CN102213581A - Object measuring method and system - Google Patents

Abstract

A method for measuring the stereo coordinate of object includes such steps as using two non-parallel image pick-up devices to pick up the calibration points of known stereo coordinate, deriving the collinearity function of light beams of said image pick-up devices, calibrating said image pick-up devices to obtain the internal and external parameters of said image pick-up devices, and finding out the collinearity function of light beams of said image pick-up devices. Then, the image capturing devices respectively capture images of the target object, so that the processing module calculates the three-dimensional coordinate of the object according to the light beam intersection collinear function. Therefore, the invention not only can rapidly obtain the three-dimensional coordinate and the posture of the target object, but also can improve the accuracy and the convenience of the target object in measurement, and is beneficial to the application of various working environments.

Description

Object measurement method and system

technical field

The present application relates to a method and system for object measurement, in particular, to an object measurement method and system for calculating the three-dimensional coordinates of an object by using two non-parallel image capture devices according to the intersection and collinear function of light beams.

Background technique

Due to the rapid evolution of science and technology, whether it is in the field of commodity design, industrial manufacturing or high-precision operation, there are more and more operating procedures that need to be carried out by automation systems such as robots or robot arms. Therefore, how to improve the automation system Operational efficiency has also become an important issue. The key lies in how to enable automatic systems such as robots or robotic arms to accurately identify the three-dimensional coordinates of objects in space. Accordingly, various measurement methods that can measure the three-dimensional coordinates of objects have emerged as the times require.

For example, in the object measurement method disclosed in US Patent Application No. US 06795200, a structured light source is first projected on the plane to be measured, and then two cameras arranged in parallel are used to respectively obtain images of objects on the plane to be measured. However, in actual use, the arrangement and configuration of the structured light source will often increase the additional burden on the user. Secondly, when using simple triangular geometry principles to calculate the three-dimensional coordinates, because the observation error of the camera itself cannot be taken into account, the accuracy of the calculated three-dimensional coordinates of the object is insufficient, and the insufficient three-dimensional coordinates will make the system Excessive errors are generated in the follow-up operation, so the above-mentioned US 06795200 patent application is not only not practical, but also cannot be applied in the field of high-precision operation.

Furthermore, in the object measurement method disclosed in the U.S. Patent Application No. US 20060088203, a plurality of fixed cameras are set up above the working area at the same time, so as to perform three-dimensional imaging operations on the objects in the working area, and then, After the image acquisition is completed, the three-dimensional coordinates of the object are calculated by using the simple triangular geometry principle. However, the three-dimensional imaging operation by fixing multiple cameras on the working area is not only expensive, but also has poor flexibility in use. At the same time, it is easy to hinder the progress of the three-dimensional imaging operation due to blind spots of vision, which is not applicable. In the field of high-precision operation.

In addition, the European patent application WO 2008076942 also discloses an object measurement method, which firstly installs a single camera on a mobile robot arm, so as to use the mobile robot arm to perform multiple and different measurements on objects in the work area. Angle imaging operation, and then calculate the three-dimensional coordinates of the object through the simple triangular geometry principle. However, it takes extra time to use a single camera to take images of objects in the working area multiple times and from different angles, which not only increases the cost, but also reduces the practicability. Secondly, the same as the aforementioned US 06795200 patent application and US 20060088203 patent application, the three-dimensional coordinates of the object calculated by the simple triangular geometry principle will also cause excessive errors in the subsequent operations of the system. It cannot be applied to extremely fine operating fields.

In view of this, how to provide an object measurement method and system for measuring the three-dimensional coordinates of objects, which can not only obtain the three-dimensional coordinates of objects conveniently, quickly and accurately, but also be applicable to high-precision operation fields, is urgently requested by all walks of life. urgent issues to be resolved.

Contents of the invention

In order to achieve the above and other objectives, the present invention proposes a method for measuring an object, which uses a set of non-parallel first image capture devices and second image capture devices that are arranged side by side and rotate inwards and are connected with the first and second image capture devices. The processing module connected to the two image capture devices measures the object, and the object measurement method includes the following steps: (1) causing the first image capture device and the second image capture device to respectively capture at least one known stereo The coordinates of the first image and the second image of the lens correction point, and then let the processing module calculate the first lens distortion parameter of the first image capture device according to the first image and the second image through the lens correction algorithm and the second lens distortion parameter of the second image capture device; (2) make the first image capture device and the second image capture device capture images of the same plurality of posture correction points with known three-dimensional coordinates Coordinates, and then let the processing module substitute the stereoscopic coordinates of the attitude correction point, the first lens distortion parameter and the second lens distortion parameter into a geometric function based on the beam intersection collinear imaging principle, wherein the geometric function includes The unknown first lens center and first pose parameter of the first image capture device and the unknown second lens center and second pose parameter of the second image capture device; and (3) making the processing module use the preset The established algorithm calculates the geometric function to solve the first lens center and the first pose parameter of the first image capture device and the second lens center and the second pose of the second image capture device parameters, and substitute the solved first lens center, the first attitude parameter, the second lens center and the second attitude parameter into the geometric function based on the beam intersection collinear imaging principle to generate a pair The first beam intersection collinear function and the second beam intersection collinear function of the first and second image capture devices.

In a preferred form, step (4) is also included, making the first image capture device and the second image capture device simultaneously capture the feature point coordinates of a target object, and capture the first image The feature point coordinates captured by the device and the feature point coordinates captured by the second image capture device are substituted into the first and second beam intersection collinear functions to calculate the three-dimensional space coordinates of the target object.

其中，展开后为In another preferred aspect, the geometric function based on the beam collinear imaging principle in the above step (2) satisfies

Among them, the expanded

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 5</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="k"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">k</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="k"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">k</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="p"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">p</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; )</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

${\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="k"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">k</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 5</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

, and (X _A , Y _A , Z _A ) are the known three-dimensional coordinates of the attitude correction point, (x _c , y _c ) are the attitude correction point captured by the first/second image capture device Image coordinates, f is the known focal length of the first/second image capture device, k ₀ , k ₁ , k ₂ , p ₁ , p ₂ are the distortion parameters of the first/second lens, (X _L , Y _L , Z _L ) is the first/second lens center, where, m ₁₁ =cosφcosκ, m ₁₂ =sinωsinφcosκ+cosωsinκ, m ₁₃ =-cosωsinφcosκ+sinωsinκ, m ₂₁ =-cosφsinκ, m ₂₂ =-sinωsinφsinκ+ cosωcosκ, m ₂₃ =cosωsinφsinκ+sinωcosκ, m ₃₁ =sinφ, m ₃₂ =-sinωcosφ and m ₃₃ =cosωcosφ, and ω, φ, κ are the first/second attitude parameters.

Secondly, the present invention also proposes a method for measuring an object, which includes the following steps: (1) making the first image capture device and the second image capture device respectively capture at least one first calibration point of known three-dimensional coordinates. image and the second image; (2) make the processing module substitute the parameters of the correction point corresponding to the known stereo coordinates in the first image and the second image into the geometric function based on the principle of beam intersection and collinear imaging, to Using a preset algorithm to calculate the geometric function and then solve the first lens distortion parameter, the first lens center and the first attitude parameter of the first image capture device and the second lens distortion parameter of the second image capture device , the second lens center and the second attitude parameter; and (3) make the processing module solve the first lens distortion parameter, the first lens center, the first attitude parameter, the second lens distortion parameter, The second lens center and the second attitude parameter are substituted into the geometric function based on the beam intersection collinear imaging principle to generate the first beam intersection collinear function and the second beam intersection collinear function corresponding to the first and second image capture devices. Two-beam intersection collinear function.

In addition, the present invention also proposes an object measurement system, including a first image capture device and a second image capture device for capturing images of calibration points and target objects, wherein the first image capture device and the A non-parallel arrangement of the second image capture devices side by side and rotated inward; and a processing module connected to the first and second image capture devices for processing according to the images captured by the first and second image capture devices The image of the correction point is used for lens correction and object measurement, wherein the processing module substitutes the parameters of the image of the correction point into a geometric function based on the principle of beam intersection and collinear imaging, so as to use a preset algorithm to calculate the geometric function and then solving the first lens distortion parameter, the first lens center and the first pose parameter of the first image capture device and the second lens distortion parameter, the second lens center and the second pose parameter of the second image capture device, Then make the first image capture device and the second image capture device simultaneously capture the feature point coordinates of the target object, and use the feature point coordinates captured by the first and the second image capture device, The first lens distortion parameter, the first lens center, the first attitude parameter, the second lens distortion parameter, the second lens center, and the second attitude parameter are substituted into the geometry based on the beam intersection collinear imaging principle function to calculate the three-dimensional space coordinates of the target object.

In summary, the present invention can use two non-parallel image capture devices to capture images of objects and derive the beam intersection collinear function of the image capture device, and then calculate the beam intersection collinear function through the beam intersection collinear function The three-dimensional coordinates of the object, because these image capture devices will first capture the image of the correction points with known three-dimensional coordinates before capturing the image of the object, and then perform lens correction and posture correction on these image capture devices , therefore, the accuracy of the measured three-dimensional coordinates of the object can be further improved.

Description of drawings

Fig. 1A is the flow chart of object measurement method of the present invention;

FIG. 1B is a flow chart of another embodiment of the object measuring method of the present invention;

Fig. 2 is the structural diagram of object measurement system of the present invention;

Fig. 3 is a diagram of beam intersection relationship of the present invention;

FIG. 4A is a schematic diagram of the image capturing devices of the present invention arranged in parallel; and

FIG. 4B is a schematic diagram of the image capture device of the present invention arranged in a non-parallel arrangement.

[Description of main component symbols]

Steps S1～S5, S1’～S4’

1 Object measurement system

10, 10’ image capture device

11. 11' steering mechanism

12 Fixed base

13 processing module

2 Image screen

A1, A2 Field of view intersection area

Detailed ways

Embodiments of the present invention are described below through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification.

Please refer to FIG. 1A , FIG. 1B and FIG. 2 at the same time. FIG. 1A and FIG. 1B are flow charts of the object measuring method of the present invention, and FIG. 2 is a structure diagram of the object measuring system of the present invention.

The process of FIG. 1A is applied to the object measurement system 1 shown in FIG. 2, which includes at least one set of non-parallel image capture devices 10 and image capture devices 10' arranged side by side and rotated inwardly, and a steering mechanism 11. and a steering mechanism 11', a fixed base 12, and a processing module 13 connected to the image capturing devices 10, 10'.

In this embodiment, the image capture device 10 and the image capture device 10' can be, for example, a video camera or a digital camera including a Charge Coupled Device (CCD), and are respectively fixed on a movable turntable, for example. The steering mechanism 11 and the steering mechanism 11 ′, and the steering mechanisms 11 and 11 ′ can be rotatably arranged on the fixed base 12 with scales. Furthermore, the processing module 13 can be a computer or a micro-processing chip with a logic operation function.

When executing step S1, the image capture device 10 and the image capture device 10' connected to the processing module 13 can be rotatably arranged on the fixed base 12 through the steering mechanism 11 and the steering mechanism 11' respectively, Then adjust the steering angle of the steering mechanism 11 and the steering mechanism 11' according to the three-dimensional coordinates of at least one calibration point, so that the image capture device 10 and the image capture device 10' are simultaneously aligned with the calibration point and arranged in a non-parallel manner. The way is set on the fixed base 12.

During specific implementation, the distance between the image capture device 10 and the image capture device 10' may be no greater than 10 cm, such as 5 cm, and the fixed base 12 may be further arranged on a robot or a robot arm (not shown), Moreover, the processing module 13 can be built into a robot or a robot arm. Of course, the number of image capture devices 10, 10' and steering mechanisms 11, 11' can increase according to user needs, and the processing module 13 can also be a simple The data conversion device transmits the data obtained by the image capture device 10 and the image capture device 10' to an external computing unit (not shown) through a USB, IEEE1394a or IEEE1394b transmission interface for subsequent calculations.

In step S2, let the image capture device 10 and the image capture device 10' respectively capture the first image and the second image of at least one lens correction point with known three-dimensional coordinates, and then let the processing module 13 pass the lens correction algorithm According to the first image and the second image, the lens distortion parameters of the image capture device 10 and the image capture device 10' are obtained respectively, and then proceed to step S3.

In this embodiment, the processing module 13 will first calculate the image coordinates of the lens correction point from the first image and the second image, and then use the lens correction algorithm, such as odd variation, according to the first image and the second image The image coordinates of the correction points in the image are used to obtain the lens distortion parameters of the image capture device 10 and the image capture device 10' respectively, and through the obtained lens distortion parameters, the distortion curve of the lens imaging edge can be adjusted as straight line. Furthermore, the above lens distortion parameters may also refer to the radial distortion and cylindrical distortion of the lens of the image capture device 10, 10'.

In step S3, the image capture device 10 and the image capture device 10' simultaneously capture the image coordinates of the same plurality of known three-dimensional coordinates of the posture correction points, and then let the processing module 13 use the three-dimensional coordinates of the posture correction points , the lens distortion parameters of the image capture device 10 and the image capture device 10' are substituted into a geometric function based on the beam intersection collinear imaging principle, wherein the geometric function includes unknown image capture device 10 and image capture Lens center and attitude parameters of the device 10 ′. Then proceed to step S4.

In this embodiment, the geometric function based on the beam collinear imaging principle satisfies And after substituting the mirror distortion vector, it can be expanded as

${\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 5</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="k"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">k</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="k"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">k</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="p"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">p</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; )</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

${\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="k"\; filenames="HSA00000091235600041.png"\; state="\{\{state\}\}">k</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 5</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; k</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

, and at this time, “X _A , Y _A , Z _A ” represent the known three-dimensional coordinates of the attitude correction points, and “x _c , y _c ” are the attitude correction points captured by the image capture devices 10 and 10 ′ The image coordinates, "f" is the known focal length of the image capture device 10, 10', and "k ₀ , k ₁ , k ₂ , p ₁ , p ₂ " are the lens distortion parameters of the image capture device 10, 10' , and "X _L , Y _L , Z _L " are the lens centers of the image capture devices 10, 10'.

Furthermore, m ₁₁ = cosφcosκ, m ₁₂ = sinωsinφcosκ + cosωsinκ, m ₁₃ = -cosωsinφcosκ + sinωsinκ, m ₂₁ = -cosφsinκ, m ₂₂ = -sinωsinφsinκ + cosωcosκ, m ₂₃ = cosωsinφsinκ + _sinωcos = φm, , m ₃₂ =-sinωcosφ and m ₃₃ =cosωcosφ, and ω, φ, κ are the attitude parameters of the image capturing devices 10 and 10 ′.

In step S4, let the processing module 13 use a preset algorithm, such as the numerical iteration method or the least square method, to calculate the geometric function, so as to simultaneously solve the lens of the image capture device 10 and the image capture device 10' Center and attitude parameters, and the solved image capture device 10 and the lens center and attitude parameters of the image capture device 10' are substituted into the aforementioned geometric functions based on the beam intersection collinear imaging principle to generate corresponding Beam intersection collinear function of image capture device 10 and image capture device 10'

In order to clearly describe the aforementioned steps S2-S4, please refer to FIG. 3 , and take the image capture device 10 as an example to illustrate the calibration point A (X _A , Y _A , Z _A ), calibration point A (X _A , Y _A , Z _A ) The image coordinates A _a (x _c , y _c ) of , and the position relationship diagram of the lens center L (X _L , Y _L , Z _L ) of the image capture device 10 in three-dimensional space.

Firstly, A(X _A , Y _A , Z _A ) is used as the lens correction point, and after the image capture device 10 performs image capture on A(X _A , Y _A , Z _A ), an image frame 2 will be captured , and the image frame 2 will have image coordinates A _a (x _c , y _c ) of A (X _A , Y _A , Z _A ). Therefore, when the processing module 13 substitutes multiple values of A(X _A , Y _A , Z _A ), multiple values of A _a (x _c , y _c ), and the focal length f of the image capture device 10 into the above-mentioned light beam After the geometric function based on the principle of collinear imaging, the processing module 13 can calculate the lens distortion parameters “k ₀ , k ₁ , k ₂ , p ₁ , p ₂ ” of the image capture device 10, and then the image capture device 10Complete lens correction. Of course, the same method can be used to complete the lens correction for the image capture device 10 ′. It is worth mentioning that the lens correction for the image capture device 10 and the image capture device 10 ′ can be implemented simultaneously or sequentially.

Next, A(X _A , Y _A , Z _A ) is used as the attitude correction point. Therefore, after the processing module 13 calculates the values of the attitude parameters "ω, ψ, and K" of the image capture device 10, because "ω, ψ, and K" can represent the deflection angles between the image capture device 10 and the direction axis in space, so the processing module 13 can complete the attitude of the image capture device 10 through the values of "ω, ψ, and K". Correction. It is worth mentioning that the posture corrections for the image capture device 10 and the image capture device 10' must be performed synchronously.

Finally, the processing module 13 can solve the lens center of the image capture device 10, which is L(X _L , Y _L , Z _L ) in the figure, so when the processing module 13 captures the solved image The values of the lens center L (X _L , Y _L , Z _L ) of the device 10, and the values of the attitude parameters "ω, ψ, and K" of the image capture device 10 are substituted into the geometry based on the principle of cross-beam collinear imaging. In the function, the beam intersection collinear function corresponding to the image capture device 10 can be generated. Of course, the beam intersection collinear function of the image capture device 10 ′ can also be generated by the same method.

In step S5, let the image capture device 10 and the image capture device 10' simultaneously capture the feature point coordinates of a target object, and let the processing module 13 capture the image capture device 10 and the image capture device 10' The obtained feature point coordinates are respectively substituted into the beam intersection collinear functions of the image capture device 10 and the image capture device 10 ′, and then the three-dimensional space coordinates of the target object are calculated.

In this embodiment, the processing module 13 can also perform matching and similarity judgments between the feature point coordinates captured by the image capture device 10 and the feature point coordinates captured by the image capture device 10', and then establish The plane equation of the target object, and calculate the three-dimensional space coordinates and posture of the target object through the established plane equation.

In order to illustrate the method of step S5 more clearly, please refer to Fig. 3 again. At this time, A(X _A , Y _A , Z _A ) represents the feature point of the target object, and A _a (x _c , y _c ) represents the feature point of the target object. The image coordinates of the feature point A (X _A , Y _A , Z _A ), and L (X _L , Y _L , Z _L ) are the same as above, representing the lens center of the image capture device 10 .

Therefore, when the image capture device 10 and the image capture device 10 respectively capture images of A(X _A , Y _A , Z _A ), two sets of A(X _A , Y A , Z _{A )} will be obtained from the image frame 2. The image coordinates A _a (x _c , y _c ) of Z _A ), whereby the processing module can replace the values of the two sets of A _a (x _c , y _c ) back to the image capture device 10 and the image capture device respectively 10′ beam intersection collinear function, and then solve the value of A(X _A , Y _A , Z _A ), and obtain the spatial coordinates of the feature points of the target object.

It should be noted here that the spatial coordinates can be accurately calculated to generate stereoscopic vision because the position of the calibration point or the target object must be covered by the intersection area of the field of view of the image capture device 10 and the image capture device 10 ′. The size of the area where the field of view intersects directly affects the accuracy of the calculation results. On the other hand, when the intersection area of the field of view of the image capture device 10 and the image capture device 10' is closer to the camera, it is less likely to occur due to the distance between the image capture device 10 and the image capture device or the calibration point. The near-distance out-of-focus situation occurs when the device 10' is too close. Therefore, the present invention sets the image capture device 10 and the image capture device 10' in a non-parallel arrangement, which can be more suitable for high-definition operation fields .

In order to clearly illustrate the advantages of the present invention that the image capture device 10 and the image capture device 10' are arranged in a non-parallel manner, please refer to FIGS. 4A and 4B, wherein FIG. 4A shows the image in FIG. 2 A schematic view of the field of view of the capture device 10 and the image capture device 10' arranged in parallel, and FIG. 4B is a diagram illustrating the non-parallel setup of the image capture device 10 and the image capture device 10' in FIG. 2 Schematic diagram of the field of view. As shown in FIG. 4A , the image capture device 10 and the image capture device 10 ′ arranged in parallel will generate a visual field intersection area A1, and the visual field intersection area A1 and the parallel image capture device 10 and image capture device The distance between the capturing device 10' is d1; and as shown in FIG. 4B, the image capturing device 10 and the image capturing device 10' arranged in a non-parallel manner will generate a field of view intersection area A2, and the field of view intersection area A2 and The distance between the non-parallel image capture device 10 and the image capture device 10' is d2. It can be seen from the comparison that the area of the visual field intersection area A2 is larger than the area of the visual field intersection area A1, and the length of the distance d1 is greater than the length of the distance d2, so the non-parallel image capture device 10 and the image capture device 10' are more suitable for In the field of extremely fine operation.

In addition, please refer to FIG. 1B again to further illustrate another embodiment of the object measuring method of the present invention.

In the step S1' of this embodiment, the first image capture device 10 and the second image capture device 10' may be firstly ordered to respectively capture a first image and a second image of at least one calibration point with known three-dimensional coordinates; Next, in step S2', let the processing module 13 substitute the parameters corresponding to the first image and the second image into the geometric function based on the principle of intersection and collinear imaging of beams, so as to calculate the geometric function, and then solve the first lens distortion parameter, the first lens center and the first attitude parameter of the first image capture device 10, and the second lens distortion parameter, the second lens center and the second image capture device 10' The second attitude parameter; in step S3', the processing module 13 can obtain the first lens distortion parameter, the first lens center, the first attitude parameter, the second lens distortion parameter, the second lens center and the second The attitude parameters are substituted into the geometric function based on the beam intersection collinear imaging principle to generate a first beam intersection collinear function and a second beam intersection collinear function corresponding to the first and second image capture devices.

Of course, in step S4' of this embodiment, the first image capture device 10 and the second image capture device 10' can be used to simultaneously capture the feature point coordinates of a target object, and the first image capture device The coordinates of the feature points captured by 10 and the coordinates of the feature points captured by the second image capture device 10' are substituted into the aforementioned first and second beam intersection collinear functions to calculate the three-dimensional space coordinates of the target object .

It should be noted here that the difference from the previous embodiments is that this embodiment only needs to capture the calibration point image once (that is, the first image capture device 10 and the second image capture device 10' in this embodiment only The correction point images are captured once for correction, but there may be multiple correction points in the image), and the processing module 13 of this embodiment obtains the first lens distortion parameter of the first image capture device 10 synchronously, the first The attitude parameter, the first lens center, and the second lens distortion parameter, the second attitude parameter, and the second lens center of the second image capture device 10 ′.

In other words, in this embodiment, for example, by adjusting the steering angles of the first image capture device 10 and the second image capture device 10 ′, a single calibration point can be used to replace the lens calibration point and the attitude calibration point of the previous embodiment. , and then make the processing module 13 synchronously complete the lens correction and posture correction of the first image capture device 10 and the second image capture device 10'. The calculation method and related parameters and functions in this embodiment are the same as those in the previous embodiment, so they will not be repeated here.

In summary, the present invention utilizes two non-parallel image capture devices to capture the image of the target object, and calculates the three-dimensional coordinates of the target object according to the beam intersection collinear function of the image capture device, therefore, it is convenient , Get the three-dimensional coordinates of the target object quickly and accurately. At the same time, since these image capture devices first capture images of calibration points with known stereo coordinates before capturing images of objects, and perform lens correction and posture correction on these image capture devices accordingly, therefore, The accuracy of the measured three-dimensional coordinates of the object can be further improved. Accordingly, the present invention can not only quickly obtain the three-dimensional coordinates and attitude of the target object, but also improve the accuracy and convenience of measuring the target object, which is beneficial to the application in various working environments.

The above implementation types are only illustrative to illustrate the principles and effects of the present invention, but are not intended to limit the present invention. Those skilled in the art can modify and change the above embodiments without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be listed in the appended claims.

Claims (20)

1. object measuring method, first image capturing device that utilizes one group of non-parallel setting and second image capturing device and the processing module that is connected with this first and second image capturing device are measured object, and this method may further comprise the steps:

(1) make this first image capturing device and this second image capturing device capture first image and second image of the check point of at least one known spatial coordinate respectively;

(2) make this processing module with in this first image and this second image to the parameter substitution of check point that should known spatial coordinate geometric function based on light beam intersection collinear image formation principle, twist parameter, second optical center and second attitude parameter with second camera lens that utilizes default algorithm to calculate this geometric function and then to solve the distortion of first camera lens parameter, first optical center and first attitude parameter and this second image capturing device of this first image capturing device; And

(3) make this processing module with this first camera lens distortion parameter of being solved, this first optical center, this first attitude parameter, this second camera lens distortion parameter, this second optical center and this second attitude parameter substitution should geometric function based on light beam intersection collinear image formation principle in, with generation to the first light beam intersection conllinear function and the second light beam intersection conllinear function that should first and second image capturing device.

2. object measuring method as claimed in claim 1, also comprise step (4), make this first image capturing device and this second image capturing device capture the unique point coordinate of a target piece simultaneously, and in this first and second light beam intersection conllinear function of unique point coordinate substitution of being captured of the unique point coordinate that this first image capturing device is captured and this second image capturing device, to calculate the solid space coordinate of this target piece.

3. object measuring method as claimed in claim 2, wherein, this step (4) comprises that also the unique point coordinate with this first image capturing device mates the plane equation formula of setting up this target piece to similar judgement to the unique point coordinate of this second image capturing device, to calculate the solid space coordinate and the attitude of this target piece by this plane equation formula.

4. object measuring method as claimed in claim 1, wherein, the algorithm of this step (2) is to utilize strange variation that the distortion curve at lens imaging edge is adjusted into straight line, and the algorithm that should preset is the numerical value process of iteration.

5. object measuring method as claimed in claim 1, wherein, should satisfy in the step (2) based on the geometric function of light-beam collinear image-forming principle $\begin{array}{c}{x}_c=-f\left[\frac{m_{11}(X_A-X_L)+m_{12}(Y_A-Y_L)+m_{13}(Z_A-Z_L)}{m_{31}(X_A-X_L)+m_{32}(Y_A-Y_L)+m_{33}(Z_A-Z_L)}\right]\\ y_c=-f\left[\frac{m_{21}(X_A-X_L)+m_{22}(Y_A-Y_L)+m_{23}(Z_A-Z_L)}{m_{31}(X_A-X_L)+m_{32}(Y_A-Y_L)+m_{33}(Z_A-Z_L)}\right]\end{array},$ Wherein, after the expansion be

$x_c=k_2{\stackrel{\&OverBar;}{x}}^5+(k_1+{2k}_2{\stackrel{\&OverBar;}{y}}^2){\stackrel{\&OverBar;}{x}}^3+\left({3p}_1\right){\stackrel{\&OverBar;}{x}}^2+(1+k_0+k_1{\stackrel{\&OverBar;}{y}}^2+k_2{\stackrel{\&OverBar;}{y}}^4+{2p}_2\stackrel{\&OverBar;}{y})\stackrel{\&OverBar;}{x}+p_1{\stackrel{\&OverBar;}{y}}^2,$

$y_c=k_2{\stackrel{\&OverBar;}{y}}^5+(k_1+{2k}_2{\stackrel{\&OverBar;}{x}}^2){\stackrel{\&OverBar;}{y}}^3+\left({3p}_2\right){\stackrel{\&OverBar;}{y}}^2+(1+k_0+k_1{\stackrel{\&OverBar;}{x}}^2+k_2{\stackrel{\&OverBar;}{x}}^4+{2p}_2\stackrel{\&OverBar;}{x})\stackrel{\&OverBar;}{y}+p_1{\stackrel{\&OverBar;}{x}}^2$

And (X _A, Y _A, Z _A) be the known spatial coordinate of this check point, (x _c, y _c) be this first/the second image capturing device this image coordinate to this check point acquisition, f is the known focal length of this first/the second image capturing device, k ₀, k ₁, k ₂, p ₁, p ₂Be this first/the second camera lens distortion parameter, (X _L, Y _L, Z _L) be this first/the second optical center, and m ₁₁=cos φ cos κ, m ₁₂=sin ω sin φ cos κ+cos ω sin κ, m ₁₃=-cos ω sin φ cos κ+sin ω sin κ, m ₂₁=-cos φ sin κ, m ₂₂=-sin ω sin φ sin κ+cos ω cos κ, m ₂₃=cos ω sin φ sin κ+sin ω cos κ, m ₃₁=sin φ, m ₃₂=-sin ω cos φ and m ₃₃=cos ω cos φ, and ω, φ, κ are this first/the second attitude parameter.

6. object measuring method as claimed in claim 1, wherein, this first/the second camera lens distortion parameter is meant the radial distortion and the tubular distortion of the camera lens of this first/the second image capturing device.

7. object measuring method, first image capturing device that utilizes one group of non-parallel setting and second image capturing device and the processing module that is connected with this first and second image capturing device are measured object, and this method may further comprise the steps:

(1) makes this first image capturing device and this second image capturing device capture first image and second image of the lens correction point of at least one known spatial coordinate respectively, make this processing module try to achieve first camera lens distortion parameter of this first image capturing device and second camera lens distortion parameter of this second image capturing device respectively according to this first image and this second image again by the lens correction algorithm;

(2) make this first image capturing device and this second image capturing device capture the image coordinate of the attitude correction point of identical a plurality of known spatial coordinates, make this processing module with the spatial coordinate of this attitude correction point, this first camera lens distortion parameter and the substitution of this second camera lens distortion parameter geometric function again based on light beam intersection collinear image formation principle, wherein, comprise first optical center of this unknown first image capturing device and second optical center and second attitude parameter of this second image capturing device of first attitude parameter and the unknown in this geometric function; And

(3) make the default algorithm of this processing module utilization calculate this geometric function, with this first optical center that solves this first image capturing device and this second optical center and this second attitude parameter of this first attitude parameter and this second image capturing device, and with this first optical center that is solved, this first attitude parameter, in geometric function that this second optical center and this second attitude parameter substitution should be based on light beam intersection collinear image formation principles, with generation to the first light beam intersection conllinear function and the second light beam intersection conllinear function that should first and second image capturing device.

8. object measuring method as claimed in claim 7, also comprise step (4), make this first image capturing device and this second image capturing device capture the unique point coordinate of a target piece simultaneously, and in this first and second light beam intersection conllinear function of unique point coordinate substitution of being captured of the unique point coordinate that this first image capturing device is captured and this second image capturing device, to calculate the solid space coordinate of this target piece.

9. object measuring method as claimed in claim 8, wherein, this step (4) comprises that also the unique point coordinate with this first image capturing device mates the plane equation formula of setting up this target piece to similar judgement to the unique point coordinate of this second image capturing device, with by this plane equation formula to calculate the solid space coordinate and the attitude of this target piece.

10. object measuring method as claimed in claim 7, wherein, the lens correction algorithm of this step (1) is to utilize strange variation that the distortion curve at lens imaging edge is adjusted into straight line.

11. object measuring method as claimed in claim 7, wherein, should satisfy in the step (2) based on the geometric function of light-beam collinear image-forming principle $\begin{array}{c}{x}_c=-f\left[\frac{m_{11}(X_A-X_L)+m_{12}(Y_A-Y_L)+m_{13}(Z_A-Z_L)}{m_{31}(X_A-X_L)+m_{32}(Y_A-Y_L)+m_{33}(Z_A-Z_L)}\right]\\ y_c=-f\left[\frac{m_{21}(X_A-X_L)+m_{22}(Y_A-Y_L)+m_{23}(Z_A-Z_L)}{m_{31}(X_A-X_L)+m_{32}(Y_A-Y_L)+m_{33}(Z_A-Z_L)}\right]\end{array},$ Wherein, after the expansion be

$x_c=k_2{\stackrel{\&OverBar;}{x}}^5+(k_1+{2k}_2{\stackrel{\&OverBar;}{y}}^2){\stackrel{\&OverBar;}{x}}^3+\left({3p}_1\right){\stackrel{\&OverBar;}{x}}^2+(1+k_0+k_1{\stackrel{\&OverBar;}{y}}^2+k_2{\stackrel{\&OverBar;}{y}}^4+{2p}_2\stackrel{\&OverBar;}{y})\stackrel{\&OverBar;}{x}+p_1{\stackrel{\&OverBar;}{y}}^2$

$y_c=k_2{\stackrel{\&OverBar;}{y}}^5+(k_1+{2k}_2{\stackrel{\&OverBar;}{x}}^2){\stackrel{\&OverBar;}{y}}^3+\left({3p}_2\right){\stackrel{\&OverBar;}{y}}^2+(1+k_0+k_1{\stackrel{\&OverBar;}{x}}^2+k_2{\stackrel{\&OverBar;}{x}}^4+{2p}_2\stackrel{\&OverBar;}{x})\stackrel{\&OverBar;}{y}+p_1{\stackrel{\&OverBar;}{x}}^2$

, and (X _A, Y _A, Z _A) be the known spatial coordinate of this attitude correction point, (x _c, y _c) be this first/the second image capturing device this image coordinate to this attitude correction point acquisition, f is the known focal length of this first/the second image capturing device, k ₀, k ₁, k ₂, p ₁, p ₂Be this first/the second camera lens distortion parameter, (X _L, Y _L, Z _L) be this first/the second optical center, and m ₁₁=cos φ cos κ, m ₁₂=sin ω sin φ cos κ+cos ω sin κ, m ₁₃=-cos ω sin φ cos κ+sin ω sin κ, m ₂₁=-cos φ sin κ, m ₂₂=-sin ω sin φ sin κ+cos ω cos κ, m ₂₃=cos ω sin φ sin κ+sin ω cos κ, m ₃₁=sin φ, m ₃₂=-sin ω cos φ and m ₃₃=cos ω cos φ, and ω, φ, κ are this first/the second attitude parameter.

12. object measuring method as claimed in claim 7, wherein, the default algorithm of this step (3) is the numerical value process of iteration.

13. object measuring method as claimed in claim 7, wherein, this first/the second camera lens distortion parameter is meant the radial distortion and the tubular distortion of the camera lens of this first/the second image capturing device.

14. an object measuring system comprises:

First image capturing device and second image capturing device, in order to the image of acquisition check point and target piece, wherein, this first image capturing device and this second image capturing device are by non-parallel setting; And

Processing module, connect this first and second image capturing device, carry out lens correction and object measurement in order to image according to this check point that this first and second image capturing device captured, wherein, this processing module is with the parameter substitution of the image of this check point geometric function based on light beam intersection collinear image formation principle, to utilize default algorithm to calculate this geometric function and then to solve first camera lens distortion parameter of this first image capturing device, second camera lens distortion parameter of first optical center and first attitude parameter and this second image capturing device, second optical center and second attitude parameter, make this first image capturing device and this second image capturing device capture the unique point coordinate of this target piece simultaneously again, and with this first and the unique point coordinate that captured of this second image capturing device, this first camera lens distortion parameter, this first optical center, this first attitude parameter, this second camera lens distortion parameter, in this geometric function of this second optical center and this second attitude parameter substitution, to calculate the solid space coordinate of this target piece based on light beam intersection collinear image formation principle.

15. object measuring system as claimed in claim 14, also comprise in order to connect first steering mechanism of this first image capturing device, in order to connect second steering mechanism of this second image capturing device, and in order to connect the fixed pedestal of this first and second steering mechanism, wherein, this first and second steering mechanism is movably set on this fixed pedestal, to adjust the steering angle of this first and second steering mechanism according to the solid space coordinate of this check point and/or this target piece, this first and second image capturing device is arranged on this fixed pedestal in nonparallel mode.

16. object measuring system as claimed in claim 15, wherein, these steering mechanism are arranged at intervals on this fixed pedestal, make between this first and second image capturing device of this non-parallel setting to have spacing.

17. object measuring system as claimed in claim 16, wherein, this spacing is less than 10 centimeters.

18. object measuring system as claimed in claim 14, wherein, described lens correction, be that this processing module of instruction is tried to achieve first camera lens distortion parameter of this first image capturing device and second camera lens distortion parameter of this second image capturing device, and described attitude correction is that this processing module of instruction solves first attitude parameter of this first image capturing device and second attitude parameter of this second image capturing device.

19. object measuring system as claimed in claim 14, wherein, this first/the second image capturing device is video camera or the camera that comprises charge coupled cell.

20. object measuring system as claimed in claim 14, wherein, this processing module is computing machine or the little process chip with logical operation function.

Images