CN110278382A

CN110278382A - A kind of focus method, device, electronic equipment and storage medium

Info

Publication number: CN110278382A
Application number: CN201910662640.1A
Authority: CN
Inventors: 陈天钧; 沈忱; 潘润发; 李准; 陈明珠
Original assignee: Zhejiang Dahua Technology Co Ltd
Current assignee: Zhejiang Dahua Technology Co Ltd
Priority date: 2019-07-22
Filing date: 2019-07-22
Publication date: 2019-09-24
Anticipated expiration: 2039-07-22
Also published as: CN110278382B

Abstract

The invention discloses a kind of focus method, device, electronic equipment and storage mediums, ball machine is according to the second coordinate information of target object and the third coordinate information of each identification point, the determining and nearest target identification point of target object space length, the corresponding relationship of each identification point and focal length is pre-saved in ball machine, after determining target identification point, the corresponding focal length of target object is determined according to initial focal length and preset focusing step-length using the corresponding focal length of target identification point as initial focal length.Because the object distance of the target identification point and target object determined is closest, the focusing focal length of target identification point and target object is closest, therefore carries out focusing on the basis of target object corresponding initial focal length and may be implemented quickly to focus.Focus program in compared to the prior art, substantially reduces focal time, improves candid photograph efficiency.

Description

A kind of focus method, device, electronic equipment and storage medium

Technical field

The present invention relates to technical field of image processing more particularly to a kind of focus method, device, electronic equipment and storage to be situated between Matter.

Background technique

With the fast development of safety-security area, the form of monitoring device is also more and more diversified, in combination with artificial intelligence After the such application of (Artificial Intelligence, AI) technology, especially recognition of face, rifle ball interlocking equipment is obtained Extensive use is arrived.Such interlocking equipment generally consists of two parts, and fixed-focus gunlock recognizes in picture for global observation When there is target object, the coordinate information of target object is sent to zoom ball machine, zoom ball machine is determined for carrying out coordinate conversion Then target object positions target object, after the completion of positioning, focus and capture.Therefore, during intelligent-tracking, It is very important for zoom ball machine focusing speed, directly affects candid photograph efficiency.

At present in such interlocking equipment, ball machine is general arbitrarily to choose an initial focal length when focusing, from initial coke Away from beginning, according to preset step-size change focal length, and one posterior focal distance of every change, one frame image of acquisition simultaneously determine the clear of image Evaluation of estimate is spent, when the clarity evaluation of estimate of image is met the requirements, it is believed that focus and complete.In many cases, it needs by multiple Focussing, can just meet the requirements, therefore, at present ball machine focus the problem is that, focal time is long, cause capture efficiency It is low.

Summary of the invention

The embodiment of the invention provides a kind of focus method, device, electronic equipment and storage mediums, to solve existing skill Focal time is long in art, leads to the problem of capturing low efficiency.

The embodiment of the invention provides a kind of focus methods, which comprises

According to first coordinate information of the target object in the first image that gunlock acquires, determine the target object in ball The second coordinate information in second image of machine acquisition；

Believed according to the third coordinate information of each identification point in second image pre-saved and second coordinate Breath, the determining and nearest target identification point of the target object space length；Wherein, the identification point is in second image The corresponding pixel logo point of immovable object；

According to the corresponding relationship of each identification point and focal length that pre-save, by the corresponding coke of the target identification point Away from as initial focal length；

According to the initial focal length and preset focusing step-length, the corresponding focal length of the target object is determined.

Further, according to the third coordinate information of each identification point in second image pre-saved, determine with The nearest target identification point of the target object space length includes:

According to determining second coordinate information, the target object is positioned, and is obtained described in ball machine positioning First positional parameter of target object；

Each identification point is positioned according to the third coordinate information of each identification point, and it is described every to obtain ball machine positioning Second positional parameter of a identification point；

According to first positional parameter and corresponding second positional parameter of each identification point, determine that described first is fixed The absolute value of position parameter and the difference of corresponding second positional parameter of each identification point；

The confidence level of each identification point is determined according to the maximum positional parameter of each absolute value and the ball machine, is judged Whether highest confidence level is greater than preset believability threshold, if so, using the corresponding identification point of highest confidence level as with The nearest target identification point of the target object space length.

Further, the positional parameter includes the step number at horizontal field of view angle, vertical field of view angle and zoom motor；

The maximum positional parameter according to each absolute value and the ball machine determines the confidence level of each identification point Include:

For each identification point, according to Determine the confidence level of the identification point；

Wherein, Depend is the confidence level of the identification point, Zoom_wholeFor the maximum step number of the zoom motor of ball machine, Delta_ZoomFor in step number the second positional parameter corresponding with the identification point of the first zoom motor in the first positional parameter The absolute value of the difference of the step number of second zoom motor, P_wholeFor the horizontal field of view angle maximum value of ball machine, Delta_PIt is fixed for first The difference at the second horizontal field of view angle in first level field angle the second positional parameter corresponding with the identification point in the parameter of position Absolute value, T_wholeFor the vertical field of view angle maximum value of ball machine, Delta_TFor the first vertical field of view angle in the first positional parameter The absolute value of the difference at the second vertical field of view angle in the second positional parameter corresponding with the identification point, A, B, C are quality coefficient.

Further, described that each identification point is positioned according to the third coordinate information of each identification point, and obtain The second positional parameter that ball machine positions each identification point includes:

Gunlock third image gathered in advance is received, by the central pixel point of second image and the third image Central pixel point is overlapped；

Receive the 4-coordinate information of each corner pixel point in the third image that the gunlock is sent；According to preparatory The coordinate transformation relation of preservation determines Five Axis information of each corner pixel point in second image；By institute The region of pixel composition of Five Axis information is stated as the calibration region in second image；

For each identification point in the calibration region, according to third coordinate of the identification point in second image Information positions the identification point；And obtain the second positional parameter that ball machine positions the identification point；

The process for pre-saving the corresponding relationship of each identification point and focal length includes:

For each identification point, to the laggard line focusing of mark point location, obtain what ball machine was focused the identification point Focal length simultaneously saves.

Further, after the determination and the nearest target identification point of the target object space length, according to preparatory Save each identification point and focal length corresponding relationship, using the corresponding focal length of the target identification point as initial focal length it Before, the method also includes:

The target object is positioned according to the second coordinate information of the target object, acquires the 4th image, and Determine the first clarity evaluation of estimate of the 4th image；

According to the corresponding clarity evaluation of estimate of each identification point pre-saved, judge the first clarity evaluation of estimate with Whether the difference of the corresponding second clarity evaluation of estimate of the target identification point is in default range, if so, carrying out subsequent Step.

Further, it is described using the corresponding focal length of the target identification point as initial focal length after, according to described initial Focal length rises and preset focusing step-length, before determining the corresponding focal length of the target object, the method also includes:

According to determining second coordinate information, the target object is positioned, using the initial focal length pair The target object is focused, and obtains the step number that ball machine focuses the first focusing motor of the target object；

According to first positional parameter, the first focusing number of motor steps, and the positional parameter pre-saved, motor step of focusing Several corresponding relationships with object distance, determine corresponding first object distance of the target object；

According to the corresponding relationship of the identification point and object distance that pre-save, corresponding second object of the target identification point is determined Away from；

According to the size relation of first object distance and the second object distance, focusing direction is determined；

According to the initial focal length and preset focusing step-length, determine that the corresponding focal length of the target object includes:

According to the initial focal length and preset focusing step-length, focuses on the focusing direction, determine the mesh Mark the corresponding focal length of object.

Further, the identification point pre-saved and the corresponding relationship of object distance include:

Identification point is focused, according to the horizontal field of view angle, vertical field of view angle of ball machine after focusing and zoom motor step number It calculates object distance and is demarcated, establish space object distance figure；

According to position of the identification point in the object distance figure of space, the corresponding relationship of identification point and object distance and preservation are determined.

On the other hand, the embodiment of the invention provides a kind of focusing device, described device includes:

First determining module, for the first coordinate information according to target object in the first image that gunlock acquires, really Fixed second coordinate information of the target object in the second image that ball machine acquires；

Second determining module, for the third coordinate information according to each identification point in second image pre-saved With second coordinate information, the determining and nearest target identification point of the target object space length；Wherein, the identification point For the corresponding pixel logo point of object immovable in second image；

Third determining module will be described for the corresponding relationship according to each identification point and focal length pre-saved The corresponding focal length of target identification point is as initial focal length；

4th determining module, for determining the target object pair according to the initial focal length and preset focusing step-length The focal length answered.

Further, second determining module, specifically for according to determining second coordinate information, to the mesh Mark object is positioned, and obtains the first positional parameter that ball machine positions the target object；According to the third of each identification point Coordinate information positions each identification point, and obtains the second positional parameter that ball machine positions each identification point；According to First positional parameter and corresponding second positional parameter of each identification point determine first positional parameter and each The absolute value of the difference of corresponding second positional parameter of identification point；According to the maximum positional parameter of each absolute value and the ball machine The confidence level for determining each identification point, judges whether highest confidence level is greater than preset believability threshold, if so, will The corresponding identification point of highest confidence level is as the target identification point nearest with the target object space length.

Second determining module is specifically used for being directed to each identification point, according toDetermine the confidence level of the identification point；Its In, Depend is the confidence level of the identification point, Zoom_wholeFor the maximum step number of the zoom motor of ball machine, Delta_ZoomIt is first The second zoom motor in the step number of the first zoom motor in positional parameter the second positional parameter corresponding with the identification point The absolute value of the difference of step number, P_wholeFor the horizontal field of view angle maximum value of ball machine, Delta_PFor first in the first positional parameter The absolute value of the difference at the second horizontal field of view angle in the second positional parameter corresponding with the identification point of horizontal field of view angle, T_wholeFor The vertical field of view angle maximum value of ball machine, Delta_TIt is corresponding with the identification point for the first vertical field of view angle in the first positional parameter The absolute value of the difference at the second vertical field of view angle in the second positional parameter, A, B, C are quality coefficient.

Further, described device further include:

Preserving module, for receiving gunlock third image gathered in advance, by the central pixel point of second image with The central pixel point of the third image is overlapped；Receive each corner pixel point in the third image that the gunlock is sent 4-coordinate information；According to the coordinate transformation relation pre-saved, determine each corner pixel point in second image In Five Axis information；The region that the pixel of the Five Axis information is constituted is as the calibration in second image Region；For each identification point in the calibration region, believed according to third coordinate of the identification point in second image Breath, positions the identification point；And obtain the second positional parameter that ball machine positions the identification point.

The preserving module is also used to, to the laggard line focusing of mark point location, obtain ball for each identification point Focal length that machine is focused the identification point and preservation.

Further, described device further include:

Judgment module is adopted for being positioned according to the second coordinate information of the target object to the target object Collect the 4th image, and determines the first clarity evaluation of estimate of the 4th image；It is corresponding according to each identification point pre-saved Clarity evaluation of estimate, judge the first clarity evaluation of estimate the second clarity evaluation of estimate corresponding with the target identification point Difference whether in default range, if so, triggering the third determining module.

Further, described device further include:

5th determining module, for positioning, adopting to the target object according to determining second coordinate information The target object is focused with the initial focal length, and obtains the first focusing motor that ball machine focuses the target object Step number；According to first positional parameter, the first focusing number of motor steps, and the positional parameter pre-saved, motor step of focusing Several corresponding relationships with object distance, determine corresponding first object distance of the target object；According to the identification point and object distance pre-saved Corresponding relationship, determine corresponding second object distance of the target identification point；According to the size of first object distance and the second object distance Relationship determines focusing direction；

4th determining module is specifically used for according to the initial focal length and preset focusing step-length, in the focusing It focuses on direction, determines the corresponding focal length of the target object.

Further, the preserving module is also used to be focused identification point, according to the horizontal field of view of ball machine after focusing Angle, vertical field of view angle and zoom motor step number calculate object distance and are demarcated, and establish space object distance figure；According to identification point in space Position in object distance figure determines the corresponding relationship of identification point and object distance and preservation.

On the other hand, the embodiment of the invention provides a kind of electronic equipment, including processor, communication interface, memory and Communication bus, wherein processor, communication interface, memory complete mutual communication by communication bus；

Memory, for storing computer program；

Processor when for executing the program stored on memory, realizes method and step described in any of the above embodiments.

In another aspect, the embodiment of the invention provides a kind of computer readable storage medium, the computer-readable storage Dielectric memory contains computer program, and the computer program realizes method step described in any of the above embodiments when being executed by processor Suddenly.

The embodiment of the invention provides a kind of focus method, device, electronic equipment and storage mediums, which comprises According to first coordinate information of the target object in the first image that gunlock acquires, determine what the target object was acquired in ball machine The second coordinate information in second image；According to the third coordinate information of each identification point in second image pre-saved With second coordinate information, the determining and nearest target identification point of the target object space length；Wherein, the identification point For the corresponding pixel logo point of object immovable in second image；According to each identification point for pre-saving with The corresponding relationship of focal length, using the corresponding focal length of the target identification point as initial focal length；According to the initial focal length and preset Focusing step-length, determine the corresponding focal length of the target object.

Due in embodiments of the present invention, electronic equipment is according to the second coordinate information of target object and each identification point Third coordinate information, it is determining with the nearest target identification point of target object space length, it has been pre-saved in electronic equipment each The corresponding relationship of identification point and focal length, after determining target identification point, using the corresponding focal length of target identification point as initial focal length, According to initial focal length and preset focusing step-length, the corresponding focal length of target object is determined.Because of the target identification point determined It is closest with the object distance of target object, therefore the focusing focal length of target identification point and target object is closest, therefore in target pair It may be implemented quickly to focus as carrying out focusing on the basis of corresponding initial focal length.Focus program in compared to the prior art, Focal time is substantially reduced, candid photograph efficiency is improved.

Detailed description of the invention

To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this For the those of ordinary skill in field, without creative efforts, it can also be obtained according to these attached drawings other Attached drawing.

Fig. 1 is accumulation process schematic diagram provided in an embodiment of the present invention；

Fig. 2 is the synchronous schematic diagram in the rifle ball provided in an embodiment of the present invention visual field；

Fig. 3 is object distance figure in space provided in an embodiment of the present invention；

Fig. 4 is that rifle ball zoom provided in an embodiment of the present invention focuses flow chart；

Fig. 5 is aggregation apparatus structural schematic diagram provided in an embodiment of the present invention；

Fig. 6 is electronic devices structure schematic diagram provided in an embodiment of the present invention.

Specific embodiment

The present invention will be describe below in further detail with reference to the accompanying drawings, it is clear that described embodiment is only this Invention a part of the embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art exist All other embodiment obtained under the premise of creative work is not made, shall fall within the protection scope of the present invention.

Fig. 1 be accumulation process schematic diagram provided in an embodiment of the present invention, the process the following steps are included:

S101: according to first coordinate information of the target object in the first image that gunlock acquires, the target pair is determined As the second coordinate information in the second image that ball machine acquires.

Focus method provided in an embodiment of the present invention is applied to electronic equipment, which can be ball machine itself, It can be the equipment such as the PC that can control ball machine, tablet computer.It in order to facilitate understanding and describes, with electronics in the embodiment of the present invention Equipment is that ball machine itself is illustrated.In practical application scene, there is the gunlock being used cooperatively with ball machine, gunlock position is fixed And fixed-focus, ball machine include rotatable holder and zoom.Fixed-focus gunlock recognizes in picture and occurs for global observation When target object, the coordinate information of target object is sent to zoom ball machine, zoom ball machine converts determining target for carrying out coordinate Then object positions target object, after the completion of positioning, focus and capture.

In embodiments of the present invention, using the image of gunlock acquisition as the first image, when target object appears in gunlock view When in open country, gunlock can acquire the first image, and identify the coordinate information of target object in the first image, by target object the For coordinate information in one image as the first coordinate information, gunlock is determining first coordinate of the target object in the first image After information, the first coordinate information is sent to ball machine.Ball machine acquires in advance the second image, wherein can operate ball machine and be in When minimum times of wide-angle position, the second image is acquired.Also, also calibration has each mark in advance in ball machine the second image gathered in advance Know point, the identification point in the embodiment of the present invention is the corresponding pixel logo point of object immovable in the second image, such as column The a certain pixel on a certain pixel, house, a certain pixel on street lamp on bar can be demarcated as identification point.This hair Bright embodiment is not defined the quantity of identification point.Due to the gunlock visual field be it is fixed, the second image is that ball machine acquires in advance 's.Therefore there are coordinate transformation relations for the first image and the second image.For example, the first corner pixel in the first image of identification Point, and determine the second corner pixel point corresponding with the first corner pixel point in the second image, according to the first corner pixel point and The coordinate information of second corner pixel point can determine horizontal offset of second image relative to the first image, vertical shift The parameters such as amount and rotation angle, the coordinate transformation relation of the first image and the second image can be determined according to above-mentioned parameter.Ball After machine receives the first coordinate information of gunlock transmission, is converted and closed according to the coordinate of the first image and the second image obtained in advance System, can determine second coordinate information of the target object in the second image.

S102: it is sat according to the third coordinate information of each identification point in second image pre-saved and described second Mark information, the determining and nearest target identification point of the target object space length；Wherein, the identification point is second figure The corresponding pixel logo point of immovable object as in.

Calibration has each identification point in ball machine the second image gathered in advance, and ball machine preserves each identification point the Third coordinate information in two images.After ball machine determines the second coordinate information of the target object in the second image, according to second The third coordinate information of coordinate information and each identification point, can determine the target identification nearest with target object space length Point.Wherein it is possible to identify the identification point nearest with the image distance of the second coordinate information in the second image, which is made For the target identification point nearest with target object space length.

S103: according to the corresponding relationship of each identification point and focal length that pre-save, by the target identification point pair The focal length answered is as initial focal length.

The corresponding relationship of each identification point and focal length is pre-saved in ball machine, wherein ball machine is directed to each mark in advance After the identification point is positioned and focused, the focal length after focusing as the corresponding focal length of the identification point and is saved for point.Ball machine After determining target identification point, the corresponding focal length of target identification point is determined as initial focal length.

S104: according to the initial focal length and preset focusing step-length, the corresponding focal length of the target object is determined.

Preset focusing step-length is preserved in ball machine, wherein preset focusing step-length can be 1 millimeter, 2 millimeters etc.. After ball machine determines initial focal length, on the basis of initial focal length, focus according to preset focusing step-length.Wherein it is possible to set in advance It sets the tone burnt direction, the direction that e.g. focal length increases is the direction of preset focusing, then on the basis of initial focal length, according to pre- If focusing step-length increase focal length, also, image can be acquired after focusing every time, and determine the clarity evaluation of estimate of image, such as The clarity evaluation of estimate of fruit image is being gradually increased, then when the clarity evaluation of estimate of image is greater than preset clarity Evaluation threshold When, it is believed that it focuses and completes, and determine that focal length at this time is the corresponding focal length of target object, or can also determine the clear of image Focal length when spending evaluation of estimate maximum is the corresponding focal length of target object.If the clarity evaluation of estimate of image is being gradually reduced, It focuses to the opposite direction in preset focusing direction, and determines the corresponding focal length of target object according to the method described above.

Since in embodiments of the present invention, ball machine is according to the second coordinate information of target object and the third of each identification point Coordinate information, it is determining with the nearest target identification point of target object space length, pre-saved in ball machine each identification point and The corresponding relationship of focal length, after determining target identification point, using the corresponding focal length of target identification point as initial focal length, according to initial Focal length rises and preset focusing step-length, determines the corresponding focal length of target object.Because of the target identification point determined and target pair The object distance of elephant is closest, therefore the focusing focal length of target identification point and target object is closest therefore corresponding in target object Focusing is carried out on the basis of initial focal length may be implemented quickly to focus.Focus program in compared to the prior art, greatly shortens Focal time, improves candid photograph efficiency.

Due to two nearest points of object distance in the nearest two points not necessarily actual scene of distance in image, in order to make The determining target identification point nearest with target object space length is more acurrate, in embodiments of the present invention, according to pre-saving Second image in each identification point third coordinate information, it is determining with the nearest target of the target object space length Identification point includes:

According to the maximum positional parameter of each absolute value and the ball machine, determines the confidence level of each identification point, sentence Whether highest confidence level of breaking is greater than preset believability threshold, if so, using the corresponding identification point of highest confidence level as The target identification point nearest with the target object space length.

It in embodiments of the present invention, can be with after ball machine determines the second coordinate information of the target object in the second image Cloud platform rotation is controlled according to the second coordinate information, to realize the positioning to target object.After the completion of positioning, ball machine is available The positional parameter of self poisoning target object, wherein positional parameter can be horizontal field of view angle, vertical field of view angle, zoom motor Step number.In embodiments of the present invention, the positional parameter of target object will be positioned as the first positional parameter.In addition, ball machine root Each identification point is positioned according to the third coordinate information of each identification point, then obtains determining for each identification point of self poisoning Position parameter, using the positional parameter as the second positional parameter.After ball machine obtains the first positional parameter, it can determine that first is fixed respectively The absolute value of position parameter and the difference of each identification point corresponding second positional parameter, then according to each absolute value, ball machine Maximum positional parameter and preset formula determine the confidence level of each identification point, preset believability threshold are preserved in ball machine, When highest confidence level be greater than preset believability threshold when, using the corresponding identification point of highest confidence level as with target object The nearest target identification point of space length.In order to improve efficiency, ball machine can first determine in the second image with target object distance Then the identification point of closer preset quantity, such as 3,4 identification points determine the first positional parameter and preset quantity respectively The absolute value of the difference of corresponding second positional parameter of identification point, then according to the maximum positioning ginseng of each absolute value and ball machine Number, determines the confidence level of each identification point, when highest confidence level is greater than preset believability threshold, by highest confidence level Corresponding identification point is as the target identification point nearest with target object space length.

Specifically, determining the confidence level packet of each identification point according to the maximum positional parameter of each absolute value and ball machine It includes:

Formula is preserved in ball machine After ball machine determines the absolute value of the difference of the first positional parameter and each second positional parameter, by each absolute value and ball machine Maximum positional parameter brings above-mentioned formula into, can determine the confidence level of each identification point, then by the highest mark of confidence level Know point and is determined as target identification point.

It is described that each identification point is positioned according to the third coordinate information of each identification point, and obtain ball machine positioning institute The second positional parameter for stating each identification point includes:

Is obtained by ball machine and is gathered to the identification point for the laggard line focusing of mark point location for each identification point Burnt focal length and preservation.

Ball machine is before saving corresponding second positional parameter of each identification point, it is necessary first to regard to the gunlock visual field and ball machine Open country synchronizes, and Fig. 2 is the synchronous schematic diagram in the rifle ball visual field, as shown in Fig. 2, gunlock acquires third image in advance, and by third figure As being sent to ball machine, it should be noted that the third image and the first image can be same piece image, be also possible to different Image.The central pixel point a of the central pixel point A of third image and the second image of itself acquisition are overlapped by ball machine, and ball machine is also Receive the 4-coordinate information of each corner pixel point in the third image that gunlock is sent, that is, tetra- pixels of BCDE in Fig. 2 4-coordinate information of the point in third image determines each corner pixel then according to the coordinate transformation relation pre-saved Five Axis information o'clock in the second image, that is, determine the 5th of tetra- pixels of bcde in Fig. 2 in the second image Coordinate information, the region that the pixel of Five Axis information is constituted complete at this time as the calibration region in the second image The rifle ball visual field is synchronous.

User can be in each identification point of selection in calibration region, wherein user can be chosen by single-click operation and be marked Know point, identification point can also be chosen by double click operation, or mark can also be chosen by inputting the coordinate information of pixel Point.It is of course also possible to determine each identification point according to object immovable in the second image by ball machine.Ball machine exists according to user Selection operation in calibration region determines each identification point, and obtains the third coordinate information of each identification point.Then for every A identification point controls cloud platform rotation according to the third coordinate information of the identification point, positions to the identification point, after the completion of positioning Ball machine is obtained to position the second positional parameter of the identification point and save.Corresponding second positioning of each identification point can be saved in this way Parameter.In addition, ball machine is directed to each identification point, it is focused after being positioned to the identification point, obtains ball machine after the completion of focusing The focal length be focused to the identification point and preservation, as the corresponding focal length of the identification point.

In order to which the initial focal length for guaranteeing determining is accurate, in embodiments of the present invention, the determination and the target object are empty Between after the nearest target identification point of distance, according to the corresponding relationship of each identification point and focal length that pre-save, by institute Before the corresponding focal length of target identification point is stated as initial focal length, the method also includes:

After ball machine determines target identification point, the target object is determined according to the second coordinate information of target object Position acquires the 4th image, and determines the first clarity evaluation of estimate of the 4th image.Ball machine is directed to each mark point location in advance Afterwards, the corresponding image of the identification point is acquired, and determines that the second clarity evaluation of estimate of the image is corresponding clear as the identification point Clear degree evaluation of estimate.Ball machine determines the corresponding second clarity evaluation of estimate of target identification point, and judge the first clarity evaluation of estimate with Whether the difference of the corresponding second clarity evaluation of estimate of target identification point is in default range.If it is, determining target mark The confidence level for knowing point is higher, can carry out the corresponding relationship according to each identification point and focal length pre-saved at this time, will The step of corresponding focal length of the target identification point is as initial focal length, if it is not, then determine the confidence level of target identification point compared with It is low, a focal length can be set at random at this time as initial focal length.

It is in embodiments of the present invention, described by the corresponding coke of the target identification point in order to further increase focusing speed After as initial focal length, according to the initial focal length and preset focusing step-length, determine that the target object is corresponding Before focal length, the method also includes:

The identification point pre-saved and the corresponding relationship of object distance include:

After ball machine determines initial focal length, according to the second coordinate information of target object, target object is positioned, is adopted Target object is focused with initial focal length, and obtains the step number of the first focusing motor of ball machine focus objects object；Ball machine The corresponding relationship of positional parameter, the number of motor steps and object distance of focusing itself is preserved, which determined by the model of ball machine 's.Ball machine according to the first positional parameter, first focusing number of motor steps, and pre-save positional parameter, focusing number of motor steps with The corresponding relationship of object distance determines corresponding first object distance of target object.In addition, ball machine is directed to each identification point in advance, to the mark After knowledge point carries out positioning focusing, the step number that ball machine focuses the second focusing motor of the identification point is obtained；Then according to the identification point The second positional parameter, second focusing motor step number, and pre-save positional parameter, focus number of motor steps and object distance pair It should be related to, determine the corresponding object distance of the identification point and save.Fig. 3 is the space object distance figure saved in ball machine, as shown in figure 3, empty Between horizontal field of view angle pan, vertical field of view angle tilt, the relationship between zoom motor step number Zoom are labelled in object distance figure.With ball The point on each concentric spherical centered on machine represents some PT angle under a burnt section (Zoom step number is identical), if It is the state focused after completing, then the focusing number of motor steps of the point can serve as absolute amount to calculate object distance and to this It is fixed to click through rower.Object distance calibration can be carried out after focusing to each spatial point according to focus data, it is final to establish sky Between object distance equipotential lines, as shown in figure 3, be differently formed several space spherical surfaces with zoom motor step number, and any in space A little all object distance will be demarcated after each focus.ABCD is closer four identification points of distance objective object O in Fig. 3, and ball machine is true The identification point is determined as target when there are the identification point that confidence level is more than threshold value by the confidence level for determining tetra- identification points of ABCD Identification point.Ball machine determines corresponding second object distance of target identification point according to the identification point pre-saved and the corresponding relationship of object distance, According to the size relation of the first object distance and the second object distance, focusing direction is determined.Specifically, when the first object distance is greater than the second object distance When, focusing direction is focal length augment direction, and when the first object distance is less than the second object distance, focusing direction is that focal length reduces direction.Root It according to initial focal length and preset focusing step-length, focuses on the focusing direction determined, determines the corresponding coke of target object Away from.To further improve focusing speed.

Fig. 4 be rifle ball zoom provided in an embodiment of the present invention focus flow chart, as shown in figure 4, focus start after, first into Row gunlock is synchronous with the coordinate of ball machine, it is, gunlock third image gathered in advance is received, by the center pixel of the second image Point is overlapped with the central pixel point of the third image；Receive each corner pixel in the third image that the gunlock is sent The 4-coordinate information of point；According to the coordinate transformation relation, determine each corner pixel point in second image Five Axis information；The region that the pixel of the Five Axis information is constituted is as the calibration area in second image Domain.Then the target identification point that confidence level is greater than preset believability threshold is judged whether there is.It is, according to determining institute The second coordinate information is stated, the target object is positioned, and obtains the first positioning ginseng that ball machine positions the target object Number；According to first positional parameter the second positional parameter corresponding with each identification point pre-saved, determine described in The absolute value of the difference of first positional parameter and corresponding second positional parameter of each identification point；According to each absolute value and described The maximum positional parameter of ball machine determines the confidence level of each identification point, judge highest confidence level whether be greater than it is preset can Confidence threshold.If it is present using the corresponding identification point of highest confidence level as nearest with the target object space length Target identification point, using the corresponding focal length of the target identification point as initial focal length；According to the initial focal length and preset Focusing step-length, determines the corresponding focal length of the target object.If it does not exist, then being gathered according to method in the prior art It is burnt.

Fig. 5 is focusing device structural schematic diagram provided in an embodiment of the present invention, which includes:

First determining module 41, for the first coordinate information according to target object in the first image that gunlock acquires, Determine second coordinate information of the target object in the second image that ball machine acquires；

Second determining module 42, for the third coordinate letter according to each identification point in second image pre-saved Breath and second coordinate information, the determining and nearest target identification point of the target object space length；Wherein, the mark Point is the corresponding pixel logo point of object immovable in second image；

Third determining module 43, for the corresponding relationship according to each identification point and focal length pre-saved, by institute The corresponding focal length of target identification point is stated as initial focal length；

4th determining module 44, for determining the target object according to the initial focal length and preset focusing step-length Corresponding focal length.

Second determining module 42, specifically for according to determining second coordinate information, to the target object It is positioned, and obtains the first positional parameter that ball machine positions the target object；Believed according to the third coordinate of each identification point Breath positions each identification point, and obtains the second positional parameter that ball machine positions each identification point；According to described One positional parameter the second positional parameter corresponding with each identification point pre-saved, determine first positional parameter and The absolute value of the difference of corresponding second positional parameter of each identification point；According to the maximum positioning of each absolute value and the ball machine Parameter determines the confidence level of each identification point, judges whether highest confidence level is greater than preset believability threshold, if It is, using the corresponding identification point of highest confidence level as the target identification point nearest with the target object space length.

The positional parameter includes the step number at horizontal field of view angle, vertical field of view angle and zoom motor；

Second determining module 42 is specifically used for being directed to each identification point, according toDetermine the confidence level of the identification point；Its In, Depend is the confidence level of the identification point, Zoom_wholeFor the maximum step number of the zoom motor of ball machine, Delta_ZoomIt is first The second zoom motor in the step number of the first zoom motor in positional parameter the second positional parameter corresponding with the identification point The absolute value of the difference of step number, P_wholeFor the horizontal field of view angle maximum value of ball machine, Delta_PFor first in the first positional parameter The absolute value of the difference at the second horizontal field of view angle in the second positional parameter corresponding with the identification point of horizontal field of view angle, T_wholeFor The vertical field of view angle maximum value of ball machine, Delta_TIt is corresponding with the identification point for the first vertical field of view angle in the first positional parameter The absolute value of the difference at the second vertical field of view angle in the second positional parameter, A, B, C are quality coefficient.

Described device further include:

Preserving module 45, for receiving gunlock third image gathered in advance, by the central pixel point of second image It is overlapped with the central pixel point of the third image；Receive each corner pixel point in the third image that the gunlock is sent 4-coordinate information；According to the coordinate transformation relation pre-saved, determine each corner pixel point in second figure Five Axis information as in；The region that the pixel of the Five Axis information is constituted is as the mark in second image Determine region；For each identification point in the calibration region, according to third coordinate of the identification point in second image Information positions the identification point；And obtain the second positional parameter that ball machine positions the identification point.

The preserving module 45 is also used to, to the laggard line focusing of mark point location, obtain for each identification point Focal length that ball machine is focused the identification point and preservation.

Described device further include:

Judgment module 46, for being positioned according to the second coordinate information of the target object to the target object, The 4th image is acquired, and determines the first clarity evaluation of estimate of the 4th image；According to each identification point pair pre-saved The clarity evaluation of estimate answered judges the first clarity evaluation of estimate the second clarity evaluation corresponding with the target identification point Whether the difference of value is in default range, if so, triggering the third determining module 43.

Described device further include:

5th determining module 47, for being positioned to the target object according to determining second coordinate information, The target object is focused using the initial focal length, and obtains the first focusing electricity that ball machine focuses the target object The step number of machine；According to first positional parameter, the first focusing number of motor steps, and the positional parameter, the focusing motor that pre-save The corresponding relationship of step number and object distance determines corresponding first object distance of the target object；According to the identification point and object pre-saved Away from corresponding relationship, determine corresponding second object distance of the target identification point；According to the big of first object distance and the second object distance Small relationship determines focusing direction；

4th determining module 44 is specifically used for according to the initial focal length and preset focusing step-length, in the tune It focuses on burnt direction, determines the corresponding focal length of the target object.

The preserving module 45, is also used to be focused identification point, according to the horizontal field of view angle of ball machine after focusing, vertical Field angle and zoom motor step number calculate object distance and are demarcated, and establish space object distance figure；According to identification point in space object distance figure In position, determine the corresponding relationship of identification point and object distance and preservation.

A kind of electronic equipment is additionally provided in the embodiment of the present invention, as shown in Figure 6, comprising: processor 501, communication interface 502, memory 503 and communication bus 504, wherein processor 501, communication interface 502, memory 503 pass through communication bus 504 complete mutual communication；

It is stored with computer program in the memory 503, when described program is executed by the processor 501, so that The processor 501 executes following steps:

Based on the same inventive concept, a kind of electronic equipment is additionally provided in the embodiment of the present invention, due to above-mentioned electronic equipment The principle solved the problems, such as is similar to focus method, therefore the implementation of above-mentioned electronic equipment may refer to the implementation of method, repetition Place repeats no more.

The communication bus that above-mentioned ball machine is mentioned can be Peripheral Component Interconnect standard (Peripheral Component Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard Architecture, EISA) bus etc..The communication bus can be divided into address bus, data/address bus, control bus etc..For just It is only indicated with a thick line in expression, figure, it is not intended that an only bus or a type of bus.

Communication interface 502 is for the communication between above-mentioned ball machine and other equipment.

Memory may include random access memory (Random Access Memory, RAM), also may include non-easy The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also To be storage device that at least one is located remotely from aforementioned processor.

Above-mentioned processor can be general processor, including central processing unit, network processing unit (Network Processor, NP) etc.；It can also be digital signal processor (Digital Signal Processing, DSP), dedicated collection At circuit, field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hard Part component etc..

When processor executes the program stored on memory in embodiments of the present invention, realize according to target object in rifle The first coordinate information in first image of machine acquisition, determines second of the target object in the second image that ball machine acquires Coordinate information；Believed according to the third coordinate information of each identification point in second image pre-saved and second coordinate Breath, the determining and nearest target identification point of the target object space length；Wherein, the identification point is in second image The corresponding pixel logo point of immovable object；According to the corresponding relationship of each identification point and focal length that pre-save, Using the corresponding focal length of the target identification point as initial focal length；According to the initial focal length and preset focusing step-length, determine The corresponding focal length of the target object.

Electronic equipment is according to the second coordinate information of target object and the third coordinate information of each identification point, determining and mesh Mark object space pre-saves the corresponding relationship of each identification point and focal length in nearest target identification point, ball machine, really After making target identification point, using the corresponding focal length of target identification point as initial focal length, according to initial focal length and preset tune Burnt step-length determines the corresponding focal length of target object.Because the object distance of the target identification point and target object determined is closest, because The focusing focal length of this target identification point and target object is closest, thus on the basis of target object corresponding initial focal length into Row focusing may be implemented quickly to focus.Focus program in compared to the prior art, substantially reduces focal time, improves and grabs Clap efficiency.

The embodiment of the invention also provides a kind of computers to store readable storage medium storing program for executing, the computer readable storage medium It is inside stored with the computer program that can be executed by electronic equipment, when described program is run on the electronic equipment, so that institute It states when electronic equipment executes and realizes following steps:

Based on the same inventive concept, a kind of computer readable storage medium is additionally provided in the embodiment of the present invention, due to place The principle and focus method that reason device is solved the problems, such as in the computer program stored on executing above-mentioned computer readable storage medium It is similar, therefore processor may refer to method in the implementation for the computer program for executing above-mentioned computer-readable recording medium storage Implementation, overlaps will not be repeated.

Above-mentioned computer readable storage medium can be any usable medium that the processor in electronic equipment can access Or data storage device, including but not limited to magnetic storage such as floppy disk, hard disk, tape, magneto-optic disk (MO) etc., optical memory Such as CD, DVD, BD, HVD and semiconductor memory such as ROM, EPROM, EEPROM, nonvolatile memory (NAND FLASH), solid state hard disk (SSD) etc..

Computer program, computer program quilt are provided in the computer readable storage medium provided in embodiments of the present invention The first coordinate information according to target object in the first image that gunlock acquires is realized when processor executes, and determines the target Second coordinate information of the object in the second image that ball machine acquires；According to each mark in second image pre-saved The third coordinate information and second coordinate information of point, the determining and nearest target identification of the target object space length Point；Wherein, the identification point is the corresponding pixel logo point of object immovable in second image；According to pre-saving Each identification point and focal length corresponding relationship, using the corresponding focal length of the target identification point as initial focal length；According to The initial focal length and preset focusing step-length, determine the corresponding focal length of the target object.

The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates, Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or The function of being specified in multiple boxes.

These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one The step of function of being specified in a box or multiple boxes.

Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as It selects embodiment and falls into all change and modification of the scope of the invention.

Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies Within, then the present invention is also intended to include these modifications and variations.

Claims

1. a kind of focus method, which is characterized in that the described method includes:

According to first coordinate information of the target object in the first image that gunlock acquires, determine that the target object is adopted in ball machine The second coordinate information in second image of collection；

According to the third coordinate information and second coordinate information of each identification point in second image pre-saved, really The fixed and nearest target identification point of the target object space length；Wherein, the identification point be second image in can not The corresponding pixel logo point of mobile object；

According to the corresponding relationship of each identification point and focal length that pre-save, the corresponding focal length of the target identification point is made For initial focal length；

2. the method as described in claim 1, which is characterized in that according to each identification point in second image pre-saved Third coordinate information, determine that the target identification point nearest with the target object space length includes:

According to determining second coordinate information, the target object is positioned, and obtains ball machine and positions the target First positional parameter of object；

Each identification point is positioned according to the third coordinate information of each identification point, and obtains ball machine and positions each mark Know the second positional parameter of point；

According to first positional parameter and corresponding second positional parameter of each identification point, the first positioning ginseng is determined The absolute value of the difference of number the second positional parameter corresponding with each identification point；

The confidence level that each identification point is determined according to the maximum positional parameter of each absolute value and the ball machine, judges highest Confidence level whether be greater than preset believability threshold, if so, using the corresponding identification point of highest confidence level as with it is described The nearest target identification point of target object space length.

3. method according to claim 2, which is characterized in that the positional parameter includes horizontal field of view angle, vertical field of view angle With the step number of zoom motor；

The maximum positional parameter according to each absolute value and the ball machine determines that the confidence level of each identification point includes:

Wherein, Depend is the confidence level of the identification point, Zoom_wholeFor the maximum step number of the zoom motor of ball machine, Delta_Zoom For the second zoom in step number the second positional parameter corresponding with the identification point of the first zoom motor in the first positional parameter The absolute value of the difference of the step number of motor, P_wholeFor the horizontal field of view angle maximum value of ball machine, Delta_PFor in the first positional parameter First level field angle the second positional parameter corresponding with the identification point in the second horizontal field of view angle difference absolute value, T_wholeFor the vertical field of view angle maximum value of ball machine, Delta_TFor in the first positional parameter the first vertical field of view angle and the identification point The absolute value of the difference at the second vertical field of view angle in corresponding second positional parameter, A, B, C are quality coefficient.

4. method according to claim 2, which is characterized in that the third coordinate information according to each identification point is to each Identification point is positioned, and is obtained ball machine and positioned the second positional parameter of each identification point and include:

Gunlock third image gathered in advance is received, by the center of the central pixel point of second image and the third image Pixel is overlapped；

Receive the 4-coordinate information of each corner pixel point in the third image that the gunlock is sent；According to pre-saving Coordinate transformation relation, determine Five Axis information of each corner pixel point in second image；By described The region that the pixel of Five Axis information is constituted is as the calibration region in second image；

For each identification point in the calibration region, believed according to third coordinate of the identification point in second image Breath, positions the identification point；And obtain the second positional parameter that ball machine positions the identification point；

5. the method as described in claim 1, which is characterized in that the determination and the nearest mesh of the target object space length After marking identification point, according to the corresponding relationship of each identification point and focal length that pre-save, by the target identification point pair Before the focal length answered is as initial focal length, the method also includes:

The target object is positioned according to the second coordinate information of the target object, acquires the 4th image, and determine First clarity evaluation of estimate of the 4th image；

According to the corresponding clarity evaluation of estimate of each identification point pre-saved, judge the first clarity evaluation of estimate with it is described Whether the difference of the corresponding second clarity evaluation of estimate of target identification point is in default range, if so, carrying out subsequent step.

6. method according to claim 2, which is characterized in that described using the corresponding focal length of the target identification point as initial After focal length, according to the initial focal length and preset focusing step-length, before determining the corresponding focal length of the target object, institute State method further include:

According to determining second coordinate information, the target object is positioned, using the initial focal length to described Target object is focused, and obtains the step number that ball machine focuses the first focusing motor of the target object；

According to first positional parameter, first focusing number of motor steps, and pre-save positional parameter, focusing number of motor steps with The corresponding relationship of object distance determines corresponding first object distance of the target object；

According to the corresponding relationship of the identification point and object distance that pre-save, corresponding second object distance of the target identification point is determined；

According to the initial focal length and preset focusing step-length, focuses on the focusing direction, determine the target pair As corresponding focal length.

7. method as claimed in claim 6, which is characterized in that the identification point that the basis pre-saves is corresponding with object distance to close System, determines that corresponding second object distance of the target identification point includes:

Identification point is focused, is calculated according to the horizontal field of view angle, vertical field of view angle of ball machine after focusing and zoom motor step number Object distance is simultaneously demarcated, and space object distance figure is established；

8. a kind of focusing device, which is characterized in that described device includes:

First determining module determines institute for the first coordinate information according to target object in the first image that gunlock acquires State second coordinate information of the target object in the second image that ball machine acquires；

Second determining module, for according to the third coordinate information of each identification point and institute in second image pre-saved State the second coordinate information, the determining and nearest target identification point of the target object space length；Wherein, the identification point is institute State the corresponding pixel logo point of immovable object in the second image；

Third determining module, for the corresponding relationship according to each identification point and focal length pre-saved, by the target The corresponding focal length of identification point is as initial focal length；

4th determining module, for determining that the target object is corresponding according to the initial focal length and preset focusing step-length Focal length.

9. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein processing Device, communication interface, memory complete mutual communication by communication bus；

Memory, for storing computer program；

Processor when for executing the program stored on memory, realizes the described in any item method steps of claim 1-7 Suddenly.

10. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium Program realizes claim 1-7 described in any item method and steps when the computer program is executed by processor.