21) is to step 24)；If so, determining that the target enters the monitor area；According in target acquisition result three times
Center range difference and the time difference, calculate the second target speed, according to second target speed, to described
Target is kept track.
Target is confirmed if being more than confirmation target gate if the highest score in the second pairing score, according to detecting target three times
Center range difference and the time difference calculate target speed, and according to calculating speed starting keep track program；Otherwise
Two classes, first detection result is covered into previous two classes, second detection result, by two classes, second detection result
The two classes third detection result is covered, and reads newest video frame, continues to execute step 21) to step 24).
Wherein, center range difference and time difference are identical as the definition in two frames initialization method for tracing.Center
Range difference refers to the distance between the target's center position of the corresponding result of target acquisition three times of highest pairing score difference.Time difference
Refer to the time difference of the corresponding target acquisition three times of highest pairing score.
After confirmation target is tracked in initialization, target is kept track, it can be with are as follows: according to target speed, adjust
Turntable revolving speed in the whole supervision equipment, makes the target be maintained at the central area of the monitor area, the specific steps are as follows:
Step 31): target speed is converted into the horizontal rotation speed and pitching rotation speed of turntable.
Wherein, target speed can be first object movement velocity or the second target speed.
Step 32): according to the horizontal rotation speed and pitching rotation speed, the turntable revolving speed is adjusted.
The target speed that tracing program calculating will be initialized is converted into the horizontal rotation speed and pitching rotation of turntable
Speed, and by this adjustment turntable revolving speed.
Step 33): by initialization tracing process, highest matches the target acquisition knot of the corresponding last frame image of score
Fruit is recorded as historical information.
The detection data for initializing the corresponding last frame image of highest pairing score that tracing program calculates is recorded as
Historical information.
Step 34): target detection is carried out to current frame image, obtains the 4th target acquisition result.
Program is detected to current frame image performance objective, obtains the target acquisition result of multiple targets；The inspection of each target
Surveying result includes: target's center position (X, Y), shared region is wide high (W, H), and the data such as identification score S.
Step 35): reading the historical information, and the 4th target acquisition result is matched with the historical information
It is right, it calculates third and matches score.
Historical information is read, the historical information of each result of this target detection and previous confirmation is matched, is pressed
Third, which is calculated, according to following formula matches score:
ω"1exp(-|Wi-Wj|-|Hi-Hj|)+ω"2Si+ω"3Mi-j+ω"4exp(-|Xi-Xj|-|Yi-Yj|)
Wherein, this testing result footnote remembers that i, historical information footnote remember j, ω "1、ω"2、ω"3、ω"4It is weight system
System, Mi-jIt is the score of Region Matching Algorithm.
Step 36): judge whether the highest pairing score in the third pairing score is lower than target gate；If so,
Think that the target is temporarily lost, predict and finds the target；If not, then it is assumed that the target is not lost.
The step can be with are as follows:
If the 4th target acquisition result is 0, alternatively, the highest pairing score in third pairing score is lower than mesh
Mark thresholding, then it is assumed that the target is temporarily lost, and is predicted and is found the target；If the 4th target acquisition result is not 0,
And the highest pairing score in the third pairing score is not less than target gate, then it is assumed that the target is not lost.
Step 37): if target is not lost, the third is matched to the corresponding detection of highest pairing score in score
As a result it updates and arrives the historical information.
Step 38): according to the horizontal rotation speed of current goal imaging position and horizontal relative velocity adjustment turntable；According to
The pitching rotation speed of current goal imaging position and pitching relative velocity adjustment turntable, repeats step 32) to step 38).
It is as follows that turntable speed adjusts process:
Assuming that speed of related movement of the target in monitor area are as follows: (VX,VY)；
VX=Xi-Xj VY=Yi-Yj
According to the horizontal velocity of this confirmation target imaging position and horizontal relative velocity Vx adjustment turntable, wherein Vx is absolute
Value is less than threshold value and is considered static, and otherwise numerical value, which is positive, is considered to the right, and numerical value, which is negative, to be considered to the left；Speed threshold absolute value table
Show revolving speed size, positive to the right, negative direction to the left, as shown in table 1.
Table 1
| Horizontal relative velocity | Imaging position | Speed adjustment |
| To the left | Right side | It is constant |
| It is static | Right side | Add 1 grade |
| To the right | Right side | Add 2 grades |
| To the left | It is intermediate | It is constant |
| It is static | It is intermediate | It is constant |
| To the right | It is intermediate | It is constant |
| To the left | Left side | Subtract 2 grade |
| It is static | Left side | Subtract 1 grade |
| To the right | Left side | It is constant |
According to the horizontal velocity of this confirmation target imaging position and pitching relative velocity Vy adjustment turntable, wherein Vy is absolute
Value is less than threshold value and is considered static, and otherwise numerical value, which is positive, is considered that numerical value, which is negative, is considered downward upwards；Speed threshold absolute value table
Show revolving speed size, positive is upward, and negative direction is downward, as shown in table 2.
Table 2
| Pitching relative velocity | Imaging position | Speed adjustment |
| Downwards | Upside | It is constant |
| It is static | Upside | Add 1 grade |
| Upwards | Upside | Add 2 grades |
| Downwards | It is intermediate | It is constant |
| It is static | It is intermediate | It is constant |
| Upwards | It is intermediate | It is constant |
| Downwards | Downside | Subtract 2 grade |
| It is static | Downside | Subtract 1 grade |
| Upwards | Downside | It is constant |
In a specific embodiment provided by the invention, the prediction simultaneously finds the target, comprising: according to the mesh
The last imaging position of target and relative velocity adjust turntable revolving speed, and the last flight path along the target searches target.
Get rid of the monitoring visual field when being tracked target, or block, obscure etc. be imaged due to fail to be detected, then basis
Last imaging position and relative velocity adjust turntable speed, search target along last flight path.The specific side of adjustment of horizontal velocity
Formula is as shown in table 3.
Table 3
| Horizontal relative velocity | Imaging position | Speed adjustment |
| To the left | Right side | It is constant |
| It is static | Right side | After keeping present speed 300ms, every 300ms adds 1 grade |
| To the right | Right side | Immediately plus 2 grades, and every 300ms adds 2 grades |
| To the left | It is intermediate | After keeping present speed 300ms, every 300ms subtracts 1 grade |
| It is static | It is intermediate | It is constant |
| To the right | It is intermediate | After keeping present speed 300ms, every 300ms adds 1 grade |
| To the left | Left side | Subtract 2 immediately grade, and every 300ms subtracts 2 grade |
| It is static | Left side | After keeping present speed 300ms, every 300ms subtracts 1 grade |
| To the right | Left side | It is constant |
The specific adjustment mode of hoisting speed is as shown in table 4:
Table 4
If detecting target, it can judge whether that target is looked for according to the step 34) kept track in method to step 36)
It returns；Section does not give target for change yet when the preset time is exceeded, then it is assumed that target is lost for a long time, stops tracking.Wherein, preset time period can
To be 3 seconds.
More than, it is a kind of target acquisition method for tracing based on video provided by the invention.
Second embodiment:
It is corresponding based on inventive concept identical with a kind of above-mentioned target acquisition method for tracing based on video,
The target acquisition tracing system based on video that the embodiment of the invention also provides a kind of.Due to system embodiment it is substantially similar with it is square
Method embodiment, so describing fairly simple, the relevent part can refer to the partial explaination of embodiments of method.
A kind of target acquisition tracing system based on video provided by the invention, comprising: processing chip, turntable, camera lens and
Camera；The camera lens is arranged on the camera；The camera is fixed on the turntable；The turntable, camera lens and phase unit
At supervision equipment；The processing chip is all connected with the turntable, camera lens and camera；The processing chip is for controlling the prison
Depending on the working condition of equipment, target acquisition method for tracing of one of the first embodiment based on video is executed.
Wherein, processing chip can be set in the camera, can also be separately provided outside camera, and smart machine is arrived in setting
In, this is all within the scope of the present invention.
Wherein, turntable is the turntable that can be rotated freely in the horizontal direction with pitch orientation；Camera lens can be high definition focal length
Varifocal zoom lens；Camera can be high-definition digital camera；Processing chip can use Embedded real-time signal processing chip.
More than, it is a kind of target acquisition tracing system based on video provided by the invention.
3rd embodiment:
It is corresponding based on inventive concept identical with a kind of above-mentioned target acquisition method for tracing based on video,
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, and the program is processed
Device realizes a kind of above-mentioned target acquisition method for tracing based on video when executing.
Fourth embodiment:
It is corresponding based on inventive concept identical with a kind of above-mentioned target acquisition method for tracing based on video,
The embodiment of the invention also provides a kind of target acquisition tracing equipment based on video, comprising: memory, processor and be stored in
On memory and the computer program that can run on a processor, the processor realize a kind of above-mentioned base when executing described program
In the target acquisition method for tracing of video.
In specification of the invention, numerous specific details are set forth.It is to be appreciated, however, that the embodiment of the present invention can be with
It practices without these specific details.In some instances, well known method, structure and skill is not been shown in detail
Art, so as not to obscure the understanding of this specification.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme should all cover within the scope of the claims and the description of the invention.
Claims (10)
1. a kind of target acquisition method for tracing based on video characterized by comprising
It adjusts supervision equipment and turns to monitor area；
Initialization tracking is carried out to the monitor area；In initialization tracing process, judge whether target enters the monitoring
Region；If it is not, then continuing to carry out initialization tracking to the monitor area；If so, keeping track to the target；
During keeping track, judge whether the target loses；If it is not, then continuing to keep track the target；
If so, predicting and finding the target；
If within a preset period of time, giving the target for change, then the target is continued to keep track；If in preset time period
It is interior, it does not give the target for change, then terminates to track.
2. the method according to claim 1, wherein the adjustment supervision equipment turns to monitor area, comprising:
The turntable angle and lens focus for adjusting supervision equipment make supervision equipment turn to monitor area.
3. being wrapped the method according to claim 1, wherein described carry out initialization tracking to the monitor area
It includes:
Method for tracing is initialized using two frames, initialization tracking is carried out to the monitor area, the specific steps are as follows:
Step 11): carrying out target detection to current frame image, obtains a kind of first detection result；
Step 12): the second detection result of one kind of target detection is once carried out before reading to picture frame；
Step 13): bond area matching algorithm carries out a kind of first detection result and a kind of second detection result
Pairing obtains the first pairing score；
Step 14): judge whether the highest pairing score in the first pairing score is more than target gate；
If it is not, then the target does not enter into the monitor area；
A kind of first detection result is covered into a kind of second detection result, newest video frame is read, continues to execute step
It is rapid 11) to step 14)；
If so, determining that the target enters the monitor area；
According in a kind of first detection result and a kind of second detection result center range difference and the time difference,
Calculate first object movement velocity；According to the first object movement velocity, the target is kept track.
4. being wrapped the method according to claim 1, wherein described carry out initialization tracking to the monitor area
It includes:
Method for tracing is initialized using three frames, initialization tracking is carried out to the monitor area, the specific steps are as follows:
Step 21): target detection is carried out to current frame image, obtains two the first detection results of class；
Step 22): two the second detection results of class and two class thirds detection knot of target detection are carried out before reading to picture frame twice
Fruit；
Step 23): two classes, first detection result, two the second detection results of class and two class third detection results are matched
It is right, obtain the second pairing score；
Step 24): judge whether the highest pairing score in the second pairing score is more than target gate；
If it is not, then the target does not enter into the monitor area；
Two classes, first detection result is covered into previous two classes, second detection result, two class second is detected
As a result the two classes third detection result is covered, and reads newest video frame, continues to execute step 21) to step 24)；
If so, determining that the target enters the monitor area；
According in target acquisition result three times center range difference and the time difference, calculate the second target speed, according to
Second target speed, keeps track the target.
5. the method according to claim 3 or 4, which is characterized in that described to keep track to the target, comprising:
According to target speed, the turntable revolving speed in the supervision equipment is adjusted, the target is made to be maintained at the surveillance zone
The central area in domain, the specific steps are as follows:
Step 31): target speed is converted into the horizontal rotation speed and pitching rotation speed of turntable；
Step 32): according to the horizontal rotation speed and pitching rotation speed, the turntable revolving speed is adjusted；
Step 33): by initialization tracing process, highest matches the target acquisition result note of the corresponding last frame image of score
Record is historical information；
Step 34): target detection is carried out to current frame image, obtains the 4th target acquisition result；
Step 35): reading the historical information, and the 4th target acquisition result is matched with the historical information, meter
It calculates third and matches score；
Step 36): judge whether the highest pairing score in the third pairing score is lower than target gate；
If so, thinking that the target is temporarily lost, predicting and finding the target；
If not, then it is assumed that the target is not lost；
Step 37): if target is not lost, the third is matched to the corresponding detection result of highest pairing score in score
Update the historical information；
Step 38): according to the horizontal rotation speed of current goal imaging position and horizontal relative velocity adjustment turntable；According to current
The pitching rotation speed of target imaging position and pitching relative velocity adjustment turntable, repeats step 32) to step 38).
6. according to the method described in claim 5, it is characterized in that, step 36) can be with are as follows:
If the 4th target acquisition result is 0, alternatively, third pairing score is lower than target gate, then it is assumed that the mesh
Mark is temporarily lost, and predicts and finds the target；
If the 4th target acquisition result is not 0, and third pairing score is not less than target gate, then it is assumed that the mesh
Mark is not lost.
7. according to the method described in claim 5, it is characterized in that, described predict and find the target, comprising:
According to the last imaging position of the target and relative velocity, turntable revolving speed is adjusted, along the last flight rail of the target
Mark searches target.
8. a kind of target acquisition tracing system based on video characterized by comprising processing chip, turntable, camera lens and phase
Machine；
The camera lens is arranged on the camera；The camera is fixed on the turntable；The turntable, camera lens and camera composition
Supervision equipment；
The processing chip is all connected with the turntable, camera lens and camera；
The processing chip is used to control the working condition of the supervision equipment, and perform claim requires described in 1 to 7 any one
Target acquisition method for tracing based on video.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
Method described in one of claim 1-7 is realized when row.
10. a kind of target acquisition tracing equipment based on video, comprising: memory, processor and storage are on a memory and can
The computer program run on a processor, which is characterized in that the processor realizes claim 1-7 when executing described program
One of described in method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811191956.9A CN109302562B (en) | 2018-10-12 | 2018-10-12 | Video-based target detection tracking method, system, medium, and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811191956.9A CN109302562B (en) | 2018-10-12 | 2018-10-12 | Video-based target detection tracking method, system, medium, and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109302562A true CN109302562A (en) | 2019-02-01 |
| CN109302562B CN109302562B (en) | 2020-09-29 |
Family
ID=65162576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811191956.9A Active CN109302562B (en) | 2018-10-12 | 2018-10-12 | Video-based target detection tracking method, system, medium, and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109302562B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751836A (en) * | 2019-09-26 | 2020-02-04 | 武汉光庭信息技术股份有限公司 | Vehicle driving early warning method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231755A (en) * | 2007-01-25 | 2008-07-30 | 上海遥薇实业有限公司 | Moving target tracking and quantity statistics method |
| CN103024344A (en) * | 2011-09-20 | 2013-04-03 | 佳都新太科技股份有限公司 | Automatic PTZ (Pan/Tilt/Zoom) target tracking method based on particle filter |
| US20160198130A1 (en) * | 2015-01-07 | 2016-07-07 | Vivotek Inc. | Surveillance method and surveillance system |
| CN106295695A (en) * | 2016-08-08 | 2017-01-04 | 中国民用航空总局第二研究所 | A kind of takeoff and landing process automatic tracing image pickup method and device |
| CN107292907A (en) * | 2017-07-14 | 2017-10-24 | 灵动科技（北京）有限公司 | A kind of method to following target to be positioned and follow equipment |
| CN107783103A (en) * | 2017-09-26 | 2018-03-09 | 武汉三江中电科技有限责任公司 | A kind of flying bird tracking intelligent method of lightweight with self-learning function |
-
2018
- 2018-10-12 CN CN201811191956.9A patent/CN109302562B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231755A (en) * | 2007-01-25 | 2008-07-30 | 上海遥薇实业有限公司 | Moving target tracking and quantity statistics method |
| CN103024344A (en) * | 2011-09-20 | 2013-04-03 | 佳都新太科技股份有限公司 | Automatic PTZ (Pan/Tilt/Zoom) target tracking method based on particle filter |
| US20160198130A1 (en) * | 2015-01-07 | 2016-07-07 | Vivotek Inc. | Surveillance method and surveillance system |
| CN106295695A (en) * | 2016-08-08 | 2017-01-04 | 中国民用航空总局第二研究所 | A kind of takeoff and landing process automatic tracing image pickup method and device |
| CN107292907A (en) * | 2017-07-14 | 2017-10-24 | 灵动科技（北京）有限公司 | A kind of method to following target to be positioned and follow equipment |
| CN107783103A (en) * | 2017-09-26 | 2018-03-09 | 武汉三江中电科技有限责任公司 | A kind of flying bird tracking intelligent method of lightweight with self-learning function |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751836A (en) * | 2019-09-26 | 2020-02-04 | 武汉光庭信息技术股份有限公司 | Vehicle driving early warning method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109302562B (en) | 2020-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | A novel real-time moving target tracking and path planning system for a quadrotor UAV in unknown unstructured outdoor scenes | |
| CN108010067B (en) | A kind of visual target tracking method based on combination determination strategy | |
| CN103149939B (en) | A kind of unmanned plane dynamic target tracking of view-based access control model and localization method | |
| US9696404B1 (en) | Real-time camera tracking system using optical flow feature points | |
| US20220024549A1 (en) | System and method for measuring the distance to an object in water | |
| Rozantsev et al. | Flight dynamics-based recovery of a UAV trajectory using ground cameras | |
| CN101511022A (en) | Method for implementing machine-carried video compression and target tracking unitedly | |
| CN102447835A (en) | Non-blind-area multi-target cooperative tracking method and system | |
| CN106875421A (en) | A kind of multi-object tracking method and device | |
| CN104616322A (en) | Onboard infrared target image identifying and tracking method and device | |
| US20180136306A1 (en) | Device and method for measuring flight data of flying objects using high speed video camera and computer readable recording medium having program for performing the same | |
| CN110009060A (en) | A kind of robustness long-term follow method based on correlation filtering and target detection | |
| Wu et al. | Multivehicle object tracking in satellite video enhanced by slow features and motion features | |
| Zachariadis et al. | 2D visual tracking for sports UAV cinematography applications | |
| CN104613928A (en) | Automatic tracking and air measurement method for optical pilot balloon theodolite | |
| Shalnov et al. | Convolutional neural network for camera pose estimation from object detections | |
| Hamdi et al. | Drotrack: High-speed drone-based object tracking under uncertainty | |
| CN109302562A (en) | Target acquisition method for tracing, system, medium and equipment based on video | |
| Diamantas et al. | Depth estimation in still images and videos using a motionless monocular camera | |
| Yang et al. | Improvement of honey bee tracking on 2D video with hough transform and Kalman filter | |
| Qi et al. | Alpine skiing tracking method based on deep learning and correlation filter | |
| KR101032098B1 (en) | Stand-alone environment traffic detecting system using thermal infra-red | |
| CN109977796A (en) | Trail current detection method and device | |
| CN107093187A (en) | The measuring method and device of a kind of unmanned plane during flying speed | |
| Xie et al. | A deep-learning-based fusion approach for global cyclone detection using multiple remote sensing data |
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 |