CN102841486A - Automatic focusing method for digital optical imaging system based on bilateral forecasting intersection - Google Patents

Abstract

The invention relates to an automatic focusing method for a digital optical imaging system, which comprises the following steps of: independently sampling and forecasting focusing evaluation function curves on left and right sides in the positions of focusing evaluation function curve peaks; calculating the position of an intersection point of the forecast left and right focusing evaluation function curves or a minimum value of an absolute value of a difference of the focusing evaluation function values in the same positions in the forecast left and right disperse focusing evaluation function value sequences; and controlling an executing mechanism to move a focal plane of the digital optical imaging system to the intersection point or minimum value position (focusing position). According to the automatic focusing method provided by the invention, independent sampling and forecasting are performed on left and right adjacent regions containing the peak position areas; a principle error caused by taking actually asymmetrical focusing evaluation function curves as symmetrical curves according to the automatic focusing method based on curve fitting and a swinging search process possibly occurring nearby the peaks according to a quick climbing method are avoided; the influence of sampling step length on the focusing precision according to the automatic focusing method based on bilateral forecasting intersection is small and the focusing precision and speed are higher.

Description

The digital optical imaging system automatic focusing method of friendship is asked in prediction based on bilateral

Technical field

The present invention relates to the automatic control technology field, relate in particular to digital optical imaging system automatic focusing method.

Background technology

In recent years, along with electronic technology and development of computer and progress, digital imagery device manufacturing technologies such as special CCD, CMOS ripe and using, the digital optical imaging system has all obtained application widely in the mankind's life and scientific research.In numerous technology of digital optical imaging system, automatic focusing method is to influence one of the image quality of digital optical imaging system and gordian technique of efficient.Technology of auto generally can be divided into initiatively automatic focusing method and passive automatic focusing method.Initiatively the automatic focusing method causes system complex, cost height owing to need extra light path or distance-measuring equipment.Passive automatic focusing method based on Flame Image Process need not extras, is beneficial to the integrated and microminiaturized of equipment, can reduce equipment cost and complex equipments degree significantly.Therefore, the automatic focusing method based on Flame Image Process has obtained using widely in the digital optical imaging system.

Technology of auto based on Flame Image Process mainly solves evaluation map two problems of automatic focusing method as focusing evaluation function with the search focal position of Pixel Information.For a long time, because the unpredictability of automatic focusing process, researcher's major side overweights the research of focusing evaluation function, and is then less to the research of automatic focusing method.Existing automatic focusing method can be divided into 3 big types:

The first kind is to use climbing method and the method for the search focusing evaluation function peak of curve position of improving one's methods.People such as Ooi (IEEE Transactions on Consumer Electronics, Vol.36, No.3,526-530,1990) have realized automatic focus with climbing method.People such as He (IEEE Transactions on Consumer Electronics; Vol.49; No.2,257-262,2003) in the focusing process according to current with on the difference of a focusing evaluation function value dynamically change the focusing step-length and proposed a kind of improved quick climbing method.Yoon and Park (International Journal of Advanced Manufacturing Technology; Vol.43; No.3; 287-293,2009) adopt minimax method of difference and two stage searching algorithm to realize climbing method, this method need not the influence that wave filter just can reduce impact noise.The topmost problem of climbing method is that near the search that focusing speed is slow, peak value is possibly be a swing process and the possibility that has out of focus.

Second type is to use binary search and Fibonacci search procedure (Fifth International Conference on Image Processing and its Applications; Pp.232-235,1995) method of search focusing evaluation function peak of curve position.Binary search and Fibonacci search procedure have the noise of being subject to, shortcomings such as error of backlash are repeatedly introduced in the long and mobile microscopical actuator to-and-fro movement of camera motion distance, have limited their application in reality.

The 3rd type is the method that the application curves match obtains focusing evaluation function peak of curve position.People such as Yazdanfar (Optics Express; Vol.16; No.12,8670-8677,2008) a kind of automatic focusing method of the focusing evaluation function value substitution empirical function of 2-3 width of cloth image being confirmed focal position proposed; This method combines focus evaluation function and curve fitting of Brenner, and only need be no more than 3 width of cloth images just can obtain focal position.Suppose asymmetrical actual focusing evaluation function curve mostly about the peak symmetry based on the automatic focusing method of curve fitting, this will introduce original reason error.

In sum, the accuracy and the efficient that how to improve the automatic focusing of digital optical imaging system still are the problems of extensive concern, and wherein the automatic focusing method of new principle is one of key of dealing with problems.

Summary of the invention

In view of this, the digital optical imaging system automatic focusing method of friendship is asked in the prediction based on bilateral that the purpose of this invention is to provide a kind of accuracy and efficient of the automatic focusing that improves the digital optical imaging system.

The objective of the invention is to realize through following technical scheme: the digital optical imaging system automatic focusing method of friendship is asked in prediction based on bilateral, comprises the steps:

1) the focusing evaluation function curve of focusing evaluation function peak of curve position and arranged on left and right sides is independently sampled respectively and predict;

2) the minimum value position of the absolute value of the difference of the focusing evaluation function value of same position in the left and right discrete focusing evaluation function value sequence of the intersection point of the left and right focusing evaluation function curve of calculating prediction or prediction;

3) focal plane of control executing mechanism mobile digital optical imaging system is to this intersection point or minimum value position (focal position).

Further, said step 1) specifically comprises the steps:

11) confirm to comprise the zone of focusing evaluation function peak of curve position, with its left and right side neighborhood as left and right sample area;

12) in the sample area at place, the focal plane of digital optical imaging system, carry out image sampling and evaluation, with obtaining the sample point sequence that focusing evaluation function value and corresponding sampling location are configured to predict;

13) predict the focusing evaluation function curve of this side or discrete focusing evaluation function value sequence through the sample point sequence of acquisition and the forecast model of this side;

14) focal plane of digital optical imaging system is moved to the sample area of focusing evaluation function peak of curve position opposite side; In this sample area, carry out image sampling and evaluation, with obtaining focusing evaluation function value and corresponding sampling location structure sample point sequence.

15) predict the focusing evaluation function curve of this side or discrete focusing evaluation function value sequence through the sample point sequence of acquisition and the forecast model of this side.

Further; If the predicted data that obtains in the step 1) is focusing evaluation function curve; Step 2 then) calculate the intersection point of the left and right focusing evaluation function curve of prediction in, the focal plane of control executing mechanism mobile digital optical imaging system is to this intersection point in the step 3).

Further; If the predicted data that obtains in the step 1) is discrete focusing evaluation function value sequence; Step 2 then) calculate the minimum value position of the absolute value of the difference of the focusing evaluation function value of same position in the left and right discrete focusing evaluation function value sequence of prediction in, the focal plane of control executing mechanism mobile digital optical imaging system is to this minimum value position in the step 3).

Compared with prior art, the of the present invention prediction based on bilateral asks the digital optical imaging system automatic focusing method of friendship to have following advantage:

1. the present invention has avoided supposing in the function-fitting method that through predicting left and right focusing evaluation function curve respectively at the evaluation function peak of curve position left and right side of focusing asymmetrical actual focusing evaluation function curve is the original reason error that symmetrical curve brought.

2. the present invention adopts the intersection point of the left and right focusing evaluation function curve that calculates prediction to confirm focal position, has avoided the swing search procedure that possibly occur at the peak annex in the quick climbing method.

3. the present invention adopts predicted value to replace actual focusing evaluation function value, can effectively reduce IMAQ and focusing evaluation function calculation times, has higher focusing efficient.

4. the present invention adopts forecast method to obtain the focal position of digital optical imaging system, and focusing accuracy receives the influence of sampling step length little, has high focusing accuracy and focusing efficient.

5. can select for use various focusing evaluation functions to come evaluation map among the present invention as Pixel Information; Can use various Forecasting Methodologies with a plurality of future position functions of few samples point prediction; Two interior Forecasting Methodology and sampling step lengths that use of sample area can be inconsistent; Make method of the present invention more flexible, the scope of application is extensive.

Other advantages of the present invention, target and characteristic will be set forth in instructions subsequently to a certain extent; And to a certain extent; Based on being conspicuous to those skilled in the art, perhaps can from practice of the present invention, obtain instruction to investigating of hereinafter.Target of the present invention and other advantages can realize and obtain through following instructions and claims.

Description of drawings

Fig. 1 is desirable focusing evaluation function curve map;

Fig. 2 is based on the principle schematic that the digital optical imaging system automatic focusing method of friendship is asked in the bilateral prediction; Wherein Fig. 2 (a) is a synoptic diagram of confirming left and right sample area; Fig. 2 (b) is the synoptic diagram in LSA sampling and prediction; Fig. 2 (c) is the synoptic diagram in RSA sampling and prediction; The synoptic diagram of the intersection point of focusing evaluation function about Fig. 2 (d) predicts when when predicting the outcome being focusing evaluation function curve, Fig. 2 (e) are the synoptic diagram of predicting when being the discrete point sequence of focusing evaluation function value and corresponding placement configurations when predicting the outcome that obtains the position of minimum value correspondence in the series

;

Fig. 3 is the schematic flow sheet of embodiment 1;

Fig. 4 is the program circuit synoptic diagram of the embodiment of the invention 1, and wherein Fig. 4 (a) is a main program flow chart, and Fig. 4 (b) is the process flow diagram that 7 climbing methods confirm to comprise the peak zone, and Fig. 4 (c) is the process flow diagram of new sample point more;

Fig. 5 is the schematic flow sheet of embodiment 2;

Fig. 6 is the schematic flow sheet of embodiment 3;

Fig. 7 is the schematic flow sheet of embodiment 4.

Embodiment

Below will carry out detailed description to the preferred embodiments of the present invention.Should be appreciated that preferred embodiment has been merely explanation the present invention, rather than in order to limit protection scope of the present invention.

Referring to Fig. 1, can know peak of curve position s from a kind of focusing evaluation function curve that characterizes the focusing process shown in Figure 1 _pWith peak of curve position neighborhood leftmost position and farthest right position s _lAnd s _rTwo zones forming, [s _ls _p] and [s _ps _r], can be used as the left and right neighborhood of peak of curve position.Therefore, the left and right focusing evaluation function curve f in the left and right neighborhood of peak _l(s) and f _r(s) can be expressed as

$\left\{\begin{array}{cc}{f}_l\left(s\right)=\mathrm{FV}\left(s\right)& s\∈\left[\begin{array}{cc}{s}_l& s_p\end{array}\right]\\ f_r\left(s\right)=\mathrm{FV}\left(s\right)& s\∈\left[\begin{array}{cc}{s}_p& s_r\end{array}\right]\end{array}\right.---\left(1\right)$

FV in the formula (s) is the focusing evaluation function value of the sampled images at the s place in the sampling location.According to Fig. 1 and Shi (1), the peak of focusing evaluation function curve can be regarded as left and right two focusing evaluation function curve f _l(s) and f _r(s) position of intersecting point.

Above-mentioned characteristic to focusing evaluation function curve; The bilateral that the present invention proposes is predicted and is asked and hand over automatic focusing method to confirm to comprise the zone of focusing evaluation function peak of curve position and the left and right neighborhood in this zone, and this left and right neighborhood is confirmed as left and right sample area (LSA and RSA); According to the focusing evaluation function value of the forecast model of the curve development trend of the both sides, evaluation function peak of curve position of representing to focus and the sampled images that in LSA and RSA, obtains respectively and the focusing evaluation function curve of two sample point sequence prediction peak left and right sides of corresponding sampling location structure; The intersection point and the control executing mechanism that obtain the left and right focusing evaluation function curve of prediction move the focal plane to this intersection point realization automatic focusing.

The technical scheme of this method is achieved in that

At first, shown in Fig. 2 (a), obtain to comprise the zone (s of peak according to existing method _Pls _Pr) and the left and right neighborhood [s in this zone _ls _Pl] and [s _Prs _r], the left and right neighborhood in this zone is used as LSA and RSA.

Secondly, shown in Fig. 2 (b) and Fig. 2 (c),, obtain two sample point series S through the sampled images that assessment obtains in LSA and RSA with certain sampling step length respectively ₁And S _r, can be expressed as respectively

S _l={(s _l(1),FV _l(1)),(s _l(2),FV _l(2)),…,(s _l(n),FV _l(n))}n∈N ^* (2)

S _r={(s _r(1),FV _r(1)),(s _r(2),FV _r(2)),…,(s _r(m),FV _r(m))}m∈N ^* (3)

N and m are respectively S in the formula _lAnd S _rSample point quantity, (s _l(i) FV _lAnd (s (i)) _r(i) FV _r(i)) represent S respectively ₁And S _rI sample point, to S _lAnd S _r, i≤n and m, wherein s _l(i) and FV _l(i) expression S _lThe sampling location and focusing evaluation function value, s of i sample point _r(i) and FV _r(i) expression S _rThe sampling location and focusing evaluation function value, N of i sample point ^*The expression positive integer.Consider that the forecast model form does

$\hat{f}\left(s\right)=f\left(s\right)---\left(4\right)$

F in the formula (s) and

represent forecast model respectively and predict the outcome.Predict the outcome

Be the focusing evaluation function value of prediction focusing evaluation function curve or prediction and the discrete point sequence of corresponding predicted position structure, according to formula (4) and the serial S of sampling _lAnd S _r, calculate the left and right focusing evaluation function curve of prediction With

Or the discrete point sequence of the left and right side of prediction

With

Wherein

With

Can be expressed as respectively

${\hat{f}}_{\mathrm{lp}}=\{({\hat{s}}_l\left(1\right),F{\hat{V}}_l\left(1\right)),({\hat{s}}_l\left(2\right),{F\hat{V}}_l\left(2\right)),...,({\hat{s}}_l\left(n\right),F{\hat{V}}_l\left(h\right))\}h\∈N^{*}---\left(4\right)$

${\hat{f}}_{\mathrm{rp}}=\{({\hat{s}}_r\left(1\right),F{\hat{V}}_r\left(1\right)),({\hat{s}}_r\left(2\right),{F\hat{V}}_r\left(2\right)),...,({\hat{s}}_r\left(m\right),F{\hat{V}}_r\left(h\right))\}h\∈N^{*}---\left(5\right)$

At last, be prediction focusing evaluation function curve if predict the outcome, shown in Fig. 2 (d), the intersection point of the left and right focusing evaluation function curve of prediction satisfies so

${\hat{f}}_l\left({\hat{s}}_p\right)-{\hat{f}}_r\left({\hat{s}}_p\right)=0---\left(6\right)$

Intersection point

is not only the peak of focusing evaluation function curve, and is the focal position of digital optical imaging equipment.

If predict the outcome is the discrete point sequence, can be expressed as according to the new sequence of formula (4) and (5) structure

${\hat{X}}_p=|F{\hat{V}}_l\left(i\right)-F{\hat{V}}_r\left(i\right)|i\≤h---\left(7\right)$

Shown in Fig. 2 (e), the position of minimum value correspondence is as focal position in the sequence

.

Mobile digital focal plane of the optical imaging system to

location or sequence

in the position corresponding to the minimum to achieve automatic focusing.

Among the present invention, the forecast model of left and right sampling step length, sample point number that left and right sample point sequence comprises and bilateral all can be inconsistent.Forecast model can usage trend outside forecast method and time series forecasting method, for example grey forecasting model, exponential forecasting model etc.

Embodiment 1

With reference to Fig. 3; The process flow diagram of the digital optical imaging system automatic focusing method method of friendship is asked in the prediction based on bilateral of present embodiment, comprises that position of intersecting point (8) is arrived in zone (1), left and right sample area (2), acquisition left samples point sequence (3) right side sample point sequence (4), left side model parameter (5) right side model parameter (6), position of intersecting point (7), the mobile focal plane of definite peak.Concrete realization for the explanation present embodiment; With the Variance function as the focusing evaluation function; 7 climbing methods are as confirming to comprise the zone of focusing evaluation function peak of curve position and the method for left and right sample area; The exponential forecasting model is as the bilateral forecast model, and least square method is as the method for the model parameter of calculating the bilateral forecast model.The concrete realization of present embodiment is following:

In the step of confirming left and right sample area; Obtain 7 Variance focusing evaluation function values and corresponding sampling location with certain sampling step length and 7 climbing methods successively according to focusing direction (pointing to the direction of focusing evaluation function peak of curve position from the initial sampled position), obtaining sample sequence S can be expressed as:

S={(s(1),FV(1)),(s(2),FV(2)),…,(s(i),FV(i)),…,(s(7),FV(7))} i＝1,2,…,7 (8)

Judge whether FV (4) among the sequence S is whether maximal value and two sequences (FV (3), FV (2), FV (1)) and (FV (5), FV (6), the FV (7)) of all focusing evaluation function values in the sequence is monotonically decreasing sequence.Do not satisfy if any a condition, increase a focusing evaluation function value and corresponding sampling location, upgrade sample sequence S by the focusing direction; FV in satisfying sequence S (4) is maximal value and two sequences (FV (3), FV (2), FV (1)) and (FV (5); FV (6); FV (7)) be monotonically decreasing sequence, the zone of confirming to comprise focusing evaluation function peak of curve position is (s (3) s (5)), realizes (1) in the present embodiment.

The left and right neighborhood that comprises the zone of focusing evaluation function peak of curve position is respectively [s (1) s (3)] and [s (5) s (7)], and this left and right neighborhood is used as left and right sample area, realizes (2) in the present embodiment.

Sequence ((s (1) FV (1)), (s (2) FV (2)), (s (3) FV (3))) and ((s (5) FV (5)), (s (6) FV (6)), (s (7) FV (7))) are used as left side and right side sample point sequence respectively, realize (3) and (4) in the present embodiment.

The exponential forecasting model can be expressed as

f(s)=ae ^bs (9)

A and b are calculative model parameters in the formula, and s is the sampling location, and f (s) is the focusing evaluation function value that the s place obtains in the sampling location.For utilizing least square method to confirm model parameter a and b, formula (9) can be expressed as

$\ln \hat{f}\left(s\right)=\ln a+\mathrm{bs}---\left(10\right)$

With sample point sequence substitution formula (10), can get

lnFV(i)=lna+bs(i) (11)

The equality of finding the solution exponential forecasting model parameter lna and b with least square method can be expressed as

$\left[\begin{array}{c}\ln a\\ b\end{array}\right]\text{=}{\left(B^{T}B\right)}^{-1}{B}^T{Y}_N---\left(12\right)$

In the formula $B=\left[\begin{array}{cc}.& .\\ .& .\\ .& .\\ 1& s\left(i\right)\\ .& .\\ .& .\\ .& .\end{array}\right],$ $Y_N=\left[\begin{array}{c}.\\ .\\ .\\ \mathrm{Ln}\left(\mathrm{FV}\left(i\right)\right)\\ .\\ .\\ .\end{array}\right].$ With left side and right side sample point sequence substitution formula (10) and according to formula (12), obtain left side exponential forecasting model parameter (lna1 bol) and right side exponential forecasting model parameter (lna through respectively _rb _r), realize (5) and (6) in the present embodiment.

The left and right focusing evaluation function group of prediction can be expressed as

$\left\{\begin{array}{c}\ln {\hat{f}}_l\left(s\right)=\ln a_l+b_{l}s\\ \ln {\hat{f}}_r\left(s\right)=\ln a_r+b_{r}s\end{array}\right.---\left(13\right)$

Therefore, the intersection point of the left and right focusing evaluation function curve of prediction can be expressed as

${\hat{s}}_p=\frac{\ln a_l-\ln a_r}{b_r-b_l}---\left(14\right)$

Calculate the intersection point of the left and right sides exponential forecasting model of prediction through following formula and realize (7) in the present embodiment.

Accomplish automatic focus through the focal plane of mobile digital optical imaging system to the position of intersecting point of the left and right sides exponential forecasting model of prediction and realize (8) in the present embodiment.

Fig. 4 is the software flow pattern of present embodiment.

Specific embodiment 2:

With reference to Fig. 5, the prediction of the bilateral of present embodiment asks the realization module map of handing over automatic focusing method to comprise that minimum value position (12) is arrived in corresponding position (11) and the mobile focal plane of minimum value in the zone (1) that comprises peak, left and right sample area (2), left samples point sequence (3), right side sample point sequence (4), left side discrete point sequence (9), right side discrete point sequence (10), the sequence

.The difference of present embodiment and embodiment 1 is that (9) and (10) in the present embodiment all are discrete point sequence but not model parameter (5) and (6) of calculating forecast model among the embodiment 1 of calculating prediction, and the difference of present embodiment and embodiment 1 also is to need the middle minimum value position (11) of sequence

of acquisition in the present embodiment but not the position of intersecting point (7) of the left and right focusing evaluation function curve of the prediction that predicts the outcome about among the embodiment 1.The difference of present embodiment and embodiment 1 is that also present embodiment moves the focal plane to minimum value position (12) but not position of intersecting point (8) among the embodiment 1.

Specific embodiment 3:

With reference to Fig. 6, the prediction of the bilateral of present embodiment asks the realization module map of handing over automatic focusing method to comprise that the zone (1) that comprises peak, left and right sample area (2), left samples point sequence (3), left side model parameter (5), right side sample point sequence (4), right side model parameter (6), position of intersecting point (7) and mobile focal plane are to position of intersecting point (8).The difference of present embodiment and embodiment 1 is that present embodiment is after definite left and right sample area (2).Can not directly obtain left and right side sample point sequence from the data that the zone of confirming to comprise peak (1) and definite left and right sample area (2) process, obtain, but need go sampling could obtain left and right side sample points certificate in left and right sample area respectively.

Specific embodiment 4:

With reference to Fig. 7; The bilateral prediction of present embodiment is asked and is handed over the realization module map of automatic focusing method to comprise the zone (1) that comprises peak, left and right sample area (2), left samples point sequence (3), left side discrete point sequence (9); Right side sample point sequence (4); Minimum value position (12) is arrived in corresponding position (11) and the mobile focal plane of minimum value in the right side discrete point sequence (10), sequence

.The difference of present embodiment and embodiment 1 is in the present embodiment after definite left and right sample area (2).Can not directly obtain left and right side sample point sequence from the data that the zone of confirming to comprise peak (1) and definite left and right sample area (2) process, obtain, but need to obtain left and right side sample points certificate in left and right sample area sampling respectively.The difference of present embodiment and embodiment 1 is that (9) and (10) in the present embodiment all are discrete point sequence but not model parameter (5) and (6) of calculating forecast model among the embodiment 1 of calculating prediction, and the difference of present embodiment and embodiment 1 also is to need the middle minimum value position (11) of sequence

of acquisition in the present embodiment but not the position of intersecting point (7) of the left and right focusing evaluation function curve of the prediction that predicts the outcome about among the embodiment 1.The difference of present embodiment and embodiment 1 is that also present embodiment moves the focal plane to minimum value position (12) but not position of intersecting point (8) among the embodiment 1.

Explanation is at last; Above embodiment is only unrestricted in order to technical scheme of the present invention to be described; Although with reference to preferred embodiment the present invention is specified, those of ordinary skill in the art should be appreciated that and can make amendment or be equal to replacement technical scheme of the present invention; And not breaking away from the aim and the scope of present technique scheme, it all should be encompassed in the middle of the claim scope of the present invention.

Claims (4)

1. predict the digital optical imaging system automatic focusing method of asking friendship based on bilateral, it is characterized in that: comprise the steps:

1) the focusing evaluation function curve of focusing evaluation function peak of curve position and arranged on left and right sides is independently sampled respectively and predict;

2) the minimum value position of the absolute value of the difference of the focusing evaluation function value of same position in the left and right discrete focusing evaluation function value sequence of the intersection point of the left and right focusing evaluation function curve of calculating prediction or prediction;

3) focal plane of control executing mechanism mobile digital optical imaging system is to this intersection point or minimum value position (focal position).

2. according to claim 1ly ask the digital optical imaging system automatic focusing method of friendship based on bilateral prediction, it is characterized in that: said step 1) specifically comprises the steps:

11) confirm to comprise the zone of focusing evaluation function peak of curve position, with its left and right side neighborhood as left and right sample area;

12) in the sample area at place, the focal plane of digital optical imaging system, carry out image sampling and evaluation, with obtaining the sample point sequence that focusing evaluation function value and corresponding sampling location are configured to predict;

13) predict the focusing evaluation function curve of this side or discrete focusing evaluation function value sequence through the sample point sequence of acquisition and the forecast model of this side;

14) focal plane of digital optical imaging system is moved to the sample area of focusing evaluation function peak of curve position opposite side; In this sample area, carry out image sampling and evaluation, with obtaining focusing evaluation function value and corresponding sampling location structure sample point sequence;

15) predict the focusing evaluation function curve of this side or discrete focusing evaluation function value sequence through the sample point sequence of acquisition and the forecast model of this side.

3. the digital optical imaging system automatic focusing method of asking friendship of predicting based on bilateral according to claim 1 and 2; It is characterized in that: if the predicted data that obtains in the step 1) is focusing evaluation function curve; Step 2 then) calculate the intersection point of the left and right focusing evaluation function curve of prediction in, the focal plane of control executing mechanism mobile digital optical imaging system is to this intersection point in the step 3).

4. the digital optical imaging system automatic focusing method of asking friendship of predicting based on bilateral according to claim 1 and 2; It is characterized in that: if the predicted data that obtains in the step 1) is discrete focusing evaluation function value sequence; Step 2 then) calculate the minimum value position of the absolute value of the difference of the focusing evaluation function value of same position in the left and right discrete focusing evaluation function value sequence of prediction in, the focal plane of control executing mechanism mobile digital optical imaging system is to the minimum value position in the step 3).

Images