CN102486869A - Calibration method of linear array camera - Google Patents

Abstract

The embodiment of the invention provides a calibration method of a linear array camera and is used for improving the positioning precision of a touch screen or an electronic white board on the basis of the linear array camera. The method comprises the steps of: obtaining an image position coordinate p of a touch object and an included angle Alpha between a reference line of an optical main point and a connecting line of the position of the touch object with the object space optical main point of the camera; respectively mapping p and Alpha to a specified interval in a linear mapping mode, obtaining a function which has a unique inverse function, establishing a function relationship, taking the effective pixel value interval of the camera to be [0, V), selecting m groups of (x-y) corresponding relationships of (x1, y1), (x2, y2) to (xm, ym), and making the (x-y) corresponding relationships meet the function; and determining parameters ak when the function error is the minimum according to the known m groups of (x-y) corresponding relationships of (x1, y1), (x2, y2) to (xm, ym).

Description

A kind of scaling method of linear array camera

Technical field

The present invention relates to the touching technique field, relate in particular to a kind of scaling method of linear array camera.

Background technology

The purpose of optical touch screen or optical electron blank camera calibration is the parameter for the camera imaging model of confirming the geometric relationship between description image coordinate system and touch area coordinate system.As shown in Figure 1, be the synoptic diagram of linear array camera calibration, some A is for touching the object position, and A ' is for touching imaging in the online battle array of the object camera, and some O is the optics principal point of linear array camera.It is exactly the corresponding relation of confirming between some A ' position and the angle α that this linear array camera is demarcated.

The inventor finds, existing linear array camera calibration method can't satisfy touch-screen or electronic whiteboard to locating high-precision requirement.

Summary of the invention

The embodiment of the invention provides a kind of scaling method of linear array camera, to improve the bearing accuracy based on the touch-screen or the electronic whiteboard of linear array camera.

The embodiment of the invention provides a kind of scaling method of linear array camera, and this method comprises:

Obtain the position coordinates p of the picture that touches object and touch the angle between the datum line of line and the said optics principal point of mistake of object position and camera object space optics principal point;

Adopt the linear mapping mode that p and α are mapped to respectively between designation area, obtain function

and this function and have unique inverse function

Set up funtcional relationship

Get

Wherein camera valid pixel value interval is that [0, V), selection m organizes x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), make it satisfy above-mentioned function;

According to known m group x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), function

Error hour definite parameter a _k

The embodiment of the invention provides a kind of scaling method of linear array camera, has improved the bearing accuracy based on the touch-screen or the electronic whiteboard of linear array camera through this method.

Description of drawings

Fig. 1 is the synoptic diagram of linear array camera calibration;

The process flow diagram of the scaling method of a kind of linear array camera that Fig. 2 provides for the embodiment of the invention.

Embodiment

The embodiment of the invention provides a kind of scaling method of linear array camera, in order to improve the bearing accuracy based on the touch-screen or the electronic whiteboard of linear array camera.

As shown in Figure 2, the process flow diagram of the scaling method of a kind of linear array camera that the embodiment of the invention provides, this method comprises:

S2001 obtains the position coordinates p of the picture that touches object and touches the angle between the datum line of line and the said optics principal point of mistake of object position and camera object space optics principal point.

S2002; Adopt the linear mapping mode that p and α are mapped to respectively between designation area, obtain function

and this function and have unique inverse function

S2003 sets up funtcional relationship Get

Wherein camera valid pixel value interval is that [0, V), selection m organizes x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), make it satisfy above-mentioned function.

S2004 is according to known m group x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), function

Error hour definite parameter a _k

The error function of function y (x) is:

Can confirm parameter a according to following method _k:

Function y (x) is a polynomial function, a _k(0≤k≤n) is a parameter;

According to the above-mentioned m (group of m＞n) x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), and satisfy: y _k=y ₁(x _k) (0≤k≤m), x _k((0≤k≤m), error function hour,

Structure n+1 rank square formations

Structure n+1 dimensional vector

Wherein: $a_{i,j}=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_l^{i+j},$ $b_i=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_k^i{y}_k;$

Parameter vector then:

Also can confirm parameter a according to following method _k:

Function y (x) is a polynomial function, a _k(0≤k≤n) is a parameter;

According to the above-mentioned m (group of m＞n) x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), and satisfy: y _k=y ₁(x _k) (0≤k≤m), y _k(0≤k≤m), error function hour,

Structure n+1 rank square formations

Structure n+1 dimensional vector

Wherein: $a_{i,j}=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_l^{i+j},$ $b_i=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_k^i{y}_k;$

Then fitting parameter is vectorial:

The scaling method of above-mentioned linear array camera also can be used for the demarcation of one of them dimension of face battle array camera.

The embodiment of the invention provides a kind of scaling method of linear array camera, has improved the bearing accuracy based on the touch-screen or the electronic whiteboard of linear array camera through this method.

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

Claims (5)

1. the scaling method of a linear array camera is characterized in that, comprising:

Obtain the position coordinates p of the picture that touches object and touch the angle between the datum line of line and the said optics principal point of mistake of object position and camera object space optics principal point;

Adopt the linear mapping mode that p and α are mapped to respectively between designation area, obtain function and this function and have unique inverse function

Set up funtcional relationship

Get

Wherein camera valid pixel value interval is that [0, V), selection m organizes x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), make it satisfy above-mentioned function;

According to known m group x-y corresponding relation (x ₁, y ₁), (x ₂, y ₂) ..., (x _m, y _m), function

Error hour definite parameter a _k

2. method according to claim 1 is characterized in that, the error function of function y (x) does $\mathrm{\Δ}=\frac{\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{(y_k-y\left(x_k\right))}^2}{m}.$

3. method according to claim 2 is characterized in that, said function

Error hour definite parameter a _kBe specially:

Structure n+1 rank square formations

Structure n+1 dimensional vector

Wherein: $a_{i,j}=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_l^{i+j},$ $b_i=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_k^i{y}_k;$

Then parameter vector is:

4. method according to claim 2 is characterized in that, said function Error hour definite parameter a _kBe specially:

Structure n+1 rank square formations

Structure n+1 dimensional vector

Wherein: $a_{i,j}=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_l^{i+j},$ $b_i=\underset{k=1}{\overset{m}{\mathrm{\Σ}}}{x}_k^i{y}_k;$

Then the fitting parameter vector is:

5. according to any described method of claim 1 to 4, it is characterized in that said method is used for the demarcation of one of them dimension of face battle array camera.

Images