CN106154689A - Atomatic focusing method and use the image capture unit of this Atomatic focusing method - Google Patents

Abstract

The present invention provides a kind of Atomatic focusing method and uses the image capture unit of this Atomatic focusing method.This image capture unit includes camera lens and photosensory assembly.This Atomatic focusing method includes first changing this camera lens position relative with this photosensory assembly with the first amount of space and obtaining a plurality of image at a plurality of positions pick-up image respectively, then the definition values of these images is calculated, again these definition values are obtained a plurality of operation values after suitable mathematical functional operation and make these operation values relation curve with position close to straight line, re-use linear interpolation to calculate and extrapolate the camera lens corresponding to desired value of definition exactly with the target of photosensory assembly relative to position, finally with the second amount of space adjust the camera lens relative position with photosensory assembly to this target relative to position to complete to focus.

Description

Atomatic focusing method and use the image capture unit of this Atomatic focusing method

Technical field

The present invention is related to a kind of Atomatic focusing method and uses the image capture unit of this Atomatic focusing method, In particular to a kind of the definition values of image is carried out computing with mathematical function after to estimate definition Atomatic focusing method that desired value thereby carrying out is focused and the image capture using this Atomatic focusing method fill Put.

Background technology

In known focusing method, having so-called two benches focusing method, the first stage is first driven by stepper motor Move and focus lens is moved and respectively obtains image definition angle value with bigger amount of space in multiple positions, then look for Go out the maximum in these image definition values, in second stage, by stepping horse near above-mentioned maximum Reach driving focus lens to be moved and in multiple positions, finds out image definition angle value with less amount of space Good value.

But owing to above-mentioned two stage focusing method is relatively time consuming, therefore another kind of known Atomatic focusing method Then disclose first to drive with bigger amount of space and focus lens is moved and respectively obtains corresponding to respectively in multiple positions The image definition angle value of individual position, then estimate out in the way of these image definition angle value are with mathematical operation with Another less amount of space moves the corresponding position to the optimum that can obtain image definition angle value during focus lens Put, as it is shown in figure 1, change the position P to focus lens in camera lens with larger space amount₀' to P_N', and Obtain the most several image definition value, then take preferably three position P of image definition angle value₁’、P₂' and P₃', and by these three position P₁’、P₂' and P₃' corresponding image definition angle value M₁’、M₂' and M₃’ Application linear interpolation calculates when moving focus lens with relatively closely spaced amount on auto-focusing curve C Best placement P₂₁', at this optimum position P₂₁' available optimal image definition angle value M_f’.I.e. such as Fig. 1 Shown in, P₁₁' it is P₁' and P₂' midpoint, P₂₂' it is P₂' and P₃' midpoint, it is assumed that P₁₁' to P₂₁’ Distance and P₂₁' to P₂₂' the ratio of distance be Δ x'/(S'-Δ x'), with linear interpolation, it meets $\frac{\mathrm{\Δ}{x}^{\′}}{S^{\′}-\mathrm{\Δ}{x}^{\′}}=\frac{{M_2}^{\′}-{M_3}^{\′}}{{M_2}^{\′}-{M_1}^{\′}},$ Therefore $\mathrm{\Δ}{x}^{\′}=\frac{M_2^{\′}-{M_3}^{\′}}{({M_2}^{\′}-{M_1}^{\′})+({M_2}^{\′}-{M_3}^{\′})}\×S^{\′},$ Known ${P_{21}}^{\′}={P_2}^{\′}-\frac{1}{2}{S}^{\′}+\mathrm{\Δ}{x}^{\′}$ And S'=(P₂'-P₁'), therefore can obtain: ${P_{21}}^{\′}={P_2}^{\′}-\frac{1}{2}({P_2}^{\′}-{P_1}^{\′})+\frac{{M_2}^{\′}-{M_3}^{\′}}{({M_2}^{\′}-{M_1}^{\′})+({M_2}^{\′}-{M_3}^{\′})}\×({P_2}^{\′}-{P_1}^{\′}).$

But in the method shown in Fig. 1, due to auto-focusing curve C non-rectilinear, and use linear interior The optimum of the image definition angle value that the method for inserting is tried to achieve on auto-focusing curve C and actual image definition The optimum of value has error, thus cannot correctly find the position that can produce sharp image.

Summary of the invention

The technical problem to be solved in the present invention is, cannot be accurate for Atomatic focusing method of the prior art Find the defect that can produce the position of sharp image, it is provided that a kind of Atomatic focusing method, can accurately find The position of sharp image can be produced.

Another object of the present invention is to provide a kind of image capture unit, the auto-focusing of the applicable present invention Method.

The present invention solves that its technical problem be employed technical scheme comprise that, it is provided that a kind of Atomatic focusing method One embodiment is used for image capture unit, and this image capture unit includes camera lens and photosensory assembly, this camera lens Can be adjusted according to parameter with the relative position of this photosensory assembly and complete focusing, this parameter can be according to first Amount of space and the second amount of space are changed, and this Atomatic focusing method includes: change according to this first amount of space Become this parameter and obtain a plurality of numerical value to adjust this camera lens position relative with this photosensory assembly, these numerical value Corresponding to via this camera lens plurality of states of imaging on this photosensory assembly；These numerical value are carried out mathematics fortune Calculate and respectively obtain a plurality of operation values；Change according to this second amount of space to estimate out based on these operation values This parameter and obtain this camera lens target with this photosensory assembly relative to position, this target relative to position corresponding to warp By this camera lens dbjective state of imaging on this photosensory assembly；And adjust this camera lens or this photosensory assembly to being somebody's turn to do Target relative to position to obtain this dbjective state.

In the above-described embodiments, these states include the first state, the second state and the third state, via The imaging on this photosensory assembly of this camera lens is more clear, and this first state, the second state and the third state are divided Dui Yingyu the first numerical value, second value and third value in these numerical value.

In the above-described embodiments, this camera lens includes focus lens, and this parameter is that this to the position of focus lens or is somebody's turn to do The position of photosensory assembly.

In the above-described embodiments, this first state, the second state and the third state correspond respectively to this focusing Eyeglass or the primary importance of this photosensory assembly, the second position and the 3rd position, this primary importance and the 3rd It is positioned adjacent to this second position.

In the above-described embodiments, this second state is to become on this photosensory assembly via this camera lens in these states The most clear state of picture.

Above-described embodiment further includes and so that mathematical function carries out computing, these numerical value is respectively obtained these computings Value.

In the above-described embodiments, this mathematical function includes logarithmic function, this target relative to position with following equations Formula speculates:

$P_{21}=P_2-\frac{1}{2}(P_2-P_1)+\frac{\log M_2-\log M_3}{(\log M_2-\log M_1)+(\log M_2-\log M_3)}\×(P_2-P_1)$

Wherein at P₁Represent this primary importance, P₂Represent this second position, P₃Represent the 3rd position, M₁Representing should First numerical value, M₂Represent this second value, M₃Represent this third value, P₂₁Represent that this target is relative to position.

Above-described embodiment further includes: adjust this parameter to plurality of parameters value with adjust this camera lens with this photosensitive group The relative position of part, these parameter values from minima with this first amount of space progression to maximum；And whenever When this parameter adjustment is to the parameter value therein of these parameter values, capture photosensitive in this via this lens imaging The image of assembly, and obtain a plurality of image；And calculated these numerical value respectively by these images.

In the above-described embodiments, these numerical value are the definition values of these images.

In the above-described embodiments, this second amount of space is little compared with this first amount of space.

One embodiment of the image capture unit of the present invention includes: photosensory assembly, is converted into by the light of reception Electric signal；Camera lens, light via this lens imaging in this photosensory assembly；Processing unit, for according to the One amount of space changes parameter and obtains a plurality of numerical value to adjust this camera lens position relative with this photosensory assembly, These numerical value, corresponding to via this camera lens plurality of states of imaging on this photosensory assembly, utilizes these numerical value Carry out computing and obtain a plurality of operation values, change according to the second amount of space to estimate out based on these operation values This parameter and obtain this camera lens target with this photosensory assembly relative to position, this target relative to position corresponding to warp By this camera lens dbjective state of imaging on this photosensory assembly, and according to this second amount of space change this parameter with Adjust this camera lens or this photosensory assembly to this target relative to position to obtain this dbjective state.

In the above-described embodiments, these states include the first state, the second state and the third state, via The imaging on this photosensory assembly of this camera lens is more clear, and this first state, the second state and the third state are divided Dui Yingyu the first numerical value, second value and third value in these numerical value.

In the above-described embodiments, this camera lens includes focus lens, and this parameter is that this to the position of focus lens or is somebody's turn to do The position of photosensory assembly.

In the above-described embodiments, this first state, the second state and the third state correspond respectively to this focusing Eyeglass or the primary importance of this photosensory assembly, the second position and the 3rd position, this primary importance and the 3rd It is positioned adjacent to this second position.

In the above-described embodiments, this second state is to become on this photosensory assembly via this camera lens in these states The most clear state of picture.

Above-described embodiment further includes this processing unit and is obtained respectively so that mathematical function carries out computing by these numerical value To these operation values.

In the above-described embodiments, this mathematical function includes logarithmic function, this target relative to position by this process list Unit speculates with following equation:

$P_{21}=P_2-\frac{1}{2}(P_2-P_1)+\frac{\log M_2-\log M_3}{(\log M_2-\log M_1)+(\log M_2-\log M_3)}\×(P_2-P_1)$

Wherein at P₁Represent this primary importance, P₂Represent this second position, P₃Represent the 3rd position, M₁Representing should First numerical value, M₂Represent this second value, M₃Represent this third value, P₂₁Represent that this target is relative to position.

In the above-described embodiments, this processing unit adjust this parameter to plurality of parameters value with adjust this camera lens with The relative position of this photosensory assembly, these parameter values from minima with this first amount of space progression to maximum； Whenever the parameter value therein of this parameter of adjustment to these parameter values, this processing unit captures via this camera lens Image in the image of this photosensory assembly, and obtain a plurality of image；This processing unit is counted respectively by these images Calculate these numerical value.

In the above-described embodiments, these numerical value are the definition values of these images.

In the above-described embodiments, this second amount of space is little compared with this first amount of space.

Implement the Atomatic focusing method of the present invention and use the image capture unit of this Atomatic focusing method, having Following beneficial effect: the numerical value of image definition is obtained operation values after suitable mathematical functional operation, And the relation curve of this operation values and position is close to straight line, after so re-using linear interpolation computing, with regard to energy Extrapolate the target phase para-position of the camera lens corresponding to the optimal values of image definition and photosensory assembly exactly Put.

Accompanying drawing explanation

In order to above and other objects of the present invention, feature and advantage can be become apparent, cited below particularly go out Embodiment also coordinates accompanying drawing to elaborate.

Fig. 1 represents the auto-focusing curve that known Atomatic focusing method is used.

Fig. 2 A is the square of an embodiment of the image capture unit of the Atomatic focusing method being applicable to the present invention Figure.

Fig. 2 B is the side of another embodiment of the image capture unit of the Atomatic focusing method being applicable to the present invention Block figure.

Fig. 3 is the flow chart of the Atomatic focusing method of the present invention.

The auto-focusing curve that the Atomatic focusing method that Fig. 4 is the present invention is used.

Detailed description of the invention

Referring to Fig. 2 A, its expression is suitable for a reality of the image capture unit of the Atomatic focusing method of the present invention Execute example.This image capture unit includes camera lens 10, photosensory assembly 20, processing unit 30, control unit 40 And driver element 50, camera lens 10 has focus lens 12.Light images in photosensitive group via camera lens 10 On part 20, and by focus lens 12 adjusts focusing, the light of reception is converted into electronics by photosensory assembly 20 Signal, processing unit 30 images in the image on photosensory assembly 20 for capturing and this image is carried out computing After process, the result of calculation process passing to control unit 40, control unit 40 controls driver element 50 To change the position to focus lens 12, thereby adjust the relative position to focus lens 12 with photosensory assembly 20 And make light clearly image on photosensory assembly 20.In the present embodiment, photosensory assembly 20 is image sense Survey assembly, such as CCD or CMOS.

Referring to Fig. 2 B, its expression is suitable for another of the image capture unit of the Atomatic focusing method of the present invention Embodiment.This image capture unit includes camera lens 10, photosensory assembly 20, processing unit 30, control unit 40 and driver element 50, camera lens 10 has focus lens 12.Light images in photosensitive via camera lens 10 On assembly 20, the light of reception is converted into electric signal by photosensory assembly 20, and processing unit 30 is used for picking After taking the image imaged on photosensory assembly 20 and this image being carried out calculation process, by the knot of calculation process Fruit passes to control unit 40, control unit 40 according to the output control driver element 50 of this computing to change Become the position of photosensory assembly 20, thereby adjust position relative with photosensory assembly 20 to focus lens 12 and make Light clearly images on photosensory assembly 20.In the present embodiment, photosensory assembly 20 is image sensing group Part, such as CCD or CMOS.

The Atomatic focusing method of the present invention can realize by the image capture unit of Fig. 2 A or Fig. 2 B, but does not limits In Fig. 2 A or the image capture unit of Fig. 2 B.The Atomatic focusing method of the present invention is adjusted by changing a parameter Whole above-mentioned completing focus lens 12 with the relative position of photosensory assembly 20 is focused, and above-mentioned parameter can be right The position of focus lens 12 or the position of photosensory assembly 20, above-mentioned parameter can according to the first amount of space with And second amount of space be changed, can be by the position of the position of focus lens 12 or photosensory assembly 20 Driver element 50 is adjusted, and driver element 50 can be stepper motor, the most aforesaid second amount of space Can be the minimum step of stepper motor driving, the first amount of space be then the multiple of minimum step.

Refer to Fig. 3, the flow process of the Atomatic focusing method of its expression present invention.In step sl, according to It is relative with photosensory assembly 20 to adjust camera lens 10 to plurality of parameters value that first amount of space changes above-mentioned parameter Position and obtain a plurality of numerical value, these parameter values from minima with this first amount of space progression to maximum, These numerical value is corresponding to via camera lens 10 plurality of states of imaging on photosensory assembly 20.Driver element 50 move focus lens 12 or photosensory assembly 20 with the first amount of space with the first amount of space, i.e. stepper motor To adjust the relative position of camera lens 10 and photosensory assembly 20, processing unit 30 is obtaining with this first amount of space The a plurality of positions arrived pick-up image respectively, as shown in Figure 4, at position P₀To P_NBy processing unit 30 points Other pick-up image, P₀For minimum position, P_NFor maximum position, then picked for each by processing unit 30 The image taken calculates its definition values (numerical value), and the definition values calculated can correspond to light via camera lens 10 on photosensory assembly 20 state of imaging.When processing unit 30 calculates the definition values of each image After, enter step S2.

In step s 2, above-mentioned numerical value is carried out computing and respectively obtain a plurality of operation values.Processing unit The a plurality of definition values calculated are performed mathematical calculations and obtain a plurality of operation values, in this reality by 30 Executing in example, arithmetic element 30 takes definition values preferably primary importance P₁, second position P₂And the 3rd position P₃The image captured, calculates the first definition values (the first numerical value) M respectively₁, the second definition values ( Two numerical value) M₂And the 3rd definition values (third value) M₃, the first definition values M₁, the second definition values M₂And the 3rd definition values M₃Represent that light images in the first shape of illuminant module 20 through camera lens 10 respectively State, the second state and the third state, wherein second position P₂The second definition values M₂For all positions P₀To P_NDefinition values in the highest, primary importance P₁With the 3rd position P₃Adjacent to second position P₂, Then would correspond to primary importance P₁, second position P₂And the 3rd position P₃First, second and third definition Value M₁、M₂And M₃Carry out computing with mathematical function, in order to improve linear interpolation in known technology be applied to right Burnt curve has the problem of error, finds after calculating checking with actual numerical value, as shown in Figure 4, passes through It is closer to straight line, hence with logarithm with the relation curve D of the operation values after logarithmic function computing Yu position Operation values after functional operation can obtain fairly small with the relation curve D application linear interpolation of position Estimate error.Subsequently enter step S3.

In step s3, above-mentioned parameter is changed to estimate out according to this second amount of space based on these operation values And obtain the camera lens 10 target with photosensory assembly 20 relative to position, this target relative to position corresponding to via mirror 10 on photosensory assembly 20 dbjective state of imaging, the second amount of space is little compared with the first amount of space, in this reality Executing in example, this dbjective state is via camera lens 10 the most clear state of imaging on photosensory assembly 20.Due to In Fig. 4, operation values after logarithmic function computing is closer to straight line with the relation curve D of position, therefore applies Relation curve D calculate the target of imaging relative to position, as shown in Figure 4, P21 is above-mentioned target phase para-position Putting, target is Mf relative to the largest image definition values corresponding to the P21 of position, largest image definition values For the logarithmic function operation values of Mf be logMf, P11 be the midpoint of primary importance P1 and second position P2, P22 is second position P2 and the midpoint of the 3rd position P3.P11 Yu P22, at a distance of S, makes P11 and P21 At a distance of Δ x, therefore P21 Yu P22 S-Δ x apart, according to the linear relationship of Fig. 4, P11 to P21 away from Difference and logM2 and logM1 of logM2 Yu logM3 it is equal to from the ratio with the distance of P21 to P22 The ratio of difference, i.e.

$\frac{\mathrm{\Δx}}{S-\mathrm{\Δx}}=\frac{\log M_2-\log M_3}{\log M_2-{\log M}_1}$ Therefore

$\mathrm{\Δx}=\frac{\log M_2-\log M_3}{(\log M_2-{\log M}_1)+(\log M_2-\log M_3)}\×S---\left(1\right)$

The most knownAnd S=(P₂-P₁), can obtain:

$P_{21}=P_2-\frac{1}{2}(P_2-P_1)+\mathrm{\Δx}---\left(2\right)$

By the available target in equation (1) and (2) relative to position P₂₁For:

$P_{21}=P_2-\frac{1}{2}(P_2-P_1)+\frac{\log M_2-\log M_3}{(\log M_2-\log M_1)+(\log M_2-{\log M}_3)}\×(P_2-P_1)---\left(3\right)$

Processing unit 30 calculates target relative to position P according to equation (3)₂₁After, subsequently enter step S4.

In step s 4, camera lens 10 or photosensory assembly 20 to target relative position P are adjusted₂₁To obtain this mesh Mark state.Processing unit 30 calculates target relative to position P according to equation (3)₂₁After, target is relative Position P₂₁Data be transferred to control unit 40, control unit 40 according to target relative to position P₂₁Data Control driver element 50 and drive and focus lens 12 or photosensory assembly 20 are moved to target relative to position P₂₁, The image optimal to obtain definition, i.e. completes focusing.

The Atomatic focusing method of the present invention utilizes mathematical function to be calculated as the definition values of picture, and it is straight to obtain approximation The relation curve of line, thereby can utilize linear interpolation to estimate imaging position the most clearly exactly, and complete Become auto-focusing.

Though the present invention is disclosed above with embodiment, but it is not used to limit the scope of the present invention, the skill of this area Art personnel, without departing from the scope of the present invention, when a little change and retouching can be done, therefore this Bright protection domain is when being as the criterion depending on as defined in claim.

Claims (12)

1. an Atomatic focusing method, for image capture unit, it is characterised in that this image capture fills Putting and include camera lens and photosensory assembly, this camera lens can be adjusted according to parameter with the relative position of this photosensory assembly Whole and complete focusing, this parameter can be changed according to the first amount of space and the second amount of space, and this is the most right Burnt method includes:

According to this first amount of space change this parameter adjusting this camera lens position relative with this photosensory assembly and Obtaining a plurality of numerical value, these numerical value is corresponding to via this camera lens a plurality of shapes of imaging on this photosensory assembly State；

These numerical value are utilized to carry out computing and obtain a plurality of operation values；

Based on these operation values with estimate out change this parameter according to this second amount of space and obtain this camera lens with The target of this photosensory assembly is relative to position, and this target corresponds to via this camera lens at this photosensory assembly relative to position The dbjective state of upper imaging；And

This parameter is changed relative to this target to adjust this camera lens or this photosensory assembly according to this second amount of space Position is to obtain this dbjective state.

2. Atomatic focusing method as claimed in claim 1, it is characterised in that these states include first State, the second state and the third state, more clear via the imaging on this photosensory assembly of this camera lens, and This second state is the most clear state in these states, and this camera lens includes focus lens, and this parameter is that this is right The position of focus lens, this first state, the second state and a third state correspond respectively to this to focus lens Primary importance, the second position and the 3rd position, this primary importance with the 3rd be positioned adjacent to this second Position, this first state, the second state and the third state correspond respectively to the first numerical value in these numerical value, Second value and third value.

3. Atomatic focusing method as claimed in claim 1, it is characterised in that these states include first State, the second state and the third state, more clear via the imaging on this photosensory assembly of this camera lens, and This second state is the most clear state in these states, and this parameter is the position of this photosensory assembly, and this is first years old State, the second state and the third state correspond respectively to this to the primary importance of focus lens, the second position with And the 3rd position, this primary importance and the 3rd is positioned adjacent to this second position, this first state, second State and the third state correspond respectively to the first numerical value in these numerical value, second value and the 3rd number Value.

4. Atomatic focusing method as claimed in claim 2 or claim 3, it is characterised in that further include these Numerical value respectively obtains these operation values so that mathematical function performs mathematical calculations.

5. Atomatic focusing method as claimed in claim 4, it is characterised in that it is right that this mathematical function includes Number function, this target speculates with following equation relative to position:

$P_{21}=P_2-\frac{1}{2}(P_2-P_1)+\frac{\log M_2-\log M_3}{(\log M_2-\log M_1)+(\log M_2-\log M_3)}\×(P_2-P_1)$

Wherein at P₁Represent this primary importance, P₂Represent this second position, P₃Represent the 3rd position, M₁Table Show this first numerical value, M₂Represent this second value, M₃Represent this third value, P₂₁Represent this target phase para-position Put.

6. Atomatic focusing method as claimed in claim 1, it is characterised in that this second amount of space is less than This first amount of space.

7. an image capture unit, it is characterised in that including:

Photosensory assembly, is converted into electric signal by the light of reception；

Camera lens, light via this lens imaging in this photosensory assembly；And

Processing unit, for changing parameter to adjust the phase of this camera lens and this photosensory assembly according to the first amount of space Position obtains a plurality of numerical value, and these numerical value is corresponding to via the imaging on this photosensory assembly of this camera lens Plurality of states, utilizes these numerical value to carry out computing and obtain a plurality of operation values, based on these operation values with Estimate out and change this parameter according to the second amount of space and obtain the target phase para-position of this camera lens and this photosensory assembly Putting, this target corresponds to via this camera lens dbjective state of imaging on this photosensory assembly relative to position, and root According to this second amount of space change this parameter with adjust this camera lens or this photosensory assembly to this target relative to position with Obtain this dbjective state.

8. image capture unit as claimed in claim 7, it is characterised in that these states include first State, the second state and the third state, more clear via the imaging on this photosensory assembly of this camera lens, and This second state is the most clear state in these states, and this camera lens includes focus lens, and this parameter is that this is right The position of focus lens, this first state, the second state and a third state correspond respectively to this to focus lens Primary importance, the second position and the 3rd position, this primary importance with the 3rd be positioned adjacent to this second Position, this first state, the second state and the third state correspond respectively to the first numerical value in these numerical value, Second value and third value.

9. image capture unit as claimed in claim 7, it is characterised in that these states include first State, the second state and the third state, more clear via the imaging on this photosensory assembly of this camera lens, and This second state is the most clear state in these states, and this parameter is the position of this photosensory assembly, and this is first years old State, the second state and the third state correspond respectively to this to the primary importance of focus lens, the second position with And the 3rd position, this primary importance and the 3rd is positioned adjacent to this second position, this first state, second State and the third state correspond respectively to the first numerical value in these numerical value, second value and the 3rd number Value.

10. image capture unit as claimed in claim 8 or 9, it is characterised in that this processing unit will These numerical value respectively obtain these operation values so that mathematical function performs mathematical calculations.

11. image capture units as claimed in claim 10, it is characterised in that this mathematical function includes Logarithmic function, this target is speculated with following equation by this processing unit relative to position:

$P_{21}=P_2-\frac{1}{2}(P_2-P_1)+\frac{\log M_2-\log M_3}{(\log M_2-\log M_1)+(\log M_2-\log M_3)}\×(P_2-P_1)$

Wherein at P₁Represent this primary importance, P₂Represent this second position, P₃Represent the 3rd position, M₁Table Show this first numerical value, M₂Represent this second value, M₃Represent this third value, P₂₁Represent this target phase para-position Put.

12. image capture units as claimed in claim 7, it is characterised in that this second amount of space is less than This first amount of space.