阅读说明：本技术 一种快速彩色条纹图相位解调算法及系统 (Rapid color fringe pattern phase demodulation algorithm and system ) 是由 赵宏 朱倩 于 2021-08-17 设计创作，主要内容包括：本发明公开了一种快速彩色条纹图相位解调算法及系统,首先使用标定平板对测量系统进行预标定,求出耦合强度系数；其次使用该系统分别向待测物投影一幅均匀强度图和一幅彩色条纹图,对两幅图分别去耦后,使用均匀强度图求取背景项和调制度,对彩色条纹图进行去背景和色彩归一化,得到三幅无背景的、色彩归一的条纹图；然后使用相移法对这三幅条纹图求取解调相位；最后对得到的解调相位进行去包裹,得到解包裹相位。经过背景项标定后实现了单幅彩色条纹图相移法解调,既保留了传统灰度相移法解调的鲁棒性,又是实现了快速相位解调,满足了动态三维形貌测量的需求。(The invention discloses a rapid color fringe pattern phase demodulation algorithm and a system, wherein a calibration flat plate is used for pre-calibrating a measurement system to obtain a coupling intensity coefficient; secondly, projecting a uniform intensity image and a color fringe image to the object to be measured by using the system, decoupling the two images respectively, then using the uniform intensity image to obtain a background item and a modulation degree, and carrying out background removal and color normalization on the color fringe image to obtain three background-free color normalized fringe images; then, a phase shift method is used for solving demodulation phases of the three fringe patterns; and finally, unwrapping the obtained demodulation phase to obtain an unwrapped phase. After background item calibration, single-frame color fringe pattern phase shift demodulation is realized, the robustness of the traditional gray phase shift demodulation is kept, the rapid phase demodulation is realized, and the requirement of dynamic three-dimensional morphology measurement is met.)

1. A fast color fringe pattern phase demodulation algorithm is characterized in that firstly, a calibration flat plate is used for pre-calibrating a measurement system to obtain a coupling intensity coefficient; secondly, projecting a uniform intensity image and a color fringe image to the object to be measured by using the system, decoupling the two images respectively, then using the uniform intensity image to obtain a background item and a modulation degree, and carrying out background removal and color normalization on the color fringe image to obtain three background-free gray fringe images with colors normalized; then, a phase shift method is used for solving demodulation phases of the three gray scale fringe patterns; and finally, unwrapping the obtained demodulation phase, solving the unwrapped phase, and reconstructing the three-dimensional morphology.

2. The fast color fringe pattern phase demodulation algorithm as claimed in claim 1, comprising the steps of:

step 1: three red uniform intensity maps with consistent intensity are projected to a calibration white boardGreen uniform intensity mapAnd blue uniform intensity mapShooting the projected uniform intensity image, and collecting the red uniform intensity imageGreen uniform intensity mapAnd blue uniform intensity map

Step 2: utilizing the three red, green and blue uniform intensity maps collected in the step 1Calculating the coupling intensity coefficient of each main color channel in the other two channels for pre-calibration;

and step 3: and (3) after the pre-calibration in the step 1 and the step 2 is finished, starting formal measurement. Under the same measuring system, a uniform intensity chart I is projected to the object 4 to be measured^puAnd synchronously shooting images to acquire a projected uniform intensity image I^cu；

And 4, step 4: under the same measuring system, a color fringe pattern I is projected to the measured object^pAnd synchronously shooting and collecting a color fringe pattern I^c；

And 5: using the coupling intensity coefficients to respectively align the uniform intensity maps I^cuAnd color stripe pattern I^cDecoupling to obtain a decoupled uniform intensity map I^cudAnd color stripe pattern I^cd；

Step 6: by utilizing a decoupling uniform intensity diagram, taking one channel as a reference channel, and solving the modulation ratio of the other two color channels to the reference channel respectively;

and 7: for the decoupled color stripe pattern I^cdRemoving the background to obtain a color stripe image I after removing the background^debackWherein Is I^debackMiddle red, green and blueThree different channel gray scale bar

And 8: using the blue channel as a reference, and using the modulation degree ratio obtained in the step 6 to contrast the gray stripe patternCarrying out modulation degree normalization to obtain three gray stripe images with normalized colors;

and step 9: solving demodulation phases for the three color normalized gray scale fringe patterns;

step 10: and unwrapping the demodulation phase, reconstructing the three-dimensional appearance, and covering the acquired uniform intensity map on the reconstructed three-dimensional map to obtain a real three-dimensional image with coexisting texture and color.

3. The algorithm for fast phase demodulation of color fringe patterns as claimed in claim 1, wherein in step 2, the coupling intensity coefficient of each main color channel in the other two channels is calculated by the intensity ratio between different channels.

4. The fast color fringe pattern phase demodulation algorithm as claimed in claim 1, wherein in said step 5, the uniform intensity pattern I^cuThe decoupling formula of (c) is:

wherein the content of the first and second substances,respectively, a uniform intensity map I^cuRed, green, blue channel images; respectively decoupled uniform intensity map I^cudRed channel, green channel, blue channel images, andthe meanings are consistent; k_rg,K_rbCoupling intensity coefficients of the red main color channel to the green channel and the blue channel in the collected red uniform intensity graph respectively; k_gr,K_gbCoupling intensity coefficients of the green main color channel to the red channel and the blue channel in the collected green uniform intensity graph respectively; k_br,K_bgCoupling intensity coefficients of the blue main color channel to the red channel and the green channel in the collected blue uniform intensity map are respectively;

color stripe pattern I^cThe decoupling formula of (c) is:

wherein the content of the first and second substances,respectively, collected color fringe patterns I^cGray stripe patterns on a middle red channel, a green channel and a blue channel;respectively a decoupled colour fringe pattern I^cdThe gray stripe patterns of the red channel, the green channel and the blue channel,are respectively asThe background item of (a) is,are respectively as The modulation degree of (2).

5. The fast color fringe pattern phase demodulation algorithm as claimed in claim 1, wherein in step 6, the modulation ratio M₁,M₂The calculation formula is as follows:

wherein corr1, corr2, corr3 and corr4 are all modulation factors,is a decoupled uniform intensity map I^cudA uniform intensity map on the red, green, and blue channels.

6. The fast color fringe pattern phase demodulation algorithm as claimed in claim 1, wherein in step 9, the demodulation phase is solved for the color-normalized three gray fringe patterns by using a three-step phase shift method.

7. The fast color fringe pattern phase demodulation algorithm as claimed in claim 1, wherein in the step 10, the demodulation phase obtained in the step 9 is unwrapped by using a quality guide algorithm.

8. A rapid color fringe pattern phase demodulation three-dimensional measurement system is characterized by comprising a color camera (1), a projector (2) and an upper computer; the output end and the input end of the color camera (1) and the projector (2) are both connected with an upper computer and used for receiving and transmitting instructions of the upper computer; the projector (2) is used for projecting a red uniform intensity image, a green uniform intensity image and a blue uniform intensity image to the calibration white board and projecting a color fringe image and a uniform intensity image to the object to be measured (4); the color camera (1) is used for shooting a projected uniform intensity image for calibration, a projected image and a uniform intensity image of a color fringe image on a measured object, transmitting the shot image to an upper computer, wherein a calculation program capable of running on the upper computer is stored on the upper computer, and when the processor executes the calculation program, the processor demodulates the phase of the received image, calculates the unwrapping phase and reconstructs the three-dimensional morphology.

Technical Field

The invention belongs to the technical field of optical three-dimensional measurement, and particularly relates to a rapid color fringe pattern phase demodulation algorithm and a system.

Background

In recent decades, dynamic three-dimensional topography measurement techniques have been widely used in various fields, such as machine vision, industrial automatic detection, product quality control, object profiling, biomedicine, and film and television trick manufacturing. In order to achieve fast three-dimensional measurement of dynamic objects, multi-camera stereo vision methods and fringe contouring are commonly employed. The stereoscopic vision has low measurement accuracy due to uncertainty of stereoscopic matching, and the multiple cameras increase measurement cost. The fringe projection profilometry has the advantages of non-contact, full field, high precision and the like, and is widely developed and applied. The fringe projection profilometry method mainly adopts two ways of gray coding and color coding at present, and reconstructs the object appearance through three processes of phase demodulation, phase unwrapping and phase-to-height mapping after acquiring the deformed coding fringes modulated by the object to be detected. The speed and accuracy of phase demodulation are of decisive significance to the measurement of three-dimensional topography, and therefore, researchers have conducted a great deal of research.

In the existing phase demodulation methods, the two most commonly used methods are a space domain method and a phase shift method, wherein although the space domain method can extract the phase by only acquiring one gray stripe image to meet the requirement of dynamic measurement, when the measurement object is an object with a complex surface, the method is easy to generate frequency spectrum aliasing to cause phase extraction errors; the phase shift method is not sensitive to ambient light, can obtain pixel-level phase points, and extracts phases using the intensity relationship of multiple images (more than three images), so the demodulation accuracy is high, however, acquiring multiple images is not acceptable for dynamic measurement, because the same point in the projector pattern sequence may be mapped to different points in the acquired image due to the change of the depth of the object to be measured, resulting in demodulation phase errors, and limiting the accuracy of dynamic measurement. However, with the development of various industries, the measurement requirement for dynamic objects is increasing, and higher requirements for speed and precision are provided, so that a dynamic three-dimensional shape measurement technology with both precision and speed is required.

Based on the above consideration, with the development of color cameras, compared with the traditional single-channel gray-white camera, the color camera has three channels, so that students encode three gray-scale fringe patterns in one color fringe pattern, the number of pictures is reduced, and the efficiency of projection acquisition is improved. But in the use of a color stripe pattern, it is inevitable that there is a problem of color coupling and imbalance between the three channels. For the problem of color coupling, the problem can be removed by a system pre-calibration method; for the problem of Color imbalance, a student Pan et al projects and collects more than three Color fringe patterns, calculates the background item and modulation degree of each channel, and performs Color normalization correction of three channels (Color-encoded digital front projection detection for high-speed 3-D shape measurement).

Disclosure of Invention

The invention aims to provide a phase demodulation method with high robustness, which can quickly solve a demodulation phase and meet the requirement of dynamic measurement, aiming at the problems that the existing dynamic three-dimensional topography measurement technology is limited by a phase demodulation method and the efficiency and the precision cannot be simultaneously considered.

In order to achieve the purpose, the invention provides a rapid color fringe pattern phase demodulation algorithm, which comprises the steps of firstly projecting three red, green and blue uniform patterns on a calibration flat plate, extracting a coupling intensity coefficient, and pre-calibrating a system; (ii) a Secondly, projecting a uniform intensity image and a color fringe image to the object to be measured by using a pre-calibrated system, decoupling the two images, calculating a background item and a modulation degree by using the uniform intensity image, and performing background removal and color normalization on the color fringe image to obtain three background-free and color-normalized gray fringe images; then, a phase shift method is used for solving demodulation phases of the three gray scale fringe patterns; and finally, unwrapping the obtained demodulation phase, solving the unwrapped phase, and reconstructing the three-dimensional morphology.

Further, the method comprises the following steps:

step 1: three red uniform intensity maps with consistent intensity are projected to a calibration white boardGreen uniform intensity mapAnd blue uniform intensity mapShooting the projected uniform intensity image, and collecting the red uniform intensity imageGreen uniform intensity mapAnd blue uniform intensity map

Step 2: utilizing the three red, green and blue uniform intensity maps collected in the step 1Calculating the coupling intensity coefficient of each main color channel in the other two channels;

and step 3: and (3) after the pre-calibration in the step 1 and the step 2 is finished, starting formal measurement. Under the same measuring system, a uniform intensity image I is projected to the object 4 to be measured^puAnd synchronously shooting images to acquire a projected uniform intensity image I^cu；

And 4, step 4: under the same measuring system, a color fringe pattern I is projected to the measured object^pAnd synchronously shooting and collecting a color fringe pattern I^c；

And 5: respectively aligning single uniform intensity chart I by using coupling intensity coefficients^cuAnd a single color stripe pattern I^cDecoupling to obtain a decoupled uniform intensity map I^cudAnd color stripe pattern I^cd；

Step 6: using decoupled uniform intensity map I^cudTaking one channel as a reference channel, and solving the modulation ratio of the other two color channels to the reference channel respectively;

and 7: removing the background of the decoupled color fringe image to obtain a background-removed color fringe image I^deback，Is I^debackGray scale fringe patterns of three different channels of middle red, green and blue;

and 8: taking the blue channel as a reference, and utilizing the modulation degree ratio obtained in the step 6 to remove the background gray fringe patternCarrying out modulation degree normalization to obtain three gray stripe images with normalized colors;

and step 9: solving a demodulation phase by using three gray scale fringe images with color normalization;

step 10: and (4) unwrapping the demodulation phase in the step (9), reconstructing a three-dimensional appearance, and covering the acquired uniform intensity map on the reconstructed three-dimensional map to obtain a real three-dimensional image with coexisting texture and color.

Further, the coupling intensity coefficient of each main color channel in the other two channels is calculated according to the intensity ratio of different channels.

Further, in step 5, uniform intensity map I^cuThe decoupling formula of (c) is:

wherein the content of the first and second substances,respectively, a uniform intensity map I^cuRed, green, blue channel images; respectively, a uniform intensity map I^cuImages of the red channel, the green channel and the blue channel after decoupling, andthe meanings are consistent; k_rg,K_rbRespectively the collected red uniform intensity mapThe coupling strength coefficient of the red main color channel to the green channel and the blue channel; k_gr,K_gbRespectively collected green uniform intensity mapsThe coupling strength coefficient of the green main color channel to the red channel and the blue channel; k_br,K_bgRespectively the collected blue uniform intensity mapThe coupling strength coefficient of the blue main color channel to the red channel and the green channel;

color stripe pattern I^cThe decoupling formula of (c) is:

wherein the content of the first and second substances,respectively a color stripe pattern I^cGray stripe patterns on the red channel, the green channel and the blue channel;respectively a decoupled colour fringe pattern I^cdThe intensities of the red, green and blue channels,are respectively asThe background item of (a) is,are respectively asThe modulation degree of (2).

Further, in step 6, the modulation ratio M₁,M₂The calculation formula is as follows:

wherein corr1, corr2, corr3 and corr4 are all modulation factors,is a decoupled uniform intensity map of the reference channel,andrespectively, the decoupled uniform intensity maps of the other two channels.

Further, in step 9, the demodulation phase is solved for the three color-normalized gray scale fringe patterns by using a three-step phase shift method.

Further, in step 10, the demodulation phase obtained in step 9 is unwrapped by using a quality guidance algorithm.

A fast color fringe pattern phase demodulation system comprises a color camera, a projector and an upper computer; the output end and the input end of the color camera and the projector are both connected with the upper computer and used for receiving and transmitting instructions of the upper computer; the projector is used for projecting a red uniform intensity map, a green uniform intensity map and a blue uniform intensity map to the calibration white board, and projecting a color fringe map and a uniform intensity map to an object to be measured; the color camera is used for shooting a projected uniform intensity image for calibration, a color fringe image and a uniform intensity image projected on a measured object, the shot image is transmitted to the upper computer, a calculation program which can run on the upper computer is stored on the upper computer, and when the processor executes the calculation program, the processor carries out phase demodulation on the received image after processing, calculates a unwrapping phase and reconstructs a three-dimensional shape.

Compared with the prior art, the invention has at least the following beneficial technical effects:

1. in the method, a uniform intensity image and a color fringe image are projected and collected on the object to be measured to realize phase shift demodulation, and more than three fringe images are required for demodulation by the traditional phase shift method, so that the method reduces the amplitude of the collected images, realizes rapid phase demodulation and is beneficial to dynamic three-dimensional measurement. Meanwhile, the size of the (R, G, B) value of each pixel point on the object to be measured can be extracted through the acquired uniform intensity map, the color information of the surface of the object is reserved, and after the three-dimensional reconstruction is completed, the three-dimensional intensity map is covered on the surface texture, so that the surface appearance of the object is more vividly displayed.

2. The invention uses a uniform intensity image to extract the background item of the object to be detected and calculates the modulation ratio between each channel, then uses the modulation ratio to remove the background and normalize the color of the gray stripe images of three channels contained in the color stripe image, and uses the normalized three gray stripe images to demodulate by phase shift method. The method realizes the demodulation of a single color fringe pattern phase shift method, not only retains the robustness of the demodulation of the traditional gray phase shift method, but also improves the measurement efficiency and meets the requirement of dynamic three-dimensional shape measurement;

3. the method of the invention codes the traditional three gray stripe images on one color stripe image, calculates the modulation ratio through a single uniform intensity image, and performs color normalization on the color stripe image, thereby realizing color imbalance correction.

Drawings

Fig. 1 is a flow chart of a fast color fringe pattern phase demodulation algorithm.

FIG. 2a is a simulated color fringe pattern with coupling and noise;

FIG. 2b is a graph of simulated uniform intensity with coupling and noise;

FIG. 2c is a cross-section corresponding to the solid white line in a;

fig. 2d is a cross-section corresponding to the solid white line in b.

FIG. 3a is a cross-section of the color stripe of FIG. 2a after it has been decoupled;

FIG. 3b is a cross-section of the color stripe image of FIG. 3a after removal of the background;

fig. 3c is a cross-section of the stripe image of fig. 3b after color normalization.

FIG. 4a is a demodulation phase of a simulation experiment;

FIG. 4b is a cross-section of the demodulation phase corresponding to the solid white line in FIG. 4 a;

FIG. 4c is a cross-section of a demodulation phase obtained by the 12-step grayscale phase shift method and the method of the present invention;

fig. 4d shows the demodulation phase error of the method of the present invention.

FIG. 5a is a red channel of a uniform intensity map collected in a human face model measurement experiment;

FIG. 5b is a green channel of a uniform intensity map collected in a human face model measurement experiment;

FIG. 5c is a blue channel of a uniform intensity map collected in a human face model measurement experiment;

fig. 5d is a cross-section corresponding to the solid white line in fig. 5a, 5b, 5 c.

FIG. 6a is a color bar collected from a human face model measurement experiment;

FIG. 6b is the color stripe of FIG. 6a after decoupling and background removal;

fig. 6c is the normalized color stripe of fig. 6 b.

FIG. 7a is a demodulation phase of a face model;

FIG. 7b is the demodulation phase of FIG. 7 a;

FIG. 7c is a cross-section of a demodulation phase obtained by the 12-step grayscale phase shift method and the method of the present invention;

FIG. 7d is a diagram of the demodulation phase error of the method of the present invention;

FIG. 8a is a unwrapped phase map of a face model;

fig. 8b is a face model phase diagram covering surface color information.

Fig. 9 is a schematic view of a measurement system.

In the attached figure, 1-a color camera, 2-a projector, 3-an upper computer and 4-an object to be measured.

Detailed Description

In order to make the objects and technical solutions of the present invention clearer and easier to understand. The present invention will be described in further detail with reference to the following drawings and examples, wherein the specific examples are provided for illustrative purposes only and are not intended to limit the present invention.

In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are used only for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention. Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless otherwise specified. In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.

The technical solution for realizing the purpose of the invention is as follows: a fast color fringe pattern phase demodulation algorithm. Firstly, pre-calibrating a measurement system by using a calibration flat plate to calculate a coupling strength coefficient; secondly, projecting a uniform intensity image and a color fringe image to the object to be measured by using the system, decoupling the two images respectively, then using the uniform intensity image to obtain a background item and a modulation degree, and carrying out background removal and color normalization on the color fringe image to obtain three background-free color normalized fringe images; then, a phase shift method is used for solving a demodulation phase; and finally, unwrapping the obtained demodulation phase to obtain an unwrapped phase. And a demodulation phase is quickly obtained by using a color fringe pattern, data is provided for subsequent phase unwrapping, and the three-dimensional shape reconstruction of the dynamic object is realized.

Referring to fig. 1, a fast color fringe pattern phase demodulation algorithm includes the following steps:

the method comprises the following steps: three red uniform intensity maps with consistent intensity are respectively projected to a calibration white board by using a projector 2 in the measurement system of FIG. 9Green uniform intensity mapAnd blue uniform intensity mapThe intensity range is 120-220 (depending on the linear range of the intensity of different cameras), the color camera 1 shoots the projected uniform intensity image, and the red uniform intensity image is collectedGreen uniform intensity mapAnd blue uniform intensity map

Note: in the method, the images acquired by projection all have three channels of red, green and blue.

The projected red, green and blue uniform intensity maps are formulated as follows:

the actually collected red, green and blue uniform intensity maps are formulated as follows:

whereinRespectively, the projected red homogeneity mapA uniform intensity map on the red, green, blue channels of (a);respectively, a projected green uniformity mapA uniform intensity map on the red, green, blue channels of (a);respectively, a projected blue homogeneity mapA uniform intensity map on the red, green, and blue channels.Respectively, the collected red uniformity mapA uniform intensity map on the red, green, blue channels of (a);respectively, the collected green uniformity mapA uniform intensity map on the red, green, blue channels of (a);respectively, the collected blue uniformity mapA uniform intensity map on the red, green, and blue channels.

Step two: three red, green and blue uniform intensity maps collected in the first step are utilizedCalculating the coupling intensity coefficient of each main color channel in other two channels according to the intensity ratio of the different channels;

wherein, K_rg,K_rbRespectively the collected red uniform intensity mapThe coupling strength coefficient of the red main color channel to the green channel and the blue channel; k_gr,K_gbRespectively collected green uniform intensity mapsThe coupling strength coefficient of the green main color channel to the red channel and the blue channel; k_br,K_bgRespectively the collected blue uniform intensity mapAnd (3) coupling strength coefficients of the blue main color channel to the red channel and the green channel.

Note: the first step and the second step are pre-calibration links which are completed before formal measurement. Since color coupling is mainly caused by system hardware equipment and is an inherent problem of overlapping of red, green and blue light spectrums, the coupling relationship can be obtained through pre-calibration in the step, and decoupling data is provided for later acquired pictures.

Step three: and after the pre-calibration in the first step and the pre-calibration in the second step, beginning formal measurement. Under the same measuring system, a projector is used for projecting a uniform intensity chart I to the object 4 to be measured^puSynchronously shooting images through a color camera 1, and acquiring a projected uniform intensity image I^cu。

Projected uniform intensity map I^puThe intensity maps of the red, green and blue channels are respectivelyAnd isAs followsShown in the figure:

actually collected uniform intensity map I^cuThe intensity maps of the red, green and blue channels are respectivelyAnd isAs follows:

wherein the content of the first and second substances,the set projection intensity is consistent with the background item intensity of the color fringe pattern projected in step four.Is the intensity of the uniform map that is actually acquired.

Step (ii) ofFourthly, the method comprises the following steps: under the same measuring system, a projector projects a color fringe pattern I to an object to be measured^p，Respectively projected colour fringe pattern I^pGray stripe patterns on the medium red, green, blue channels, andsynchronously shooting by a color camera to acquire a color fringe pattern I^c，Respectively, collected color fringe patterns I^cGray stripe patterns on the medium red, green, blue channels, andspecifically, the formula is shown as follows:

projected colour fringe pattern I^pThe gray scale fringe patterns of the three channels are respectively:

collected color fringe pattern I^cThe gray scale fringe patterns of the three channels are respectively:

wherein the content of the first and second substances,are respectively asThe background item of (a) is,are respectively asThe modulation degree of (3);are respectively asThe background item of (a) is,are respectively asPhi is the wrapped phase. Note that: in the input fringe patternHowever, due to the influence of coupling and color imbalance, the acquired fringe pattern is causedThis is a problem of color not being normalized, as shown in fig. 2. FIG. 2a is a simulated color fringe plot with coupling and Gaussian error, and FIG. 2c is a plot of FIG. 2a with solid white lines corresponding to solid white linesThe stripe amplitudes of the three channels of the color stripe graph are inconsistent after the coupling sum and the noise are added, namely the colors are not normalized; fig. 2b is a simulated uniform intensity diagram with coupling and gaussian error, and fig. 2d is a cross section corresponding to the white solid line in fig. 2b, it can be seen that after noise is added, the intensity uniform diagram has burrs, and after coupling is added, the intensities of the three uniform diagrams are inconsistent. Wherein the coupling coefficient in simulation is set to K_rg＝0.1,K_rb＝0.05，K_gr＝0.23,K_gb＝0.12，K_br＝0.05,K_bg0.12. Fig. 5 a-5 c are images of three channels of uniform intensity of an actual acquired face model, and it can be seen from fig. 5d that the uniform image intensities of the three channels are not consistent due to the effects of color coupling and imbalance, which requires decoupling and color normalization in subsequent steps.

Step five: respectively carrying out single uniform intensity graph I acquired in the third step and the fourth step on the coupling intensity coefficient obtained in the second step^cuAnd a single color stripe pattern I^cDecoupling to obtain a decoupled uniform intensity map I^cudAnd decoupled color fringe pattern I^cdThe calculation formula is as follows:

for the collected uniform intensity chart I^cuDecoupling:

respectively, a uniform intensity map I^cuRed, green, blue channel images; respectively, decoupled uniform intensity map I^cudRed channel, green channel, blue channel,andthe meanings are consistent.

Firstly, toPerforming decoupling correction to obtainThen using the corrected intensityTo pairAre decoupled to obtainFinally using the corrected intensityAndto pairAre decoupled to obtain

For collectionColor stripe pattern I^cDecoupling:

respectively, collected color fringe patterns I^cGray scale fringe images of the red channel, the green channel and the blue channel;respectively a decoupled colour fringe pattern I^cdThe gray stripe patterns of the red channel, the green channel and the blue channel,are respectively asThe background item of (a) is,are respectively asThe modulation degree of (3);

firstly, toPerforming decoupling correction to obtainThen using the corrected intensityTo pairAre decoupled to obtainFinally using the corrected intensityAndto pairAre decoupled to obtainThe decoupling correction idea of the step is to use the corrected channel strength to perform decoupling correction on other channels, so that the decoupling effect is better.

Step six: decoupling uniform intensity graphs of the red channel, the green channel and the blue channel obtained in the fifth step Taking the blue channel as a reference, calculating the modulation ratio M of the uniform intensity map of the red channel and the green channel to the blue channel respectively₁,M₂As shown in the following formula:

wherein corr1, corr2, corr3 and corr4 are all modulation factors, and take the value of-5: 5, the use is carried outSo as to ensure that the normalization degree of the three gray scale fringe patterns is better; (x, y) are the coordinates of the pixels of the uniform intensity map,is thatThe intensity value of the pixel point with the coordinate of (x, y),is thatThe intensity value of the pixel point with the coordinate of (x, y),is thatIntensity value of pixel point with coordinate of (x, y), M₁(x, y) is M₁Value of coordinate (x, y), M₂(x, y) is M₂The coordinates are the values at (x, y).

Similarly, in this step, the modulation ratio may be obtained based on the red channel and the green channel. For example, when the red channel is taken as the reference, the modulation ratio M of the uniform intensity map of the green channel and the blue channel to the red channel is obtained₁,M₂As shown in the following formula:

when the green channel is taken as a reference, the modulation degree ratio M of the uniform intensity map of the red channel and the uniform intensity map of the blue channel to the green channel is obtained₁,M₂As shown in the following formula:

no matter which channel of red, green and blue is used as a reference to obtain the modulation ratio, the measurement result is not influenced.

Step seven: from the second step and the third step, the projected uniform map intensity I^puIntensity and color fringe pattern I^pIs equal, but due to problems of color coupling and imbalance, the actually acquired uniform image intensity I is caused^cuIntensity and color fringe pattern I^cThe background item intensities of (a) are not consistent. Obtaining a decoupled uniform intensity chart I through decoupling correction of the step five^cudIntensity and color fringe pattern I^cdThe background items of (2) are uniform in intensity, i.e.For the decoupled color stripe pattern I^cdGray scale fringe pattern of three channelsRemoving the background to obtain a color stripe image I after removing the background^deback，The calculation formula is as follows:

due to the fact thatThen:

wherein the content of the first and second substances,are each I^debackGrayscale fringe pattern of middle red, green, blue channel. The fringe pattern with coupling and gaussian noise in fig. 2a is subjected to decoupling, background removal and normalization processing to obtain the result of fig. 3, wherein fig. 3a is the cross section of the color fringe pattern after decoupling, the background of fig. 3a is removed by using the step to obtain fig. 3b, and it can be seen from fig. 3b that the fringe pattern after background removal is centered on 0, so that complete background removal is realized, and the effectiveness of the decoupling correction method is also shown. Fig. 6 is a decoupling, background removing and normalization result when the human face model is actually measured, wherein fig. 6a is a cross-sectional view of a decoupled fringe pattern, fig. 6b is a result of background removing of fig. 6a, and it can be seen that after background removing, the fringe pattern is a sine curve with 0 as a center line, that is, complete background removing is realized.

Step eight: taking the blue channel as a reference, and utilizing the modulation degree ratio M obtained in the step six₁,M₂And for the gray scale fringe pattern obtained in the step seven after the background is removedThe modulation degree normalization is performed as shown in the following formula:

wherein, I₁,I₂,I₃Is a three-gray-scale stripe image after color normalization. Fig. 3c shows the cross section of the simulated stripe image after the color normalization is finally realized, and it can be seen that the three gray stripe images realize good color normalization after the normalization in this step. Fig. 6c shows a cross-sectional view of a stripe image after color normalization finally obtained when a face model which is actually acquired is measured, and it can be seen that in actual measurement, color normalization of the stripe image is finally realized through the method of the present invention.

Note: the coupling intensity coefficient, the background term and the modulation ratio in the steps can also be obtained by projecting three red, green and blue uniform images on the object to be measured, and then decoupling, background removing and color normalization are carried out on one color fringe image.

Step nine: utilizing a three-step phase shift method to normalize the three gray scale fringe patterns I of the colors obtained in the step six₁,I₂,I₃The demodulation phase phi is solved. As shown in the following formula

Fig. 4a is a diagram of the demodulation phase of the simulation plane obtained by the method of the present invention, and fig. 4b is a diagram of the demodulation phase corresponding to the white solid line in fig. 4a, and it can be seen that the obtained phase amplitude is (-pi, pi). Since the demodulation precision is higher as the number of phase shift steps is larger in the phase shift demodulation, the demodulation phase error of the method is evaluated by taking the 12-step phase shift demodulation method as a reference. Fig. 4c is a cross-sectional view of the demodulation phases obtained by the demodulation of the method of the present invention and the 12-step phase shift method, and fig. 4d is a phase error obtained by subtracting the demodulation phases obtained by the two methods in fig. 4c, and it can be seen that the error in the simulation result of the method of the present invention is within 0.04 rad. Fig. 7a to 7d are results obtained by measuring the face model by using the method of the present invention, and the demodulation phase error is within 0.08rad, which is twice of the simulation result, because in the actual measurement, the environmental error and the system error are random and uncontrollable, the error will be larger than the simulation result, but the subsequent unwrapping will not be affected, and the errors of both are within the error range, thereby satisfying the requirement of high precision measurement.

Step ten: and D, unwrapping the demodulation phase obtained in the step nine by using a quality-guided unwrapping algorithm, reconstructing a three-dimensional shape, and covering the acquired uniform intensity map on the reconstructed three-dimensional map to obtain a real three-dimensional image with coexisting texture and color. Fig. 8a and 8b show the reconstructed three-dimensional shape and the images covered with colors, which show that the unwrapping phase of the object can be solved by a uniform image and a color fringe image in the method of the present invention, so as to realize the reconstruction of the three-dimensional shape of the object.

Referring to fig. 9, a dynamic three-dimensional topography measuring apparatus based on a single color stripe pattern comprises a color camera 1, a projector 2 and an upper computer 3, wherein the output end and the input end of the color camera 1 and the projector 2 are both connected with the upper computer and are used for receiving and transmitting instructions of the upper computer. The projector 2 is used for projecting a red uniform intensity image, a green uniform intensity image and a blue uniform intensity image to the calibration white board and projecting a color fringe image and a uniform intensity image to the object 4 to be measured; the color camera 1 is used for shooting a projected uniform intensity image for calibration and a projected image and a uniform intensity image of a color fringe image on a measured object, and transmitting the shot images to an upper computer, wherein a calculation program capable of running on the upper computer is stored on the upper computer, and when the processor executes the calculation program, the steps of the phase demodulation method are realized, the demodulation phase phi of the phase demodulation method is calculated according to the received images, and unwrapping is carried out according to the demodulation phase phi, the unwrapped phase is obtained, and the three-dimensional morphology is reconstructed.

The computer program may be partitioned into one or more modules/units that are stored in the memory and executed by the processor to implement the invention.

The upper computer can be a desktop computer, a notebook, a palm computer, a cloud server and other computing equipment. The upper computer can include, but is not limited to, a processor and a memory.

The processor may be a Central Processing Unit (CPU), other general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The method disclosed by the invention provides a new idea for measuring the dynamic three-dimensional morphology, and firstly, the system is pre-calibrated to obtain the coupling strength coefficient. Then, formal measurement is carried out, a uniform intensity image and a color fringe image are projected on an object to be measured in sequence, a color camera synchronously collects the two images, a computer carries out fast decoupling, background removal and color normalization on the collected images, three gray-scale phase-shift fringe images without background and color normalization are separated, and then a phase-shift method is adopted for demodulation, so that a demodulation phase is obtained. The method of the invention reduces the number of the stripe images required by the phase shift method from a plurality of stripe images to one stripe image, and extracts the required information by collecting a single frame image, so that the measurement precision is not influenced by the movement speed of the object, and the synchronous three-dimensional shape reconstruction of the moving object can be realized; meanwhile, the method extracts the color information of the object to be detected, enriches the appearance of the reconstructed object and more vividly shows the appearance of the three-dimensional object; for the problem of color imbalance of the color fringe image, the background of the uniform intensity image is extracted to obtain a modulation ratio, and then color normalization processing is carried out on the color fringe image, so that the efficiency problem of the traditional color normalization method is solved. In conclusion, through simulation and experimental verification, the method well considers the efficiency and the precision for the measurement of the moving object, and can realize the rapid and high-precision reconstruction of the three-dimensional shape of the dynamic object.

The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.