CN105066906A - Fast high dynamic range three-dimensional measurement method - Google Patents

Abstract

The invention proposes a fast high dynamic range three-dimensional measurement method, which uses a computer to generate four grating stripes; respectively places a polarizer on the optical axis of the projector and the optical axis of the camera, rotates any polarizer, and aligns the optical axis of the projector with the optical axis of the camera The angle between the axes is adjusted to 90 degrees; use a projector to project the generated four grating fringes to the measured object, and use a camera to simultaneously capture the four fringe images generated by the reflection of the measured object; analyze the four fringe images to obtain high-frequency Wrap phase and low-frequency phase; Unwrap the high-frequency wrapped phase to obtain the high-frequency unwrapped phase, and reconstruct the three-dimensional scene of the measured object according to the unwrapped phase. The invention can realize three-dimensional measurement with high dynamic range for dynamic scenes.

Description

A Fast High Dynamic Range 3D Measurement Method

technical field

The invention belongs to the technical field of optical measurement, and in particular relates to a fast three-dimensional measurement method with high dynamic range.

Background technique

The optical three-dimensional measurement method has become an important means for people to obtain three-dimensional profile data due to its advantages of non-contact, high precision and easy implementation. According to different lighting methods, optical three-dimensional measurement is usually divided into active and passive three-dimensional measurement. Passive 3D measurement uses two or more cameras to shoot the measured object from different angles, and recovers the 3D information of the measured surface from the acquired 2D images. The advantage of this type of method is that the measurement system is simple and the data collection is convenient, so it has a wide range of applications in machine vision. The binocular vision method is a representative method of this kind of method. It constructs a function similar to the human eye according to the principle of bionics, and determines the distance from two-dimensional images of different viewing angles. In the field of aeronautical surveying, the aircraft is equipped with a large-field-of-view high-resolution camera to capture image sequences along the flight direction to obtain terrain and landforms. Although the passive method can effectively obtain the three-dimensional information of the surface of the measured object, it has two main disadvantages: the amount of calculation is very large, and a high-speed processor is usually required to realize the measurement; the excessive dependence on the surface texture of the object makes this method not suitable for measuring smooth or Surfaces with fewer feature points.

Active three-dimensional measurement is to use the method of structured light illumination to actively project light signals to the measured object, and then demodulate the three-dimensional information of the measured object from the modulated light field obtained by shooting. Since this type of method uses modulated light to actively modulate the measured surface, it solves the problem of dependence on the surface texture of the object itself, and avoids the use of related algorithms, thereby reducing the amount of data calculation. Commonly used active methods include fringe projection method, time-of-flight method, modulation degree contour method and so on. Among them, the fringe projection method is the most commonly used measurement method at present. In this method, the grating fringes are projected to the measured object, and the camera shoots the grating fringes modulated by the measured object from another angle, and after the fringe decoding, the three-dimensional profile of the measured surface is demodulated. With the continuous development of digital projection equipment, this technology is gradually used in dynamic scene 3D scene measurement. With the high-speed projection of the stripes to the measured surface and the synchronous acquisition of the two-dimensional grating pattern by the camera, and finally through high-speed data processing, high-speed three-dimensional measurement of moving and deformable objects can be realized.

High-speed 3D measurement is of great significance to the fields of online inspection, biomedicine, deformation analysis and so on. However, in the actual measurement process, researchers still face many problems that need to be solved urgently. One of the most frequently encountered problems is how to achieve high-speed measurements with high dynamic range. Because the surface of the object to be measured is currently highly reflective, it is easy to form local highlights. Especially for alloy metal workpieces, the highlights on the surface are particularly obvious. The high light on the surface will cause the pixel saturation of the camera, so the camera cannot obtain the useful light signal at the high light, and finally it is difficult to restore the three-dimensional outline of the area. At the same time, although the traditional multi-exposure technology can achieve three-dimensional measurement with high dynamic range, it is only suitable for the measurement of static scenes because it requires multiple exposures.

Contents of the invention

The purpose of the present invention is to provide a fast three-dimensional measurement method with high dynamic range, which can realize three-dimensional measurement with high dynamic range for dynamic scenes.

In order to solve the above-mentioned technical problems, the present invention provides a fast three-dimensional measurement method with high dynamic range, using a computer to generate four grating stripes; respectively placing a polarizer on the optical axis of the projector and the optical axis of the camera, rotating any polarizer, and projecting The angle between the optical axis of the instrument and the optical axis of the camera is adjusted to 90 degrees; use the projector to project the generated four grating stripes to the object under test, and use the camera to simultaneously capture the four fringe images generated by the reflection of the object under test; for the four fringes The image is analyzed to obtain the high-frequency wrapped phase and the low-frequency phase; the high-frequency wrapped phase is unwrapped to obtain the high-frequency unwrapped phase, and the three-dimensional scene of the measured object is reconstructed according to the unwrapped phase.

Further, the optical expressions of the four grating stripes I ^p ₁ ~ I ^p ₄ generated by computer are as follows,

I ^p ₁ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πfx ^p )

I ^p ₂ (x ^p ,y ^p )=A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p )

I ^p ₃ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p +2π/3)

I ^p ₄ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p +4π/3)

Among them, (x ^p , y ^p ) is the pixel coordinates of the projector, A ^p is the DC component, B ^p is the modulation degree, A ^p =B ^p =127.5, the image pixel depth is 8 bits, f is the first sinusoidal grating fringe Frequency of.

Further, the expression of the intensity distribution of the four fringe images I ^c ₁ to I ^c ₄ captured synchronously by the camera is as follows,

I ^c ₁ (x ^c ,y ^c )=A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _h )

I ^c ₂ (x ^c ,y ^c )=A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l )

I ^c ₃ (x ^c ,y ^c )＝A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l +2π/3)

I ^c ₄ (x ^c ,y ^c )＝A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l +4π/3)

Among them, (x ^c , y ^c ) is the pixel coordinates of the camera, A ^c is the background light intensity, B ^c is the degree of phase modulation, and φ _h is the first fringe image taken The high-frequency phase contained in , φ _l is the second to fourth fringe images taken The low-frequency phase contained in .

Further, perform two-dimensional Fourier transform on the first fringe image I ^c ₁ , use Hanning window to filter the positive first-order spectrum in the obtained spectrum, and then perform two-dimensional inverse Fourier transform to obtain the high-frequency package Phase φ _h .

Further, use the three-step phase shift method to analyze the second to fourth fringe images I ^c ₂ , I ^c ₃ and I ^c ₄ to obtain the low-frequency phase φ _l , the calculation formula is as follows,

${\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; the\; tan</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\sqrt{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; .</span>$

Further, the specific method of obtaining the phase Φ after unwrapping is shown in the following formula:

$\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; R</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; \&lsqb;</span>\frac{{\mathrm{<span\; class="notranslate">\; f\&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}<span\; class="notranslate">\; \&rsqb;</span>$

Among them, Round is to find the nearest integer.

Compared with the prior art, the present invention has the remarkable advantage that the present invention uses a polarizer to well solve the highlight problem existing in the prior art, so that three-dimensional measurement with a high dynamic range can be realized under the condition of a single exposure. In addition, according to traditional methods, such as the dual-frequency three-step phase-shift method, dynamic scene measurement usually requires six fringes, while the method of the present invention only needs to use four grating fringes, so the present invention is very suitable for the measurement of fast-moving deformed objects.

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Description of drawings

Fig. 1 is a schematic flow chart of the rapid high dynamic range three-dimensional measurement method of the present invention.

Fig. 2 is a measurement object and its scene image used in the embodiment of the present invention, the measurement object is a white water cup with local highlights in the image.

Fig. 3 is four fringe images taken to the measurement object shown in Fig. 2 in the embodiment of the present invention, wherein, (a) is the first fringe image taken (b) is the second fringe image taken (c) is the third fringe image taken (d) is the fourth fringe image taken

Fig. 4 is a schematic diagram of phases solved in an embodiment of the present invention, where (a) is the high-frequency phase φ _h , and (b) is the low-frequency phase φ _l .

Fig. 5 is a comparison of 3D reconstructed images obtained in the embodiment of the present invention, wherein (a) is a 3D image reconstructed using a traditional method, and (b) is a 3D image reconstructed using a method of the present invention.

Detailed ways

It is easy to understand that, according to the technical solution of the present invention, those skilled in the art can imagine various implementations of the rapid high dynamic range three-dimensional measurement method of the present invention without changing the essence of the present invention. Therefore, the following specific embodiments and drawings are only exemplary descriptions of the technical solution of the present invention, and should not be regarded as the entirety of the present invention or as a limitation or limitation on the technical solution of the present invention.

In conjunction with Fig. 1, the fast high dynamic range three-dimensional measurement method of the present invention, the steps are as follows:

Step 1, using a computer to generate four required grating stripes. Taking the generation of vertical stripes as an example, four grating stripes I ^p ₁ to I ^p ₄ and four grating stripes I ^p ₁ to I ^p ₄ are generated as follows:

I ^p ₁ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πfx ^p )

I ^p ₂ (x ^p ,y ^p )=A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p )

I ^p ₃ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p +2π/3)

I ^p ₄ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p +4π/3)

Among them, (x ^p , y ^p ) is the pixel coordinates of the projector, A ^p is the DC component, B ^p is the modulation degree, A ^p =B ^p =127.5, the image pixel depth is 8 bits, f is the first sinusoidal grating fringe Frequency of.

Step two, place the polarizer. Place a polarizer on the optical axis of the projector and the optical axis of the camera, and rotate any one of the polarizers to adjust the angle between the optical axis of the projector and the light transmission axis of the optical axis of the camera to 90 degrees.

Step 3, collecting the grating fringes and solving the phase.

First, the projector is used to project the generated four grating fringes to the measured object, and the camera synchronously collects the four fringe images generated by the reflection of the measured object, and the four fringe images I ^c ₁ (x ^c ,y ^c )～I ^c ₄ (x ^c ,y ^c ) can be expressed as:

I ^c ₁ (x ^c ,y ^c )=A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _h )

I ^c ₂ (x ^c ,y ^c )=A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l )

I ^c ₃ (x ^c ,y ^c )＝A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l +2π/3)

I ^c ₄ (x ^c ,y ^c )＝A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l +4π/3)

Among them, (x ^c , y ^c ) is the pixel coordinates of the camera, A ^c is the background light intensity, B ^c is the degree of phase modulation, and φ _h is the first fringe image taken The high-frequency phase contained in , φ _l is the second to fourth fringe images taken The low-frequency phase contained in .

Second, for the first fringe image Analysis was performed using Fourier profilometry to obtain the high-frequency wrapping phase φ _h . Using two-dimensional Fourier transform, in the obtained spectrum, use a Hanning window with a size of M×N to filter the positive first-order spectrum, and finally perform two-dimensional inverse Fourier transform to obtain the high-frequency wrapping phase φ _h .

Finally, for the second to fourth fringe images Using the three-step phase-shift method analysis, the low-frequency phase φ _l is obtained. The low-frequency phase φ _l is specifically obtained by the following formula:

${\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; the\; tan</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\sqrt{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span>}$

Step 4, three-dimensional reconstruction of the scene under test. Unwrap the high-frequency wrapped phase φ _h to obtain the high-frequency unwrapped phase Φ, and reconstruct the measured 3D scene according to the unwrapped phase Φ. The specific method to obtain the phase Φ after unwrapping is as follows:

$\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; R</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; \&lsqb;</span>\frac{{\mathrm{<span\; class="notranslate">\; f\&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}<span\; class="notranslate">\; \&rsqb;</span>$

Among them, Round is to find the nearest integer. After obtaining the phase Φ after dewrapping, the reconstruction of the measured 3D scene according to the phase Φ after dewrapping can refer to the phase depth conversion method described in the document "Automatic identification and removal of outliers for high-speed fringe projection profilometry" (OptEng2013; 52013605-013605).

Example

In this embodiment, as shown in FIG. 2 , the measured object is a water cup with high light reflection on the surface. The projection of four grating fringes generated in step 1 of the present invention is used for measurement. When the polarizer is not added, the three-dimensional measurement result is shown in Figure 5 (a). It can be clearly seen that pits appear on the surface of the highlight area, indicating that the highlight is There are measurements that affect the surface of that part. After adding a polarizer on the optical axis of the projector and the optical axis of the camera according to the method of the present invention in step 2, the four fringe images collected are successively shown in (a) to (d) in Figure 3, and it is not difficult to find from Figure 3 , the original highlight has been weakened. Subsequently, the fringe image shown in Figure 3(a) was analyzed by Fourier profilometry in step 3, and the high-frequency _wrapping phase _φh was obtained, which is shown in Figure 4(a). Then use the three-step phase shift method in step 3 to process the three fringe images shown in (b) to (d) in Figure 4 to obtain the low-frequency phase φ _l , which is shown in _Figure 4(b). Finally, use the three-dimensional reconstruction method in step four to obtain the three-dimensional outline of the water cup under test, as shown in Figure 5(b), where mm represents millimeters, and Pixel represents pixels. It can be seen that by using the method of the present invention, the three-dimensional image can also be correctly measured and reconstructed on the object surface with high light.

Claims (6)

1. A fast high dynamic range three-dimensional measurement method, characterized in that, Use computer to generate four raster stripes; Place a polarizer on the optical axis of the projector and the optical axis of the camera respectively, rotate any polarizer, and adjust the angle between the optical axis of the projector and the optical axis of the camera to 90 degrees; Use the projector to project the generated four grating fringes to the measured object, and use the camera to simultaneously capture the four fringe images generated by the reflection of the measured object; analyze the four fringe images to obtain the high-frequency wrapped phase and low-frequency phase; Unwrap the high-frequency wrapped phase, obtain the high-frequency unwrapped phase, and reconstruct the three-dimensional scene of the measured object according to the unwrapped phase. 2. The fast high dynamic range three-dimensional measurement method according to claim 1, wherein the optical expressions of the four grating stripes I ^p ₁ to I ^p ₄ generated by the computer are as follows, I ^p ₁ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πfx ^p ) I ^p ₂ (x ^p ,y ^p )=A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p ) I ^p ₃ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p +2π/3) I ^p ₄ (x ^p ,y ^p )＝A ^p (x ^p ,y ^p )+B ^p (x ^p ,y ^p )cos(2πx ^p +4π/3) Among them, (x ^p , y ^p ) is the pixel coordinates of the projector, A ^p is the DC component, B ^p is the modulation degree, A ^p =B ^p =127.5, the image pixel depth is 8 bits, f is the first sinusoidal grating fringe Frequency of. 3. The fast high dynamic range three-dimensional measurement method according to claim 2, wherein the intensity distribution expressions of the four fringe images I ^c ₁ to I ^c ₄ captured synchronously by the camera are as follows, I ^c ₁ (x ^c ,y ^c )=A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _h ) I ^c ₂ (x ^c ,y ^c )=A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l ) I ^c ₃ (x ^c ,y ^c )＝A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l +2π/3) I ^c ₄ (x ^c ,y ^c )＝A ^c (x ^c ,y ^c )+B ^c (x ^c ,y ^c )cos(φ _l +4π/3) Among them, (x ^c , y ^c ) is the pixel coordinates of the camera, A ^c is the background light intensity, B ^c is the degree of phase modulation, and φ _h is the first fringe image taken The high-frequency phase contained in , φ _l is the second to fourth fringe images taken The low-frequency phase contained in . 4. fast high dynamic range three-dimensional measurement method as claimed in claim 3, is characterized in that, carry out two-dimensional Fourier transform to the first fringe image I ^c ₁ , use Hanning window in the frequency spectrum that obtains to correct one level The spectrum is filtered and then subjected to two-dimensional inverse Fourier transform to obtain the high-frequency wrapped phase φ _h . 5. fast high dynamic range three-dimensional measurement method as claimed in claim 4, is characterized in that, use three-step phase-shift method analysis to the second to the fourth fringe images I ^c ₂ , I ^c ₃ and I ^c ₄ , obtain Low-frequency phase φ _l , the calculation formula is as follows, ${\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; the\; tan</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\sqrt{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; .</span>$ 6. fast high dynamic range three-dimensional measurement method as claimed in claim 5, is characterized in that, the specific method of obtaining the phase Φ after dewrapping is shown in the following formula: $\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; R</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; d</span>}<span\; class="notranslate">\; \&lsqb;</span>\frac{{\mathrm{<span\; class="notranslate">\; f\&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&phi;</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}<span\; class="notranslate">\; \&rsqb;</span>$ Among them, Round is to find the nearest integer.