CN109525780A - A kind of video interlink camera lens Zooming method - Google Patents
A kind of video interlink camera lens Zooming method Download PDFInfo
- Publication number
- CN109525780A CN109525780A CN201811582680.7A CN201811582680A CN109525780A CN 109525780 A CN109525780 A CN 109525780A CN 201811582680 A CN201811582680 A CN 201811582680A CN 109525780 A CN109525780 A CN 109525780A
- Authority
- CN
- China
- Prior art keywords
- beta
- alpha
- field angle
- signal
- linkage
- 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
- H04N23/67—Focus control based on electronic image sensor signals
-
- 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
-
- 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
- H04N23/69—Control of means for changing angle of the field of view, e.g. optical zoom objectives or electronic zooming
Abstract
The present invention discloses a kind of video interlink camera lens Zooming method, using table look-up, machine learning, the methods of neural network, work as the corresponding field angle tolerance band [alpha of front direction according to historical data estimation prediction, beta], in same direction same type target, the field angle tolerance band that prediction technique provides is varied less for the prediction technique guarantee;In actual monitored occasion, the value of beta is generally ten times of alpha or so, i.e. target sizes can clearly see target clearly from video pictures are filled up to 1/10th size of picture is accounted for;Camera lens current field angle theta falls in [alpha, beta] it is Great possibility in section, it is when field angle is fallen in the section, do not execute zoom action, this patent method finally realizes camera lens during linking final goal, the zoom action frequency of occurrences is reduced, zoom mechanism service life is extended.
Description
Technical field
The present invention relates to a kind of video interlink camera lens Zooming methods, belong to information technology field.
Background technique
In monitor camera cluster, after target alert video camera finds early warning target, goal verification video camera needs basis
Target position and size constantly change its direction and lens focus.Common monitoring camera lens zoom device projected life is five
Ten thousand times, the camera lens service life can effectively be extended by reducing zoom action.
There are two types of main methods to reduce zoom action at present.
One, focal length is segmented
Several focal length values are set, and by configuring the corresponding relationship for establishing target sizes and focal length.Linkage camera is according to early warning
Target sizes, by computation of table lookup and the immediate setting value of target sizes, if corresponding focal length value with it is current consistent, no
Do zoom operation.
The program has that one is difficult to overcome.In actual application, especially at night, due to environmental disturbances,
In tracking and early warning object procedure, the target sizes detected can frequently occur large jump, cause the zoom frequency to improve limited.
Two, ABC three-stage process
According to monitoring area apart from feature, monitoring area is divided into tri- sections of A, B, C, focal length value is set as short focus appropriate at the end A
Away from C-terminal is set as long-focus appropriate.When target enters a-quadrant, zoom to short focus;When target enters the region C, zoom is extremely
Long-focus;When target enters B area, focal length is constant.
The strategy works well in practical applications, but there are problems that being difficult to overcome as follows:
1, when burnt sections more than two for complex scene setting, the specific implementation process of setting process and strategy is complicated.
2, target type of different sizes, such as pedestrian target and excavator both target types are directed to, according to target institute
Determine that lens focus is not to be able to satisfy actual demand in region.
Already present related patents retrieval is as follows:
Patent " laser and visible lightseeking missile linkage control device " (application number: 201721008387.0) discloses a kind of sharp
Light and visible light autotracker.The patent is not related to lens zoom optimization problem.
Patent " a kind of video camera varifocal sensing mechanism and holder camera assembly " (application number: 201610949949.5)
Disclose a kind of video camera varifocal sensing mechanism and holder camera assembly.The patent is not related to lens zoom optimization problem.
Patent " a kind of video camera visible area call method " (application number: 201610605261.5) discloses a kind of camera shooting
The method of machine and geography information linkage, is not related to the zoom optimization problem of object tracking process.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of video interlink video camera Zooming methods, match not increasing system
Under the premise of setting with maintenance load, by adaptive learning strategy, the zoom reduced in camera lens tracking object procedure is dynamic
Make the frequency, extends relevant device service life.
In order to solve the technical problem, the technical solution adopted by the present invention is that: a kind of video interlink camera lens change
Burnt method, comprising the following steps: S01), source of early warning unit to linkage camera unit issue linkage signal (x, y, z,
Alpha, beta), wherein x is linkage target level coordinate, and y is the linkage vertical coordinate of target, and z is target type, and alpha is
The minimum field angle that this linkage allows, beta are the maximum field of view angles that this linkage allows;S02), linkage camera unit will
Linkage signal (x, y, z, alpha, beta) is sent to signal smoothing processing unit;S03), signal smoothing processing unit according to
Output alpha ', beta ' are smoothed to alpha, beta signal of input according to the value of x, y, z;S04), linkage control list
Member takes out target type from signal smoothing processing unit and is equal to z, and target position is the minimum allowable field angle for (x, y)
Alpha ' and maximum allowable field angle beta ', alpha ' are that the target type recorded in signal smoothing processing unit is equal to the position of z
Set the average value with the minimum allowable field angle alpha of (x, y) immediate n item record;Beta ' is signal smoothing processing unit
The target type of middle record is averaged equal to the maximum allowable field angle beta's of the position of z and (x, y) immediate n item record
Value;S05), linkage control unit issues turn signal (x, y) to cradle head control unit;S06), linkage control unit is from camera lens
Control unit reads current lens field angle theta, if alpha '≤theta≤beta ', does not execute zoom action;Such as
Fruit theta≤alpha ', linkage control unit issue zoom signal to lens control unit, and zoom field angle is alpha ';
If beta '≤theta, linkage control unit issues zoom signal to lens control unit, and zoom field angle is beta '.
Further, step S03 signal smoothing processing unit take out target type be equal to z, target position be for (x,
Y) minimum allowable field angle alpha ' and maximum allowable field angle beta ', alpha ' is recorded in signal smoothing processing unit
Position of the target type equal to z and (x, y) immediate n item record minimum allowable field angle alpha average value;
Beta ' is that the target type recorded in signal smoothing processing unit records most equal to the position of z and (x, y) immediate n item
The big average value for allowing field angle beta.
Further, the signal smoothing processing unit of step S03 is remembered using signal of the artificial nerve network model to submission
Record executes on-line training, and the input of artificial nerve network model training sample is (x, y, z), output label be (alpha,
Beta), the signal smoothing processing unit of step S03 determines target position according to input signal (x, y, z, alpha, beta)
The minimum allowable field angle alpha ' of (x, y) and maximum allowable field angle beta '.
Further, signal smoothing processing unit using circulation coverage mode, data be filled with after from starting position recycle to
After cover.
Further, signal smoothing processing unit uses random coverage mode, is recording at random when new data being written every time
One position of Unit selection overrides legacy data.
Further, n=3.
Beneficial effects of the present invention: the present invention has recorded field angle license of the video camera at azimuth (x [i], y [i])
Range (alpha [i], beta [i]);For any direction (x, y), this patent method using table look-up, machine learning, nerve net
The methods of network works as the corresponding field angle tolerance band [alpha, beta] of front direction, the prediction according to historical data estimation prediction
In same direction same type target, the field angle tolerance band that prediction technique provides is varied less for method guarantee;In practical prison
Occasion is controlled, the value of beta is generally ten times of alpha or so, i.e. target sizes are big to picture 1/10th is accounted for from video pictures are filled up
It is small, it can clearly see target clearly;It is general that camera lens current field angle theta, which is fallen in the section [alpha, beta],
Rate event is not execute zoom action when field angle is fallen in the section, and this patent method finally realizes that camera lens is final in linkage
In object procedure, the zoom action frequency of occurrences is reduced, extends zoom mechanism service life.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Specific embodiment
The present invention is further illustrated in the following with reference to the drawings and specific embodiments.
Embodiment 1
The present embodiment discloses a kind of video interlink camera lens Zooming method, as shown in Figure 1, comprising the following steps:
S01), source of early warning unit issues linkage signal (x, y, z, alpha, beta) to linkage camera unit, wherein x
It is linkage target level coordinate, y is linkage target value coordinate, and z is target type, and alpha is the minimum that this linkage allows
Field angle, beta are the maximum field of view angles that this linkage allows;
S02), linkage camera unit submits linkage signal (x, y, z, alpha, beta) to signal smoothing processing unit;
In the present embodiment, signal smoothing processing unit be filled with using circulation coverage mode storing data, data after from starting position
Circulation covers backward;
Certainly, signal smoothing processing unit can also use random coverage mode storing data, be written every time random when new data
A position is selected to override legacy data in recording unit.
S03), there are random error, value of the signal smoothing processing unit according to x, y, z by the alpha in linkage signal, beta
Output alpha ', beta ' are smoothed to alpha, beta signal of input, to eliminate this random error;Specifically
Implementation are as follows: signal smoothing processing unit takes out target type and is equal to z, and target position is to regard for the minimum allowable of (x, y)
Rink corner alpha ' and maximum allowable field angle beta ', alpha ' are that the target type recorded in signal smoothing processing unit is equal to z
Position and (x, y) immediate n item record minimum allowable field angle alpha average value;Beta ' is signal smoothing processing
The target type recorded in unit is flat equal to the maximum allowable field angle beta's of the position of z and (x, y) immediate n item record
Mean value.
Specifically, can be using artificial nerve network models such as BP neural network, RBF radial base neural nets to submission
Signal record executes on-line training, and the input of artificial nerve network model training sample is (x, y, z), and output label is
The signal smoothing processing unit of (alpha, beta), step S03 are determined according to input signal (x, y, z, alpha, beta)
The minimum allowable field angle alpha ' of target position (x, y) and maximum allowable field angle beta '.
S04), linkage control unit from signal smoothing processing unit take out target type be equal to z, target position be (x,
Y) minimum allowable field angle alpha ' and maximum allowable field angle beta ';
Wherein alpha ' is position of the target type equal to z recorded in signal smoothing processing unit and (x, y) immediate n item
The average value of the minimum allowable field angle alpha of record;
The target type recorded in beta ' signal smoothing processing unit is recorded equal to the position of z and (x, y) immediate n item
The average value of maximum allowable field angle beta, n=3.
S05), linkage control unit issues turn signal (x, y) to cradle head control unit;
S06), linkage control unit lens control unit reads current lens field angle theta, if alpha '≤theta
≤ beta ', does not execute zoom action;
If theta≤alpha ', linkage control unit issues zoom signal to lens control unit, (new_theta=
Alpha '), zoom field angle is alpha ';
If beta '≤theta, linkage control unit to lens control unit issue zoom signal (new_theta=
Beta '), zoom field angle is beta '.
The present invention have recorded video camera at azimuth (x [i], y [i]) field angle tolerance band (alpha [i],
beta[i]);For any direction (x, y), this patent method using table look-up, machine learning, the methods of neural network, according to going through
The corresponding field angle tolerance band [alpha, beta] of front direction is worked as in the estimation prediction of history data, and the prediction technique guarantee is same
Direction same type target, the field angle tolerance band that prediction technique provides vary less;In actual monitored occasion, the value of beta
Generally ten times of alpha or so, i.e. target sizes can clearly be seen from video pictures are filled up to 1/10th size of picture is accounted for
Clear target;It is Great possibility that camera lens current field angle theta, which is fallen in the section [alpha, beta], works as field angle
Falling in the section is not execute zoom action, and this patent method finally realizes that camera lens during linking final goal, reduces
The zoom action frequency of occurrences, extends zoom mechanism service life.
Described above is only basic principle and preferred embodiment of the invention, and those skilled in the art do according to the present invention
Improvement and replacement out, belong to the scope of protection of the present invention.
Claims (6)
1. a kind of video interlink camera lens Zooming method, it is characterised in that: the following steps are included: S01), source of early warning list
Member issues linkage signal (x, y, z, alpha, beta) to linkage camera unit, and wherein x is linkage target level coordinate,
Y is the linkage vertical coordinate of target, and z is target type, and alpha is the minimum field angle that this linkage allows, and beta is this connection
The dynamic maximum field of view angle allowed;S02), linkage signal (x, y, z, alpha, beta) is sent to by linkage camera unit
Signal smoothing processing unit;S03), signal smoothing processing unit carries out alpha, beta signal of input according to the value of x, y, z
Smoothing processing exports alpha ', beta ';S05), linkage control unit issues turn signal (x, y) to cradle head control unit;
S06), linkage control unit reads current lens field angle theta from lens control unit, if alpha '≤theta≤
Beta ' does not execute zoom action;If theta≤alpha ', linkage control unit issues zoom letter to lens control unit
Number, zoom field angle is alpha ';If beta '≤theta, linkage control unit issues zoom letter to lens control unit
Number, zoom field angle is beta '.
2. video interlink camera lens Zooming method according to claim 1, it is characterised in that: the signal of step S03
Smoothing processing unit takes out target type and is equal to z, and target position is for the minimum allowable field angle alpha ' for (x, y) and most
It is big that field angle beta ', alpha ' is allowed to be position of the target type equal to z recorded in signal smoothing processing unit and (x, y)
The average value of the minimum allowable field angle alpha of immediate n item record;Beta ' is recorded in signal smoothing processing unit
Average value of the target type equal to the position of z and the maximum allowable field angle beta of (x, y) immediate n item record.
3. video interlink camera lens Zooming method according to claim 1, it is characterised in that: the signal of step S03
Smoothing processing unit executes on-line training, artificial nerve network model using signal record of the artificial nerve network model to submission
The input of training sample is (x, y, z), and output label is (alpha, beta), the signal smoothing processing unit of step S03 according to
The minimum allowable field angle alpha ' and most of target position (x, y) is determined according to input signal (x, y, z, alpha, beta)
It is big to allow field angle beta '.
4. video interlink camera lens Zooming method according to claim 1, it is characterised in that: signal smoothing processing is single
Using circulation coverage mode, data cover after being filled with from starting position circulation member backward.
5. video interlink camera lens Zooming method according to claim 1, it is characterised in that: signal smoothing processing is single
Member uses random coverage mode, selects a position to override legacy data in recording unit at random when new data being written every time.
6. video interlink camera lens Zooming method according to claim 2, it is characterised in that: n=3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811582680.7A CN109525780B (en) | 2018-12-24 | 2018-12-24 | Video linkage camera lens zooming method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811582680.7A CN109525780B (en) | 2018-12-24 | 2018-12-24 | Video linkage camera lens zooming method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109525780A true CN109525780A (en) | 2019-03-26 |
CN109525780B CN109525780B (en) | 2020-08-21 |
Family
ID=65795804
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811582680.7A Active CN109525780B (en) | 2018-12-24 | 2018-12-24 | Video linkage camera lens zooming method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109525780B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110942579A (en) * | 2019-12-02 | 2020-03-31 | 山东神戎电子股份有限公司 | Unattended low-power multispectral night vision alarm system |
CN113315902A (en) * | 2020-02-26 | 2021-08-27 | 深圳英飞拓科技股份有限公司 | Lens stretching control signal hedging optimization method under high-speed dome camera |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090109299A1 (en) * | 2007-10-31 | 2009-04-30 | Hitachi, Ltd. | Zoom camera with manual focus function |
CN101860732A (en) * | 2010-06-04 | 2010-10-13 | 天津市亚安科技电子有限公司 | Method of controlling holder camera to automatically track target |
CN102045509A (en) * | 2010-12-28 | 2011-05-04 | 天津市亚安科技电子有限公司 | Method and device for linking zoom illumination device with zoom camera |
CN103795909A (en) * | 2012-10-29 | 2014-05-14 | 株式会社日立制作所 | Shooting optimization device, image-pickup device and shooting optimization method |
CN104284090A (en) * | 2014-09-28 | 2015-01-14 | 天津市亚安科技股份有限公司 | Method for linking zoom lighting device with camera with zoom lens |
CN104822019A (en) * | 2015-03-31 | 2015-08-05 | 深圳市莫孚康技术有限公司 | Method for calculating camera field angle |
CN105141828A (en) * | 2015-07-08 | 2015-12-09 | 成都西可科技有限公司 | Method for carrying out recording of motion camera by automatically adjusting view angle after locking scene |
CN105763795A (en) * | 2016-03-01 | 2016-07-13 | 苏州科达科技股份有限公司 | Focusing method and apparatus, cameras and camera system |
CN106210643A (en) * | 2016-07-29 | 2016-12-07 | 林玉峰 | A kind of video camera viewing area call method |
CN106878679A (en) * | 2017-02-24 | 2017-06-20 | 深圳英飞拓科技股份有限公司 | Intelligent monitoring linked system and control method |
CN107211090A (en) * | 2015-02-02 | 2017-09-26 | 富士胶片株式会社 | Operation device, tracking system, operating method and program |
CN108171723A (en) * | 2017-12-22 | 2018-06-15 | 湖南源信光电科技股份有限公司 | Based on more focal length lens of Vibe and BP neural network algorithm linkage imaging camera machine system |
-
2018
- 2018-12-24 CN CN201811582680.7A patent/CN109525780B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090109299A1 (en) * | 2007-10-31 | 2009-04-30 | Hitachi, Ltd. | Zoom camera with manual focus function |
CN101860732A (en) * | 2010-06-04 | 2010-10-13 | 天津市亚安科技电子有限公司 | Method of controlling holder camera to automatically track target |
CN102045509A (en) * | 2010-12-28 | 2011-05-04 | 天津市亚安科技电子有限公司 | Method and device for linking zoom illumination device with zoom camera |
CN103795909A (en) * | 2012-10-29 | 2014-05-14 | 株式会社日立制作所 | Shooting optimization device, image-pickup device and shooting optimization method |
CN104284090A (en) * | 2014-09-28 | 2015-01-14 | 天津市亚安科技股份有限公司 | Method for linking zoom lighting device with camera with zoom lens |
CN107211090A (en) * | 2015-02-02 | 2017-09-26 | 富士胶片株式会社 | Operation device, tracking system, operating method and program |
CN104822019A (en) * | 2015-03-31 | 2015-08-05 | 深圳市莫孚康技术有限公司 | Method for calculating camera field angle |
CN105141828A (en) * | 2015-07-08 | 2015-12-09 | 成都西可科技有限公司 | Method for carrying out recording of motion camera by automatically adjusting view angle after locking scene |
CN105763795A (en) * | 2016-03-01 | 2016-07-13 | 苏州科达科技股份有限公司 | Focusing method and apparatus, cameras and camera system |
CN106210643A (en) * | 2016-07-29 | 2016-12-07 | 林玉峰 | A kind of video camera viewing area call method |
CN106878679A (en) * | 2017-02-24 | 2017-06-20 | 深圳英飞拓科技股份有限公司 | Intelligent monitoring linked system and control method |
CN108171723A (en) * | 2017-12-22 | 2018-06-15 | 湖南源信光电科技股份有限公司 | Based on more focal length lens of Vibe and BP neural network algorithm linkage imaging camera machine system |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110942579A (en) * | 2019-12-02 | 2020-03-31 | 山东神戎电子股份有限公司 | Unattended low-power multispectral night vision alarm system |
CN113315902A (en) * | 2020-02-26 | 2021-08-27 | 深圳英飞拓科技股份有限公司 | Lens stretching control signal hedging optimization method under high-speed dome camera |
CN113315902B (en) * | 2020-02-26 | 2022-12-20 | 深圳英飞拓科技股份有限公司 | Lens stretching control signal hedging optimization method under high-speed dome camera |
Also Published As
Publication number | Publication date |
---|---|
CN109525780B (en) | 2020-08-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11380232B2 (en) | Display screen quality detection method, apparatus, electronic device and storage medium | |
CN107368776B (en) | Vehicle loss assessment image acquisition method and device, server and terminal equipment | |
US6924832B1 (en) | Method, apparatus & computer program product for tracking objects in a warped video image | |
JP2021503414A (en) | Intelligent driving control methods and devices, electronics, programs and media | |
WO2020034896A1 (en) | Method and apparatus for image processing, and mobile terminal | |
JP4279790B2 (en) | Car number recognition device | |
JP5400195B2 (en) | Method and camera for determining image adjustment parameters | |
CN111246091B (en) | Dynamic automatic exposure control method and device and electronic equipment | |
CN103716535A (en) | Method for switching photographing mode, and electronic device | |
CN105654508B (en) | Monitor video method for tracking moving target and system based on adaptive background segmentation | |
CN109525780A (en) | A kind of video interlink camera lens Zooming method | |
CN113840076B (en) | Shooting zooming method, electronic equipment and storage medium | |
CN111083388A (en) | Light supplement lamp control method and device, electronic equipment and storage medium | |
CN106791341A (en) | Image pickup method and device based on dual camera | |
CN109326125B (en) | Picture quality diagnosis system and method based on embedded system | |
CN115174861B (en) | Method and device for automatically tracking moving target by holder camera | |
CN109151257A (en) | A kind of method and video camera of image procossing | |
CN103024344A (en) | Automatic PTZ (Pan/Tilt/Zoom) target tracking method based on particle filter | |
CN202488602U (en) | All-dimensional vision detection analysis prewarning system based on fisheye correction technology | |
CN114359351A (en) | Target tracking method, system, device and equipment | |
US20230005239A1 (en) | Image capturing method and device, apparatus, and storage medium | |
CN113542695B (en) | Full-automatic regional monitoring system for internet of things | |
CN104767934A (en) | Information processing method and electronic equipment | |
CN205092923U (en) | Surveillance camera equipment | |
CN111797796A (en) | Construction method of parking specification detection model, parking specification detection method, system, terminal and medium |
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 |