CN103685950A

CN103685950A - Method and device for preventing shaking of video image

Info

Publication number: CN103685950A
Application number: CN201310656782.XA
Authority: CN
Inventors: 孔晓东; 尹志叶; 王栋
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2013-12-06
Filing date: 2013-12-06
Publication date: 2014-03-26

Abstract

The invention discloses a method and device for preventing shaking of a video image. The method for preventing shaking of the video image comprises the steps that the focusing positional value of a video image to be processed when focusing shooting is conducted by shooting equipment is obtained and serves as a current focusing positional value; a shooting object distance corresponding to the current focusing positional value is determined based on the corresponding relation between a focusing positional value and a shooting object distance which are established in advance, and the shooting object distance corresponding to the current focusing positional value serves as the current shooting object distance; the image offset corresponding to the current shooting object distance and a current camera lens rotating angle is determined based on the corresponding relation among the shooting object distance, a camera lens rotating angle and image offset which are established in advance, and the image offset corresponding to the current shooting object distance and the current camera lens rotating angle serves as the adjustment amount of image offset, wherein a camera rotating angle, of the video image to be processed, which is obtained when focusing shooting is conducted by the shooting equipment serves as the current camera lens rotating angle, and video image anti-shaking offset processing is conducted on the video image to be processed by means of the adjustment amount of image offset. By the adoption of the method for preventing shaking of the video image, the anti-shaking effect of the video image is improved.

Description

A kind of video image anti-fluttering method and device

Technical field

The present invention relates to video capture technical field and technical field of image processing, relate in particular to a kind of video image anti-fluttering method and device.

Background technology

In recent years, fast development along with video capture equipment and technology, a lot of capture apparatus have all started the shooting of supporting that high definition is even full HD, for example, slr camera, card form camera, band camera function mobile phone, video monitoring equipment, drive recorder etc., therefore, the shooting quality of video has become more and more important for evaluating capture apparatus.

The factor that affects video capture quality is a lot, comprising: the stability of resolution, frame per second, noise, color saturation, image definition, video etc.Wherein, the stability of video is very important influencing factor.Once because unstable while taking, can make whole float, so not only can cause not fogging clearly, and make human eye be easy to fatigue, subjective feeling is very poor.Especially for current conventional smart mobile phone, card form camera, because the quality of this kind equipment own is light, be again by the hand-held shooting of user, the shake that is easy to produce picture.For this type of jitter problem, the most effective settling mode is to carry out anti-shake processing to taking the image obtaining, and makes capture apparatus possess video anti-trembling function.

The anti-shake scheme of a kind of video image adopting in currently available technology is as follows:

For capture apparatus, set up anti-shake algorithm model, be specifically as follows: use capture apparatus to take default object, and the rotation of lens angular speed while taking a secondary video image by the gyroscope collection in capture apparatus, and according to the time interval between rotation of lens angular speed and adjacent two two field pictures, determine rotation of lens angle, and this video image that relatively shooting obtains and the standard picture of default object, obtain the image shift amount of this video image of comparison with standard image, as the image shift amount corresponding with this rotation of lens angle, by repeatedly taking, can determine image shift amount corresponding to many arrangements of mirrors head anglec of rotation difference, thereby set up the corresponding relation between rotation of lens angle and image shift amount, as anti-shake algorithm model,

After having set up anti-shake algorithm model, can be based on this anti-shake algorithm model, the video image that uses this capture apparatus to take to obtain is carried out to anti-shake processing, be specifically as follows: for using this capture apparatus to take the pending video image obtaining, obtain the rotation of lens angle while taking, and the rotation of lens angle based on setting up and the corresponding relation between image shift amount, determine image shift amount corresponding to this rotation of lens angle that is with shooting, as image shift adjustment amount, then adopt this image shift adjustment amount, this pending video image is carried out to the anti-shake processing of video image, for example, pending video image is offset to this image shift adjustment amount, obtain being offset rear video image, and from skew rear video image, shear the image of predeterminable area, as the video image after anti-shake processing.

Adopt the anti-shake scheme of above-mentioned video image, can carry out anti-shake processing to video image, improved video quality, yet in actual applications, only the rotation of lens angle when taking is carried out anti-shakely based on capture apparatus, makes the anti-shake effect of video image poor.

Summary of the invention

The embodiment of the present invention provides a kind of video image anti-fluttering method and device, in order to solve in prior art, exist carry out the anti-shake poor problem of anti-shake effect of video image.

First aspect, provides a kind of video image anti-fluttering method, comprising:

Obtain the focusing position value of pending video image when being taken by capture apparatus focusing, as this focusing position value;

Focusing position value based on setting up in advance and the corresponding relation of reference object distance, determine the reference object distance corresponding with described this focusing position value, as this reference object distance;

Corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as image shift adjustment amount, wherein, described this rotation of lens angle is the rotation of lens angle of described pending video image when being taken by described capture apparatus;

Adopt described image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image.

In conjunction with first aspect, in the possible implementation of the first, corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as image shift adjustment amount, specifically comprise:

Reference object distance based on setting up in advance, the x lens shaft anglec of rotation of rotating along x axle and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus;

Reference object distance based on setting up in advance, the y lens shaft anglec of rotation of rotating along y axle and the corresponding relation between y axle image shift amount, determine the y axle image shift amount corresponding with described this reference object distance and this y lens shaft anglec of rotation, as y axle image shift adjustment amount, wherein, described this y lens shaft anglec of rotation is described pending video image y lens shaft anglec of rotation along the rotation of y axle when being taken by described capture apparatus;

Wherein, the direction that the camera lens of described capture apparatus points to is vertical with the plane that y axle forms with x axle;

Adopt described image shift adjustment amount, described pending video image carried out to the anti-shake processing of video image, specifically comprise:

Adopt described x axle image shift adjustment amount and described y axle image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image.

In conjunction with first aspect, or the possible implementation of the first of first aspect, in the possible implementation of the second, also comprise:

Determine described pending video image z lens shaft anglec of rotation along the rotation of z axle when being taken by described capture apparatus, as image rotation adjustment amount, wherein, the direction that the camera lens of described capture apparatus points to is identical with z direction of principal axis;

Adopt described image rotation adjustment amount, described pending video image is carried out to the anti-shake rotation of video image and process.

In conjunction with first aspect, in the third possible implementation, set up in the following way the corresponding relation of focusing position value and reference object distance:

For each predeterminable range in a plurality of different predeterminable ranges, obtain described capture apparatus in the position of this predeterminable range of distance of camera lens reference object, focusing position value when default object is taken in focusing, as the focusing position value corresponding with this predeterminable range;

Based on each self-corresponding focusing position value of described a plurality of different predeterminable ranges, set up the corresponding relation of focusing position value and reference object distance.

In conjunction with first aspect, in the 4th kind of possible implementation, set up in the following way the corresponding relation between reference object distance, rotation of lens angle and image shift amount:

For each predeterminable range in a plurality of different predeterminable ranges, obtain described capture apparatus in the position of this predeterminable range of distance of camera lens reference object, the rotation of lens angle while taking described default object, as the rotation of lens angle corresponding with this predeterminable range;

For each predeterminable range, more described capture apparatus takes in the position of this predeterminable range of distance of camera lens reference object the sample image that described default object obtains, standard picture with described default object, obtain comparing the image shift amount of sample image described in described standard picture, as the image shift amount corresponding with this predeterminable range;

Based on described a plurality of different predeterminable ranges each self-corresponding rotation of lens angle and image shift amount, set up the corresponding relation between reference object distance, rotation of lens angle and image shift amount.

Second aspect, provides a kind of video image anti-shake apparatus, comprising:

Focusing position value acquiring unit, for obtaining the focusing position value of pending video image when being taken by capture apparatus focusing, as this focusing position value;

Distance determining unit, for the focusing position value based on setting up in advance and the corresponding relation of reference object distance, determines the reference object distance corresponding with described this focusing position value, as this reference object distance;

Skew adjustment amount determining unit, for the corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as image shift adjustment amount, wherein, described this rotation of lens angle is the rotation of lens angle of described pending video image when being taken by described capture apparatus;

Anti-shake processing unit, for adopting described image shift adjustment amount, carries out the anti-shake migration processing of video image to described pending video image.

In conjunction with second aspect, in the possible implementation of the first, described distance determining unit, specifically for the reference object distance based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus; And

Described anti-shake processing unit, specifically for adopting described x axle image shift adjustment amount and described y axle image shift adjustment amount, carries out the anti-shake migration processing of video image to described pending video image.

In conjunction with second aspect, or the possible implementation of the first of second aspect, in the possible implementation of the second, also comprise:

Rotation adjustment amount determining unit, for determining described pending video image z lens shaft anglec of rotation along the rotation of z axle when being taken by described capture apparatus, as image rotation adjustment amount, wherein, the direction that the camera lens of described capture apparatus points to is identical with z direction of principal axis;

Described anti-shake processing unit, also for adopting described image rotation adjustment amount, carries out the anti-shake rotation of video image to described pending video image and processes.

In conjunction with second aspect, in the third possible implementation, also comprise:

First sets up unit, for setting up in the following way the corresponding relation of focusing position value and reference object distance:

In conjunction with second aspect, in the 4th kind of possible implementation, also comprise:

Second sets up unit, for setting up in the following way the corresponding relation between reference object distance, rotation of lens angle and image shift amount:

The video image anti-fluttering method providing according to first aspect, the video image anti-shake apparatus that second aspect provides, in carrying out the anti-shake process of video image, focusing position value according to pending image when focusing is taken, and the focusing position value of setting up in advance and the corresponding relation of reference object distance, determine this reference object distance, and according to this reference object distance and this rotation of lens angle when taking, and the reference object distance of setting up in advance, corresponding relation between rotation of lens angle and image shift amount, determine corresponding image shift adjustment amount, thereby while making to adopt this image shift adjustment amount to carry out the anti-shake migration processing of video image to pending video image, the impact of rotation of lens angle difference while not only having considered to take on video image dithering, camera lens and distance different impacts on video image dithering between reference object while also having considered to take, thereby improved, carry out the anti-shake anti-shake effect of video image.

The third aspect, provides a kind of video image anti-fluttering method, comprising:

Corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as the first image shift adjustment amount, wherein, described this rotation of lens angle is the rotation of lens angle of described pending video image when being taken by described capture apparatus;

Corresponding relation between reference object distance, camera lens side-play amount and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this camera lens side-play amount, as the second image shift adjustment amount, wherein, described this camera lens side-play amount is the camera lens side-play amount of described pending video image when being taken by described capture apparatus;

According to described the first image shift adjustment amount and described the second image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image.

In conjunction with the third aspect, in the possible implementation of the first, corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as the first image shift adjustment amount, specifically comprise:

Reference object distance based on setting up in advance, the x lens shaft anglec of rotation of rotating along x axle and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as an x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus;

Reference object distance based on setting up in advance, the y lens shaft anglec of rotation of rotating along y axle and the corresponding relation between y axle image shift amount, determine the y axle image shift amount corresponding with described this reference object distance and this y lens shaft anglec of rotation, as a y axle image shift adjustment amount, wherein, described this y lens shaft anglec of rotation is described pending video image y lens shaft anglec of rotation along the rotation of y axle when being taken by described capture apparatus;

Corresponding relation between reference object distance, camera lens side-play amount and image shift amount based on setting up in advance, determines the image shift amount corresponding with described this reference object distance and this camera lens side-play amount, as the second image shift adjustment amount, specifically comprises:

Reference object distance based on setting up in advance, along the x lens shaft side-play amount of x axle offset and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft side-play amount, as the 2nd x axle image shift adjustment amount, wherein, described this x lens shaft side-play amount is described pending video image x lens shaft side-play amount along x axle offset when being taken by described capture apparatus;

Reference object distance based on setting up in advance, along the y lens shaft side-play amount of y axle offset and the corresponding relation between y axle image shift amount, determine the y axle image shift amount corresponding with described this reference object distance and this y lens shaft side-play amount, as the 2nd y axle image shift adjustment amount, wherein, described this y lens shaft side-play amount is described pending video image y lens shaft side-play amount along y axle offset when being taken by described capture apparatus;

According to described the first image shift adjustment amount and described the second image shift adjustment amount, described pending video image is carried out to the anti-shake processing of video image, specifically comprise:

According to a described x axle image shift adjustment amount, a described y axle image shift adjustment amount, described the 2nd x axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image.

In conjunction with the possible implementation of the first of the third aspect, in the possible implementation of the second, according to a described x axle image shift adjustment amount, a described y axle image shift adjustment amount, described the 2nd x axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image, specifically comprises:

According to a described x axle image shift adjustment amount and described the 2nd x axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd x axle image shift adjustment amount;

According to a described y axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd y axle image shift adjustment amount;

Adopt described the 3rd x axle image shift adjustment amount and described the 3rd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake processing of video image.

The possible implementation of the second in conjunction with the possible implementation of the first of the third aspect, the third aspect or the third aspect, in the third possible implementation, also comprises:

In conjunction with the third aspect, in the 4th kind of possible mode, set up in the following way the corresponding relation of focusing position value and reference object distance:

In conjunction with the third aspect, in the 5th kind of possible mode, set up in the following way the corresponding relation between reference object distance, rotation of lens angle and image shift amount:

In conjunction with the third aspect, in the 6th kind of possible mode, set up in the following way the corresponding relation between reference object distance, camera lens side-play amount and image shift amount:

For each predeterminable range in a plurality of different predeterminable ranges, obtain described capture apparatus in the position of this predeterminable range of distance of camera lens reference object, the camera lens side-play amount while taking described default object, as the camera lens side-play amount corresponding with this predeterminable range;

Based on described a plurality of different predeterminable ranges each self-corresponding camera lens side-play amount and image shift amount, set up the corresponding relation between reference object distance, camera lens side-play amount and image shift amount.

In conjunction with the possible implementation of the second of the possible implementation of the first of the third aspect, the third aspect, the third aspect, the third possible implementation of the third aspect, in the 7th kind of possible implementation, also comprises:

Determine described pending video image z lens shaft side-play amount along z axle offset when being taken by described capture apparatus, as this z lens shaft side-play amount, wherein, the direction that the camera lens of described capture apparatus points to is identical with z direction of principal axis;

Reference object distance based on setting up in advance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio, determine the image scaling ratio corresponding with described this reference object distance and this z lens shaft side-play amount;

Adopt the image scaling ratio of described correspondence, described pending video image is carried out to the anti-shake convergent-divergent of video image and process.

In conjunction with the 7th kind of possible implementation of the third aspect, in the 8th kind of possible implementation, set up in the following way reference object distance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio:

For each predeterminable range in a plurality of different predeterminable ranges, obtain described capture apparatus in the position of this predeterminable range of distance of camera lens reference object, the z lens shaft side-play amount along z axle offset while taking described default object, as the z lens shaft side-play amount corresponding with this predeterminable range;

For each predeterminable range, determine that described capture apparatus takes the area of default object described in the sample image that described default object obtains in the position of this predeterminable range of distance of camera lens reference object, ratio with the area of default object described in standard picture, obtain comparing the image scaling ratio of sample image described in described standard picture, as the image scaling ratio corresponding with this predeterminable range;

Based on described a plurality of different predeterminable ranges each self-corresponding z lens shaft side-play amount and image scaling ratio, set up reference object distance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio.

Fourth aspect, provides a kind of video image anti-shake apparatus, comprising:

The first skew adjustment amount determining unit, for the corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as the first image shift adjustment amount, wherein, described this rotation of lens angle is the rotation of lens angle of described pending video image when being taken by described capture apparatus;

The second skew adjustment amount determining unit, for the corresponding relation between reference object distance, camera lens side-play amount and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this camera lens side-play amount, as the second image shift adjustment amount, wherein, described this camera lens side-play amount is the camera lens side-play amount of described pending video image when being taken by described capture apparatus;

Anti-shake processing unit, for according to described the first image shift adjustment amount and described the second image shift adjustment amount, carries out the anti-shake migration processing of video image to described pending video image.

In conjunction with fourth aspect, in the possible implementation of the first, described the first skew adjustment amount, specifically for the reference object distance based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as an x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus;

Described the second skew adjustment amount, specifically for the reference object distance based on setting up in advance, along the x lens shaft side-play amount of x axle offset and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft side-play amount, as the 2nd x axle image shift adjustment amount, wherein, described this x lens shaft side-play amount is described pending video image x lens shaft side-play amount along x axle offset when being taken by described capture apparatus;

Described anti-shake processing unit, specifically for according to a described x axle image shift adjustment amount, a described y axle image shift adjustment amount, described the 2nd x axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image.

In conjunction with the possible implementation of the first of fourth aspect, in the possible implementation of the second, described anti-shake processing unit, specifically for according to a described x axle image shift adjustment amount and described the 2nd x axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd x axle image shift adjustment amount; And according to a described y axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd y axle image shift adjustment amount; And adopt described the 3rd x axle image shift adjustment amount and described the 3rd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake processing of video image.

The possible implementation of the second in conjunction with the possible implementation of the first of fourth aspect, fourth aspect or fourth aspect, in the third possible implementation, also comprises:

In conjunction with fourth aspect, in the 4th kind of possible mode, also comprise:

In conjunction with fourth aspect, in the 5th kind of possible mode, also comprise:

In conjunction with fourth aspect, in the 6th kind of possible mode, also comprise:

The 3rd sets up unit, for setting up in the following way the corresponding relation between reference object distance, camera lens side-play amount and image shift amount:

In conjunction with the possible implementation of the second of the possible implementation of the first of fourth aspect, fourth aspect, fourth aspect, the third possible implementation of fourth aspect, in the 7th kind of possible implementation, also comprises:

Image scaling ratio determining unit, for determining described pending video image z lens shaft side-play amount along z axle offset when being taken by described capture apparatus, as this z lens shaft side-play amount, wherein, the direction that the camera lens of described capture apparatus points to is identical with z direction of principal axis; And the distance of the reference object based on setting up in advance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio, determine the image scaling ratio corresponding with described this reference object distance and this z lens shaft side-play amount;

Described anti-shake processing unit, also for adopting the image scaling ratio of described correspondence, carries out the anti-shake convergent-divergent of video image to described pending video image and processes.

In conjunction with the 7th kind of possible implementation of fourth aspect, in the 8th kind of possible implementation, also comprise:

The 4th sets up unit, for setting up in the following way reference object distance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio:

The video image anti-fluttering method providing according to the third aspect, the video image anti-shake apparatus that fourth aspect provides, in carrying out the anti-shake process of video image, focusing position value according to pending image when focusing is taken, and the focusing position value of setting up in advance and the corresponding relation of reference object distance, determine this reference object distance, and according to this reference object distance and this rotation of lens angle when taking, and the reference object distance of setting up in advance, corresponding relation between rotation of lens angle and image shift amount, determine the first corresponding image shift adjustment amount, and according to this reference object distance and this camera lens side-play amount when taking, and the reference object distance of setting up in advance, corresponding relation between camera lens side-play amount and image shift amount, determine the second corresponding image shift adjustment amount, thereby while making to adopt this first image shift adjustment amount and the second image shift adjustment amount to carry out the anti-shake migration processing of video image to pending video image, the impact of rotation of lens angle difference while not only having considered to take on video image dithering, camera lens and distance different impacts on video image dithering between reference object while also having considered to take, and the impact of the camera lens skew while taking on video image dithering, thereby improved, carry out the anti-shake anti-shake effect of video image.

The application's further feature and advantage will be set forth in the following description, and, partly from specification, become apparent, or understand by implementing the application.The application's object and other advantages can be realized and be obtained by specifically noted structure in the specification write, claims and accompanying drawing.

Accompanying drawing explanation

Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for specification, is used from explanation the present invention with the embodiment of the present invention one, is not construed as limiting the invention.In the accompanying drawings:

One of flow chart of the video image anti-fluttering method that Fig. 1 provides for the embodiment of the present invention 1;

Fig. 2 is for adopting contrast mensuration to carry out the schematic diagram of focusing automatically;

Fig. 3 is the flow chart of setting up the corresponding relation of focusing position value and reference object distance in the embodiment of the present invention 1;

Fig. 4 is the schematic diagram of the corresponding relation curve of focusing position value and reference object distance in the embodiment of the present invention 1;

Fig. 5 is the flow chart of setting up the corresponding relation between reference object distance, rotation of lens angle and image shift amount in the embodiment of the present invention 1;

Fig. 6 is the schematic diagram that is related to of the coordinate system of x-axis, y-axis and z-axis foundation in the embodiment of the present invention 1 and the camera lens sensing of capture apparatus;

Fig. 7 is image shift amount P in the embodiment of the present invention 1 _gand the linear relationship schematic diagram between rotation of lens angle θ;

Fig. 8 is coefficient f in the embodiment of the present invention 1 _gand the schematic diagram of the relation curve between reference object distance;

Two of the flow chart of the video image anti-fluttering method that Fig. 9 provides for the embodiment of the present invention 1;

One of flow chart of the video image anti-fluttering method that Figure 10 provides for the embodiment of the present invention 2;

Figure 11 is the flow chart of setting up the corresponding relation between reference object distance, camera lens side-play amount and image shift amount in the embodiment of the present invention 2;

Figure 12 is image shift amount P in the embodiment of the present invention 2 _aand the linear relationship schematic diagram between camera lens side-play amount S;

Figure 13 is coefficient f in the embodiment of the present invention 2 _aand the schematic diagram of the relation curve between reference object distance;

Two of the flow chart of the video image anti-fluttering method that Figure 14 provides for the embodiment of the present invention 2;

The structural representation of the video image anti-shake apparatus that Figure 15 provides for the embodiment of the present invention 3;

The structural representation of the video image anti-shake apparatus that Figure 16 provides for the embodiment of the present invention 4.

Embodiment

In order to provide, improve the implementation that carries out the anti-shake anti-shake effect of video image, the embodiment of the present invention provides a kind of video image anti-fluttering method and device, below in conjunction with Figure of description, the preferred embodiments of the present invention are described, be to be understood that, preferred embodiment described herein only, for description and interpretation the present invention, is not intended to limit the present invention.And in the situation that not conflicting, embodiment and the feature in embodiment in the application can combine mutually.

Embodiment 1:

The embodiment of the present invention 1 provides a kind of video image anti-fluttering method, as shown in Figure 1, comprising:

Step 101, obtain the focusing position value of pending video image when being taken by capture apparatus focusing, as this focusing position value.

Step 102, the focusing position value based on setting up in advance and the corresponding relation of reference object distance, determine the reference object distance corresponding with this focusing position value, as this reference object distance.

Corresponding relation between step 103, the distance of the reference object based on setting up in advance, rotation of lens angle and image shift amount, determine the image shift amount corresponding with this reference object distance and this rotation of lens angle, as image shift adjustment amount, wherein, this rotation of lens angle is the rotation of lens angle of pending video image when being taken by capture apparatus.

Step 104, employing image shift adjustment amount, carry out the anti-shake migration processing of video image to pending video image.

Further, on the basis of method shown in above-mentioned Fig. 1, can also carry out anti-shake rotation processing to this pending video image, specifically can be as follows:

Determine pending video image z lens shaft anglec of rotation along the rotation of z axle when being taken by capture apparatus, as image rotation adjustment amount, wherein, the direction that the camera lens of capture apparatus points to is identical with z direction of principal axis;

Adopt image rotation adjustment amount, pending video image is carried out to the anti-shake rotation of video image and process.

In the step 102 of said method shown in Fig. 1, focusing position value based on setting up in advance and the corresponding relation of reference object distance, determining the reference object distance corresponding with this focusing position value, first to how setting up the corresponding relation of focusing position value and reference object distance, being described in detail below.

At present, the method for the automatic focusing (auto focus) that capture apparatus adopts when focusing is taken is a lot, is mainly divided into: light triangulation, phase-detection method, contrast mensuration etc.No matter which kind of method can provide data that represent camera lens current location after focusing finishes, and this position data is at a distance of v, and, the distance at a distance of v, focal distance f and object distance u(camera lens to reference object) between relation, meet following formula:

1/u+1/v＝1/f；

When focal distance f is fixedly time, object distance u and be one to one at a distance of v, therefore, during shooting, reference object is also just corresponding to different apart from difference.

For example, for the contrast mensuration generally adopting in current automatic focusing, as shown in Figure 2, the automatic focusing based on contrast mensuration, is by program, to measure the contrast of present image, according to the increase of contrast, reduce direction adjustable lens position, finally find the place that contrast is the highest, so can realize focusing clearly, in Fig. 2, abscissa is the data that represent lens location, as at a distance of v.

Wherein, at present in actual applications, what represent lens location can be apart the value within the scope of 0-1024, now may also be referred to as focusing position value apart, after using capture apparatus to take, for every width video image, capture apparatus can collect the focusing position value when this width video image is taken in focusing.

When setting up the corresponding relation of focusing position value and reference object distance for capture apparatus, first use capture apparatus respectively in the position of a plurality of different predeterminable ranges of distance of camera lens reference object, default object is taken in focusing, for example, the plurality of different predeterminable ranges can be 10cm, 20cm, 30cm, 50cm, 80cm, 100cm, 150cm, 200cm, 250cm, 300cm etc.

In the process of taking, for every width video image of taking, capture apparatus can gather focusing position value automatically, after shooting, can adopt flow process as shown in Figure 3 to set up the corresponding relation of focusing position value and reference object distance, specifically comprise following treatment step:

Step 301, for each predeterminable range in a plurality of different predeterminable ranges, obtain capture apparatus in the position of this predeterminable range of distance of camera lens reference object, focusing position value when default object is taken in focusing, as the focusing position value corresponding with this predeterminable range.

Step 302, based on each self-corresponding focusing position value of the plurality of different predeterminable ranges, set up the corresponding relation of focusing position value and reference object distance.

In this step, the focusing position value of setting up and the corresponding relation of reference object distance, can be discrete a plurality of focusing position values and the mapping table of reference object distance, also can be to carry out curve fitting according to each self-corresponding focusing position value of the plurality of different predeterminable ranges, the focusing position value obtaining and the corresponding relation curve of reference object distance, this corresponding relation curve can as shown in Figure 4, also can represent by following formula:

D＝f(position)；

Wherein, D is reference object distance, and position is focusing position value, and the span of position is 0-1024.

In the step 103 of said method shown in Fig. 1, corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with this reference object distance and this rotation of lens angle, first the corresponding relation of how setting up between reference object distance, rotation of lens angle and image shift amount is described in detail below.

When the corresponding relation of setting up for capture apparatus between reference object distance, rotation of lens angle and image shift amount, first use capture apparatus respectively in the position of a plurality of different predeterminable ranges of distance of camera lens reference object, default object is taken in focusing, for example, the plurality of different predeterminable ranges can be 10cm, 20cm, 30cm, 50cm, 80cm, 100cm, 150cm, 200cm, 250cm, 300cm etc.

Preferably, the same shooting process in the time of can using with the above-mentioned corresponding relation of setting up focusing position value and reference object distance.

In the process of taking, for every width video image of taking, capture apparatus can gather rotation of lens angular speed automatically, after shooting, can adopt flow process as shown in Figure 5 to set up the corresponding relation between reference object distance, rotation of lens angle and image shift amount, specifically comprise following treatment step:

Step 501, for each predeterminable range in a plurality of different predeterminable ranges, obtain capture apparatus in the position of this predeterminable range of distance of camera lens reference object, the rotation of lens angular speed gathering while taking default object, as the rotation of lens angular speed corresponding with this predeterminable range.

In this step, rotation of lens angular speed when capture apparatus can be used gyroscope collection shooting wherein.

Further, specifically can gather along the x lens shaft angular velocity of rotation of x axle rotation, and the y lens shaft angular velocity of rotation rotating along y axle.

Further, can also gather along the z lens shaft angular velocity of rotation of z axle rotation.

Wherein, as shown in Figure 6, the direction that the camera lens of capture apparatus points to is vertical with the plane that y axle forms with x axle, and the direction that the camera lens of capture apparatus points to is identical with z direction of principal axis.

In this step, because gyrostatic sample rate is generally higher than the frame per second of video capture, within the time of video one frame, gyroscope can complete multi collect, so, can adopt following formula to determine the mean value of a plurality of rotation of lens angular speed that multi collect obtains in video one frame time, as the rotation of lens angular speed corresponding with this predeterminable range:

G _avg＝ΣG _i/N；

Wherein, N is the quantity of the rotation of lens angular speed that collects in video one frame time, G _avgfor the mean value of N rotation of lens angular speed, G _ifor i rotation of lens angular speed in N rotation of lens angular speed.

Accordingly, the rotation of lens angular speed G obtaining in this step _avgcan comprise G _x, G _yand G _z, correspond to respectively x lens shaft angular velocity of rotation, y lens shaft angular velocity of rotation and z lens shaft angular velocity of rotation.

Step 502, for each predeterminable range in a plurality of different predeterminable ranges, according to the rotation of lens angular speed corresponding with this predeterminable range, determine the rotation of lens angle corresponding with this predeterminable range.

Because rotation of lens angle θ when taking is less, θ ≈ sin (θ) ≈ tan (θ) now, so, in this step, can adopt following formula to determine rotation of lens angle:

θ＝G _avg*Δt；

Wherein, Δ t is the time interval between adjacent two frames of video.

Accordingly, as rotation of lens angular speed G _avgcomprise x lens shaft angular velocity of rotation G _x, y lens shaft angular velocity of rotation G _ywith z lens shaft angular velocity of rotation G _ztime, adopt following formula to determine x lens shaft anglec of rotation θ _x, y lens shaft anglec of rotation θ _ywith z lens shaft anglec of rotation θ _z:

θ _x＝G _x*Δt；

θ _y＝G _y*Δt；

θ _z＝G _z*Δt。

The determined z lens shaft anglec of rotation in this step, the image rotation adjustment amount using in the time of can directly processing as subsequent video images is anti-shake.

Step 503, for each predeterminable range, relatively capture apparatus takes in the position of this predeterminable range of distance of camera lens reference object the sample image that default object obtains, standard picture with default object, obtain the position of default object in comparison with standard image, the image shift amount of the position of default object in sample image, as the image shift amount corresponding with this predeterminable range.

In this step, the standard picture that comprises default object can be that capture apparatus is taken the image that default object obtains in non-jitter situation at this predeterminable range.

For the ease of determining image shift amount, can construct the clearly default object of central point, and the position in sample image and the skew between the position in standard picture by this central point relatively, determine image shift amount, can be more definite by manpower comparing, also can be automatically more definite by capture apparatus.

In this step, the determined image shift amount P corresponding with this predeterminable range _g, can comprise x axle image shift amount P _gxwith y axle image shift amount P _gy.

Step 504, based on a plurality of different predeterminable ranges each self-corresponding rotation of lens angle and image shift amount, set up the corresponding relation between reference object distance, rotation of lens angle and image shift amount.

In this step, can be to carry out curve fitting according to each self-corresponding rotation of lens angle of the plurality of different predeterminable ranges and image shift amount, obtain the corresponding relation curve between reference object distance, rotation of lens angle and image shift amount, this form that comprises the corresponding relation curve of three parameters can adopt various curve forms, for example, can adopt the represented relation curve of following formula:

P _G＝f _G(D)*θ；

From the formula of above-mentioned expression relation curve, image shift amount P _gand between rotation of lens angle θ, as shown in Figure 7, be linear relationship, f _gfor coefficient.

And, coefficient f _g=f _g(D), i.e. f _gand meet relation curve as shown in Figure 8 between reference object distance.

In this step, specifically can set up reference object distance, the x lens shaft anglec of rotation of rotating along x axle and the corresponding relation between x axle image shift amount, corresponding relation curve is as follows:

P _Gx＝f _Gx(D)*θ _x；

And set up reference object distance, the y lens shaft anglec of rotation of rotating along y axle and the corresponding relation between y axle image shift amount, corresponding relation curve is as follows:

P _Gx＝f _Gx(D)*θ _x。

By the flow process of above-mentioned Fig. 3 and Fig. 5, set up in advance the corresponding relation of focusing position value and reference object distance, and after the corresponding relation between the reference object of setting up in advance distance, rotation of lens angle and image shift amount, corresponding relation that can be based on setting up, the video image that uses this capture apparatus to take is carried out to anti-shake processing, specifically can adopt flow process as shown in Figure 9, comprise:

Step 901, the pending video image obtaining for use capture apparatus focusing shooting, obtain the focusing position value of this pending video image when focusing is taken, as this focusing position value.

Step 902, the focusing position value based on setting up in advance and the corresponding relation of reference object distance, determine the reference object distance corresponding with this focusing position value, as this reference object distance.

When the corresponding relation of focusing position value and reference object distance is mapping table, can determine by tabling look-up corresponding reference object distance, while there is not this focusing position value in this mapping table, can be according to two focusing position values adjacent with this focusing position value in this mapping table, and these two adjacent reference object distances corresponding to focusing position values difference, carry out linear interpolation, obtain the reference object distance corresponding with this focusing position value, mode is no longer described in detail at this in detail.

When the corresponding relation of focusing position value and reference object distance is corresponding relation curve, can, directly by the formula D=f (position) in the above-mentioned steps 302 of this corresponding relation curve of expression, determine the reference object distance corresponding with this focusing position value.

Step 903, obtain the rotation of lens angular speed of this pending video image when being taken by capture apparatus, specifically can obtain x lens shaft angular velocity of rotation, y lens shaft angular velocity of rotation and z lens shaft angular velocity of rotation.

Within the rotation of lens angular speed obtaining comprises a frame video time, gather a plurality of time, can also determine the mean value of a plurality of rotation of lens angular speed, as this rotation of lens angular speed, comprise this x lens shaft angular velocity of rotation, this y lens shaft angular velocity of rotation and this z lens shaft angular velocity of rotation.

Product between the time interval between step 904, definite this rotation of lens angular speed two frames adjacent with video that obtain, result, as this rotation of lens angle, specifically can be determined this x lens shaft anglec of rotation, this y lens shaft anglec of rotation and this z lens shaft anglec of rotation.

Wherein, determined this z lens shaft anglec of rotation is directly as image rotation adjustment amount Rotation.

Above-mentioned steps 903 and step 904, and between above-mentioned steps 901 and step 902, there is no strict sequencing.

Corresponding relation between step 905, the distance of the reference object based on setting up in advance, rotation of lens angle and image shift amount, determines the image shift amount corresponding with this reference object distance and this rotation of lens angle, as image shift adjustment amount.

In this step, reference object distance that specifically can be based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with this reference object distance and this x lens shaft anglec of rotation, as x axle image shift adjustment amount p _x;

And the distance of the reference object based on setting up in advance, the y lens shaft anglec of rotation of rotating along y axle and the corresponding relation between y axle image shift amount, determine the y axle image shift amount corresponding with this reference object distance and this y lens shaft anglec of rotation, as y axle image shift adjustment amount P _y.

Step 906, employing image shift adjustment amount and image rotation adjustment amount, carry out the anti-shake migration processing of video image and the anti-shake rotation processing of video image to this pending video image.

The anti-shake processing of video image in this step, can comprise and adopt x axle image shift adjustment amount p _xwith y axle image shift adjustment amount P _y, pending video image is carried out to the anti-shake migration processing of video image, and adopt image rotation adjustment amount Rotation, this pending video image is carried out to the anti-shake rotation of video image and process.

Concrete, can be first for this pending video image, by this pending image along this x axle image shift adjustment amount of x axle offset p _x, image is carried out to the translational motion compensation along x axle, along this y axle image shift adjustment amount of y axle offset P _y, image is carried out to the translational motion compensation along y axle, along z axle, rotate this image rotation adjustment amount Rotation, image is rotated to compensation, be compensated and process rear image;

Then, from these compensation deals, in image, shear the image of predeterminable area, as the video image after anti-shake processing.

The above-mentioned video image anti-fluttering method that adopts the embodiment of the present invention 1 to provide, when video image is carried out to anti-shake processing, the impact of rotation of lens angle difference while not only having considered to take on video image dithering, while also having considered to take, camera lens and different impacts on video image dithering of distance between reference object, carry out the anti-shake anti-shake effect of video image thereby improved.

Embodiment 2:

The embodiment of the present invention 2 provides a kind of video image anti-fluttering method, as shown in figure 10, comprising:

Step 1001, obtain the focusing position value of pending video image when being taken by capture apparatus focusing, as this focusing position value.

Step 1002, the focusing position value based on setting up in advance and the corresponding relation of reference object distance, determine the reference object distance corresponding with this focusing position value, as this reference object distance.

Corresponding relation between step 1003, the distance of the reference object based on setting up in advance, rotation of lens angle and image shift amount, determine the image shift amount corresponding with this reference object distance and this rotation of lens angle, as the first image shift adjustment amount, wherein, this rotation of lens angle is the rotation of lens angle of pending video image when being taken by capture apparatus.

Corresponding relation between step 1004, the distance of the reference object based on setting up in advance, camera lens side-play amount and image shift amount, determine the image shift amount corresponding with this reference object distance and this camera lens side-play amount, as the second image shift adjustment amount, wherein, this camera lens side-play amount is the camera lens side-play amount of pending video image when being taken by capture apparatus.

Step 1005, according to the first image shift adjustment amount and the second image shift adjustment amount, pending video image is carried out to the anti-shake migration processing of video image.

Between above-mentioned steps 1003 and step 1004, there is no strict sequencing.

Further, on the basis of method shown in above-mentioned Figure 10, can also carry out anti-shake rotation processing to this pending video image, specifically can be as follows:

Then, adopt image rotation adjustment amount, pending video image is carried out to the anti-shake rotation of video image and process.

Further, on the basis of method shown in above-mentioned Figure 10, can also carry out anti-shake convergent-divergent processing to this pending video image, specifically can be as follows:

Determine pending video image z lens shaft side-play amount along z axle offset when being taken by capture apparatus, as this z lens shaft side-play amount, wherein, the direction that the camera lens of capture apparatus points to is identical with z direction of principal axis;

Reference object distance based on setting up in advance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio, determine the image scaling ratio corresponding with this reference object distance and this z lens shaft side-play amount;

Adopt corresponding image scaling ratio, pending video image is carried out to the anti-shake convergent-divergent of video image and process.

In the embodiment of the present invention 2, the focusing position value of using in the step 1002 of said method shown in Figure 10 and the corresponding relation of reference object distance, can adopt with mode identical shown in Fig. 2-Fig. 4 in above-described embodiment 1 and set up in advance, at this, no longer be described in detail.

In the embodiment of the present invention 2, corresponding relation between reference object distance, rotation of lens angle and the image shift amount of using in the step 1003 of said method shown in Figure 10, can adopt with mode identical shown in Fig. 5-Fig. 8 in above-described embodiment 1 and set up in advance, at this, no longer be described in detail.

In the step 1004 of said method shown in Figure 10, corresponding relation between reference object distance, camera lens side-play amount and image shift amount based on setting up in advance, determine the image shift amount corresponding with this reference object distance and this camera lens side-play amount, first the corresponding relation of how setting up between reference object distance, camera lens side-play amount and image shift amount is described in detail below.

When the corresponding relation of setting up for capture apparatus between reference object distance, camera lens side-play amount and image shift amount, first use capture apparatus respectively in the position of a plurality of different predeterminable ranges of distance of camera lens reference object, default object is taken in focusing, for example, the plurality of different predeterminable ranges can be 10cm, 20cm, 30cm, 50cm, 80cm, 100cm, 150cm, 200cm, 250cm, 300cm etc.

Preferably, can use with set up focusing position value and reference object distance corresponding relation time same shooting process.

In the process of taking, for every width video image of taking, capture apparatus can gather camera lens offset acceleration automatically, after shooting, can adopt flow process as shown in figure 11 to set up the corresponding relation between reference object distance, camera lens side-play amount and image shift amount, specifically comprise following treatment step:

Step 1101, for each predeterminable range in a plurality of different predeterminable ranges, obtain capture apparatus in the position of this predeterminable range of distance of camera lens reference object, the camera lens offset acceleration gathering while taking default object, as the camera lens offset acceleration corresponding with this predeterminable range.

In this step, camera lens offset acceleration when capture apparatus can be used linear accelerometer collection shooting wherein.

Further, specifically can gather the x lens shaft offset acceleration along x axle offset, and along the y lens shaft offset acceleration of y axle offset.

Further, can also gather the z lens shaft offset acceleration along z axle offset.

In this step, because the sample rate of linear accelerometer is generally higher than the frame per second of video capture, within the time of video one frame, linear accelerometer can complete multi collect, so, can adopt following formula to determine the mean value of a plurality of camera lens offset acceleration that multi collect obtains in video one frame time, as the camera lens offset acceleration corresponding with this predeterminable range:

A _avg＝ΣA _i/M；

Wherein, M is the quantity of the camera lens offset acceleration that collects in video one frame time, A _avgfor the mean value of M camera lens offset acceleration, A _ifor i camera lens offset acceleration in M camera lens offset acceleration.

Accordingly, the camera lens offset acceleration A obtaining in this step _avgcan comprise A _x, A _yand A _z, correspond to respectively x lens shaft offset acceleration, y lens shaft offset acceleration and z lens shaft offset acceleration.

Step 1102, for each predeterminable range in a plurality of different predeterminable ranges, according to the camera lens offset acceleration corresponding with this predeterminable range, determine the camera lens side-play amount corresponding with this predeterminable range.

In this step, can adopt following formula to determine camera lens side-play amount:

S＝A _avg*Δt ²/2；

Wherein, S is camera lens side-play amount, and Δ t is the time interval between adjacent two frames of video.

Accordingly, as camera lens offset acceleration A _avgcomprise x lens shaft offset acceleration A _x, y lens shaft offset acceleration A _ywith z lens shaft offset acceleration A _ztime, adopt following formula to determine x lens shaft side-play amount S _x, y lens shaft side-play amount S _ywith z lens shaft side-play amount S _z:

S _x＝A _x*Δt ²/2；

S _y＝A _y*Δt ²/2；

S _z＝A _z*Δt ²/2。

Step 1103, for each predeterminable range, relatively capture apparatus takes in the position of this predeterminable range of distance of camera lens reference object the sample image that default object obtains, standard picture with default object, obtain the position of default object in comparison with standard image, the image shift amount of the position of default object in sample image, as the image shift amount corresponding with this predeterminable range.

In this step, the determined image shift amount P corresponding with this predeterminable range _a, can comprise x axle image shift amount P _axwith y axle image shift amount P _ay.

Step 1104, based on a plurality of different predeterminable ranges each self-corresponding camera lens side-play amount and image shift amount, set up the corresponding relation between reference object distance, camera lens side-play amount and image shift amount.

In this step, can be to carry out curve fitting according to each self-corresponding camera lens side-play amount of the plurality of different predeterminable ranges and image shift amount, obtain the corresponding relation curve between reference object distance, camera lens side-play amount and image shift amount, this form that comprises the corresponding relation curve of three parameters can adopt various curve forms, for example, can adopt the represented relation curve of following formula:

P _A＝f _A(D)*S；

From the formula of above-mentioned expression relation curve, image shift amount P _aand between camera lens side-play amount S, as shown in figure 12, be linear relationship, f _afor coefficient.

And, coefficient f _a=f _a(D), i.e. f _aand meet relation curve as shown in figure 13 between reference object distance.

In this step, specifically can set up reference object distance, along the x lens shaft side-play amount of x axle offset and the corresponding relation between x axle image shift amount, corresponding relation curve is as follows:

P _Ax＝f _Ax(D)*S _x；

And set up reference object distance, along the y lens shaft side-play amount of y axle offset and the corresponding relation between y axle image shift amount, corresponding relation curve is as follows:

P _Ax＝f _Ax(D)*S _x。

Further, when also comprising that when carrying out the anti-shake processing of video image anti-shake convergent-divergent is processed, can also adopt following steps 1105-step 1106 set up reference object distance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio:

Step 1105, for each predeterminable range, determine that capture apparatus takes the area of default object in the sample image that default object obtains in the position of this predeterminable range of distance of camera lens reference object, ratio with the area of default object in standard picture, obtain the image scaling ratio of this sample image of comparison with standard image, as the image scaling ratio corresponding with this predeterminable range.

Standard picture in this step can be with above-mentioned steps 1103 identical, can be that capture apparatus is taken the image that default object obtains in non-jitter situation at this predeterminable range.

For the ease of determining the area of default object in image, can construct the default object of rectangle, and by the ratio of the area of this default object in the area of this default object in sample image and sample image, be defined as the image scaling ratio R atio corresponding with this predeterminable range, can be more definite by manpower comparing, also can be automatically more definite by capture apparatus.

Step 1106, based on a plurality of different predeterminable ranges each self-corresponding z lens shaft side-play amount and image scaling ratio, set up reference object distance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio.

In this step, can be to carry out curve fitting according to each self-corresponding z lens shaft side-play amount of the plurality of different predeterminable ranges and image scaling ratio, obtain the corresponding relation curve between reference object distance, z lens shaft side-play amount and image scaling ratio, this form that comprises the corresponding relation curve of three parameters can adopt various curve forms, for example, can adopt the represented relation curve of following formula:

Ratio＝f _Az(D)*S _z；

From the formula of above-mentioned expression relation curve, image scaling ratio R atio and z lens shaft side-play amount S _zbetween be linear relationship, f _azfor coefficient.

And, coefficient f _az=f _az(D), i.e. f _azand between reference object distance, meet the relation curve of matching.

Passing through above-mentioned Fig. 3, the flow process of Fig. 5 and Figure 11, set up in advance the corresponding relation of focusing position value and reference object distance, and the reference object distance of setting up in advance, corresponding relation between rotation of lens angle and image shift amount, and set up in advance reference object distance, corresponding relation between camera lens side-play amount and image shift amount, and set up in advance reference object distance, after the z lens shaft side-play amount and image scaling ratio of z axle offset, corresponding relation that can be based on setting up, the video image that uses this capture apparatus to take is carried out to anti-shake processing, specifically can adopt flow process as shown in figure 14, comprise:

Step 1401, the pending video image obtaining for use capture apparatus focusing shooting, obtain the focusing position value of this pending video image when focusing is taken, as this focusing position value.

Step 1402, the focusing position value based on setting up in advance and the corresponding relation of reference object distance, determine the reference object distance corresponding with this focusing position value, as this reference object distance.

Step 1403, obtain the rotation of lens angular speed of this pending video image when being taken by capture apparatus, specifically can obtain x lens shaft angular velocity of rotation, y lens shaft angular velocity of rotation and z lens shaft angular velocity of rotation.

Product between the time interval between step 1404, definite this rotation of lens angular speed two frames adjacent with video that obtain, result, as this rotation of lens angle, specifically can be determined this x lens shaft anglec of rotation, this y lens shaft anglec of rotation and this z lens shaft anglec of rotation.

Above-mentioned steps 1403 and step 1404, and between above-mentioned steps 1401 and step 1402, there is no strict sequencing.

Corresponding relation between step 1405, the distance of the reference object based on setting up in advance, rotation of lens angle and image shift amount, determines the image shift amount corresponding with this reference object distance and this rotation of lens angle, as the first image shift adjustment amount.

In this step, reference object distance that specifically can be based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with this reference object distance and this x lens shaft anglec of rotation, as an x axle image shift adjustment amount p _x1;

And the distance of the reference object based on setting up in advance, the y lens shaft anglec of rotation of rotating along y axle and the corresponding relation between y axle image shift amount, determine the y axle image shift amount corresponding with this reference object distance and this y lens shaft anglec of rotation, as a y axle image shift adjustment amount P _y1.

Step 1406, obtain the camera lens offset acceleration of this pending video image when being taken by capture apparatus, specifically can obtain x lens shaft offset acceleration, y lens shaft offset acceleration and z lens shaft offset acceleration.

Within the camera lens offset acceleration of obtaining comprises a frame video time, gather a plurality of time, can also determine the mean value of a plurality of camera lens offset acceleration, as this camera lens offset acceleration, comprise this x lens shaft offset acceleration, this y lens shaft offset acceleration and this z lens shaft offset acceleration.

Step 1407, determine time interval between this camera lens offset acceleration two frames adjacent with video obtain square between product, and this product is divided by 2 quotient, result, as this camera lens side-play amount, specifically can be determined this x lens shaft side-play amount, this y lens shaft side-play amount and this z lens shaft side-play amount.

Corresponding relation between step 1408, the distance of the reference object based on setting up in advance, camera lens side-play amount and image shift amount, determines the image shift amount corresponding with described this reference object distance and this camera lens side-play amount, as the second image shift adjustment amount.

In this step, reference object distance that specifically can be based on setting up in advance, along the x lens shaft side-play amount of x axle offset and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with this reference object distance and this x lens shaft side-play amount, as the 2nd x axle image shift adjustment amount P _x2;

And the distance of the reference object based on setting up in advance, along the y lens shaft side-play amount of y axle offset and the corresponding relation between y axle image shift amount, determine the y axle image shift amount corresponding with this reference object distance and this y lens shaft side-play amount, as the 2nd y axle image shift adjustment amount P _y2.

Step 1409, the distance of the reference object based on setting up in advance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio, determine the image scaling ratio R atio corresponding with this reference object distance and this z lens shaft side-play amount.

Above-mentioned steps 1406-step 1409, and between above-mentioned steps 1403-step 1405, there is no strict sequencing.

Step 1410, according to the first image shift adjustment amount and the second image shift adjustment amount, pending video image is carried out to the anti-shake migration processing of video image, and employing image rotation adjustment amount, pending video image is carried out to the anti-shake rotation of video image to be processed, and adopt corresponding image scaling ratio, pending video image is carried out to the anti-shake convergent-divergent of video image and process.

First the anti-shake processing of video image in this step, can adopt an x axle image shift adjustment amount P _x1with the 2nd x axle image shift adjustment amount P _x2, pending video image is carried out to the anti-shake migration processing of video image, specifically can be as follows:

According to an x axle image shift adjustment amount P _x1with described the 2nd x axle image shift adjustment amount P _x2, and each self-corresponding weights W _x1and W _x2, be weighted summation, obtain the 3rd x axle image shift adjustment amount P _x:

P _x＝W _x1*P _x1+W _x2*P _x2；

And by this pending image along x axle offset the 3rd x axle image shift adjustment amount P _x, image is carried out to the translational motion compensation along x axle;

Then, adopt a y axle image shift adjustment amount P _y1with the 2nd y axle image shift adjustment amount P _y2, pending video image is carried out to the anti-shake migration processing of video image, specifically can be as follows:

According to a y axle image shift adjustment amount P _y1with described the 2nd y axle image shift adjustment amount P _y2, and each self-corresponding weights W _y1and W _y2, be weighted summation, obtain the 3rd y axle image shift adjustment amount P _y:

P _y＝W _y1*P _y1+W _y2*P _y2；

And by this pending image along y axle offset the 3rd y axle image shift adjustment amount P _y, image is carried out to the translational motion compensation along y axle;

And adopt image rotation adjustment amount Rotation, and this pending video image is carried out to the anti-shake rotation of video image and process, along z axle, rotate this image rotation adjustment amount Rotation, image is rotated to compensation;

And adopt corresponding image scaling ratio R atio, and pending video image is carried out to the anti-shake convergent-divergent of video image and process, to this pending video image zooming, image is carried out to convergent-divergent compensation;

After above-mentioned processing, be compensated and process rear image.

The above-mentioned video image anti-fluttering method that adopts the embodiment of the present invention 2 to provide, when video image is carried out to anti-shake processing, the impact of rotation of lens angle difference while not only having considered to take on video image dithering, camera lens and distance different impacts on video image dithering between reference object while also having considered to take, and the impact of the camera lens skew while taking on video image dithering, thereby improved, carry out the anti-shake anti-shake effect of video image.

And, further, it is also conceivable that the impact of along the camera lens skew of z axle, video image zooming being shaken while taking, thereby further improved, carry out the anti-shake anti-shake effect of video image.

Embodiment 3:

Based on same inventive concept, the video image anti-fluttering method providing according to the above embodiment of the present invention, correspondingly, the embodiment of the present invention 3 also provides a kind of video image anti-shake apparatus, and its structural representation as shown in figure 15, specifically comprises:

Focusing position value acquiring unit 1501, for obtaining the focusing position value of pending video image when being taken by capture apparatus focusing, as this focusing position value;

Distance determining unit 1502, for the focusing position value based on setting up in advance and the corresponding relation of reference object distance, determines the reference object distance corresponding with described this focusing position value, as this reference object distance;

Skew adjustment amount determining unit 1503, for the corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as image shift adjustment amount, wherein, described this rotation of lens angle is the rotation of lens angle of described pending video image when being taken by described capture apparatus;

Anti-shake processing unit 1504, for adopting described image shift adjustment amount, carries out the anti-shake migration processing of video image to described pending video image.

Further, skew adjustment amount determining unit 1503, specifically for the reference object distance based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus; And

Anti-shake processing unit 1504, specifically for adopting described x axle image shift adjustment amount and described y axle image shift adjustment amount, carries out the anti-shake migration processing of video image to described pending video image.

Further, above-mentioned video image anti-shake apparatus, also comprises:

Rotation adjustment amount determining unit 1505, for determining described pending video image z lens shaft anglec of rotation along the rotation of z axle when being taken by described capture apparatus, as image rotation adjustment amount, wherein, the direction that the camera lens of described capture apparatus points to is identical with z direction of principal axis;

Anti-shake processing unit 1504, also for adopting described image rotation adjustment amount, carries out the anti-shake rotation of video image to described pending video image and processes.

Further, above-mentioned video image anti-shake apparatus, also comprises:

First sets up unit 1506, for setting up in the following way the corresponding relation of focusing position value and reference object distance:

Further, above-mentioned video image anti-shake apparatus, also comprises:

Second sets up unit 1507, for setting up in the following way the corresponding relation between reference object distance, rotation of lens angle and image shift amount:

The function of above-mentioned each unit can, corresponding to the respective handling step in flow process shown in Fig. 1, Fig. 3, Fig. 5 and Fig. 9, not repeat them here.

Embodiment 4:

Based on same inventive concept, the video image anti-fluttering method providing according to the above embodiment of the present invention, correspondingly, the embodiment of the present invention 4 also provides a kind of video image anti-shake apparatus, and its structural representation as shown in figure 16, specifically comprises:

Focusing position value acquiring unit 1601, for obtaining the focusing position value of pending video image when being taken by capture apparatus focusing, as this focusing position value;

Distance determining unit 1602, for the focusing position value based on setting up in advance and the corresponding relation of reference object distance, determines the reference object distance corresponding with described this focusing position value, as this reference object distance;

The first skew adjustment amount determining unit 1603, for the corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as the first image shift adjustment amount, wherein, described this rotation of lens angle is the rotation of lens angle of described pending video image when being taken by described capture apparatus;

The second skew adjustment amount determining unit 1604, for the corresponding relation between reference object distance, camera lens side-play amount and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this camera lens side-play amount, as the second image shift adjustment amount, wherein, described this camera lens side-play amount is the camera lens side-play amount of described pending video image when being taken by described capture apparatus;

Anti-shake processing unit 1605, for according to described the first image shift adjustment amount and described the second image shift adjustment amount, carries out the anti-shake migration processing of video image to described pending video image.

Further the first skew adjustment amount determining unit 1603, specifically for the reference object distance based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as an x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus;

The second skew adjustment amount determining unit 1604, specifically for the reference object distance based on setting up in advance, along the x lens shaft side-play amount of x axle offset and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft side-play amount, as the 2nd x axle image shift adjustment amount, wherein, described this x lens shaft side-play amount is described pending video image x lens shaft side-play amount along x axle offset when being taken by described capture apparatus;

Anti-shake processing unit 1605, specifically for according to a described x axle image shift adjustment amount, a described y axle image shift adjustment amount, described the 2nd x axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image.

Further, anti-shake processing unit 1605, specifically for according to a described x axle image shift adjustment amount and described the 2nd x axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd x axle image shift adjustment amount; And according to a described y axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd y axle image shift adjustment amount; And adopt described the 3rd x axle image shift adjustment amount and described the 3rd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake processing of video image.

Further, above-mentioned video image anti-shake apparatus, also comprises:

Rotation adjustment amount determining unit 1606, for determining described pending video image z lens shaft anglec of rotation along the rotation of z axle when being taken by described capture apparatus, as image rotation adjustment amount, wherein, the direction that the camera lens of described capture apparatus points to is identical with z direction of principal axis;

Anti-shake processing unit 1605, also for adopting described image rotation adjustment amount, carries out the anti-shake rotation of video image to described pending video image and processes.

Further, above-mentioned video image anti-shake apparatus, also comprises:

First sets up unit 1607, for setting up in the following way the corresponding relation of focusing position value and reference object distance:

Further, above-mentioned video image anti-shake apparatus, also comprises:

Second sets up unit 1608, for setting up in the following way the corresponding relation between reference object distance, rotation of lens angle and image shift amount:

Further, above-mentioned video image anti-shake apparatus, also comprises:

The 3rd sets up unit 1609, for setting up in the following way the corresponding relation between reference object distance, camera lens side-play amount and image shift amount:

Further, above-mentioned video image anti-shake apparatus, also comprises:

Image scaling ratio determining unit 1610, for determining described pending video image z lens shaft side-play amount along z axle offset when being taken by described capture apparatus, as this z lens shaft side-play amount, wherein, the direction that the camera lens of described capture apparatus points to is identical with z direction of principal axis; And the distance of the reference object based on setting up in advance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio, determine the image scaling ratio corresponding with described this reference object distance and this z lens shaft side-play amount;

Anti-shake processing unit 1605, also for adopting the image scaling ratio of described correspondence, carries out the anti-shake convergent-divergent of video image to described pending video image and processes.

Further, above-mentioned video image anti-shake apparatus, also comprises:

The 4th sets up unit 1611, for setting up in the following way reference object distance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio:

The function of above-mentioned each unit can, corresponding to the respective handling step in flow process shown in Figure 10, Figure 11 and Figure 14, not repeat them here.

The video image anti-shake apparatus that the application's embodiment provides can be realized by computer program.Those skilled in the art should be understood that; above-mentioned Module Division mode is only a kind of in numerous Module Division modes; if be divided into other modules or do not divide module, as long as video image anti-shake apparatus has above-mentioned functions, all should be within the application's protection range.

The application is with reference to describing according to flow chart and/or the block diagram of the method for the embodiment of the present application, equipment (system) and computer program.Should understand can be in computer program instructions realization flow figure and/or block diagram each flow process and/or the flow process in square frame and flow chart and/or block diagram and/or the combination of square frame.Can provide these computer program instructions to the processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing device to produce a machine, the instruction of carrying out by the processor of computer or other programmable data processing device is produced for realizing the device in the function of flow process of flow chart or a plurality of flow process and/or square frame of block diagram or a plurality of square frame appointments.

These computer program instructions also can be stored in energy vectoring computer or the computer-readable memory of other programmable data processing device with ad hoc fashion work, the instruction that makes to be stored in this computer-readable memory produces the manufacture that comprises command device, and this command device is realized the function of appointment in flow process of flow chart or a plurality of flow process and/or square frame of block diagram or a plurality of square frame.

These computer program instructions also can be loaded in computer or other programmable data processing device, make to carry out sequence of operations step to produce computer implemented processing on computer or other programmable devices, thereby the instruction of carrying out is provided for realizing the step of the function of appointment in flow process of flow chart or a plurality of flow process and/or square frame of block diagram or a plurality of square frame on computer or other programmable devices.

Obviously, those skilled in the art can carry out various changes and modification and not depart from the spirit and scope of the present invention the present invention.Like this, if within of the present invention these are revised and modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.

Claims

1. a video image anti-fluttering method, is characterized in that, comprising:

2. the method for claim 1, it is characterized in that, corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as image shift adjustment amount, specifically comprise:

3. method as claimed in claim 1 or 2, is characterized in that, also comprises:

4. the method for claim 1, is characterized in that, sets up in the following way the corresponding relation of focusing position value and reference object distance:

5. the method for claim 1, is characterized in that, sets up in the following way the corresponding relation between reference object distance, rotation of lens angle and image shift amount:

6. a video image anti-fluttering method, is characterized in that, comprising:

7. method as claimed in claim 6, it is characterized in that, corresponding relation between reference object distance, rotation of lens angle and image shift amount based on setting up in advance, determine the image shift amount corresponding with described this reference object distance and this rotation of lens angle, as the first image shift adjustment amount, specifically comprise:

8. method as claimed in claim 7, it is characterized in that, according to a described x axle image shift adjustment amount, a described y axle image shift adjustment amount, described the 2nd x axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake migration processing of video image, specifically comprises:

9. the method as described in as arbitrary in claim 6-8, is characterized in that, also comprises:

10. method as claimed in claim 6, is characterized in that, sets up in the following way the corresponding relation of focusing position value and reference object distance:

11. methods as claimed in claim 6, is characterized in that, set up in the following way the corresponding relation between reference object distance, rotation of lens angle and image shift amount:

12. methods as claimed in claim 6, is characterized in that, set up in the following way the corresponding relation between reference object distance, camera lens side-play amount and image shift amount:

13. methods as described in as arbitrary in claim 6-9, is characterized in that, also comprise:

14. methods as claimed in claim 13, is characterized in that, set up in the following way reference object distance, along the z lens shaft side-play amount of z axle offset and the corresponding relation between image scaling ratio:

15. 1 kinds of video image anti-shake apparatus, is characterized in that, comprising:

16. devices as claimed in claim 15, it is characterized in that, described skew adjustment amount determining unit, specifically for the reference object distance based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus; And

17. devices as described in claim 15 or 16, is characterized in that, also comprise:

18. devices as claimed in claim 15, is characterized in that, also comprise:

19. devices as claimed in claim 15, is characterized in that, also comprise:

20. 1 kinds of video image anti-shake apparatus, is characterized in that, comprising:

21. devices as claimed in claim 20, it is characterized in that, described the first skew adjustment amount determining unit, specifically for the reference object distance based on setting up in advance, along the x lens shaft anglec of rotation of x axle rotation and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft anglec of rotation, as an x axle image shift adjustment amount, wherein, described this x lens shaft anglec of rotation is described pending video image x lens shaft anglec of rotation along the rotation of x axle when being taken by described capture apparatus,

Described the second skew adjustment amount determining unit, specifically for the reference object distance based on setting up in advance, along the x lens shaft side-play amount of x axle offset and the corresponding relation between x axle image shift amount, determine the x axle image shift amount corresponding with described this reference object distance and this x lens shaft side-play amount, as the 2nd x axle image shift adjustment amount, wherein, described this x lens shaft side-play amount is described pending video image x lens shaft side-play amount along x axle offset when being taken by described capture apparatus;

22. devices as claimed in claim 21, it is characterized in that, described anti-shake processing unit, specifically for according to a described x axle image shift adjustment amount and described the 2nd x axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd x axle image shift adjustment amount; And according to a described y axle image shift adjustment amount and described the 2nd y axle image shift adjustment amount, and each self-corresponding weight, be weighted summation, obtain the 3rd y axle image shift adjustment amount; And adopt described the 3rd x axle image shift adjustment amount and described the 3rd y axle image shift adjustment amount, described pending video image is carried out to the anti-shake processing of video image.

23. devices as described in as arbitrary in claim 20-22, is characterized in that, also comprise:

24. devices as claimed in claim 20, is characterized in that, also comprise:

25. devices as claimed in claim 20, is characterized in that, also comprise:

26. devices as claimed in claim 20, is characterized in that, also comprise:

27. devices as described in as arbitrary in claim 20-23, is characterized in that, also comprise:

28. devices as claimed in claim 27, is characterized in that, also comprise: