CN113012279B - Non-contact three-dimensional imaging measurement method and system and computer readable storage medium - Google Patents
Non-contact three-dimensional imaging measurement method and system and computer readable storage medium Download PDFInfo
- Publication number
- CN113012279B CN113012279B CN202110289260.5A CN202110289260A CN113012279B CN 113012279 B CN113012279 B CN 113012279B CN 202110289260 A CN202110289260 A CN 202110289260A CN 113012279 B CN113012279 B CN 113012279B
- Authority
- CN
- China
- Prior art keywords
- picture
- dimensional
- centroid
- water surface
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003384 imaging method Methods 0.000 title claims abstract description 16
- 238000000691 measurement method Methods 0.000 title claims abstract description 7
- 238000003860 storage Methods 0.000 title claims description 9
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 65
- 238000004364 calculation method Methods 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims description 41
- 238000005259 measurement Methods 0.000 claims description 23
- 230000009466 transformation Effects 0.000 claims description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 10
- 230000001788 irregular Effects 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 8
- 230000033001 locomotion Effects 0.000 claims description 7
- 238000003491 array Methods 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 5
- 230000003287 optical effect Effects 0.000 abstract description 13
- 230000008859 change Effects 0.000 abstract description 12
- 238000012545 processing Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- 239000007787 solid Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002706 hydrostatic effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30244—Camera pose
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to a non-contact three-dimensional imaging measurement method, which can accurately measure the three-dimensional change of a water surface in a large range, and can further eliminate various optical errors including lens distortion influence by using interpolation of multilayer still water surface reconstruction results as compensation of distortion errors; for one identification point, only the centroid is iteratively recalculated near the known position by using the corresponding result of the previous picture, and the whole picture does not need to be processed again; therefore, the calculation efficiency is greatly improved, more importantly, the centroid position of the identification point can be automatically tracked in the face of the dynamic change of the water wave surface, and therefore the position of the identification point in each camera is automatically matched.
Description
Technical Field
The invention relates to a non-contact three-dimensional imaging measurement method, a non-contact three-dimensional imaging measurement system and a computer readable storage medium.
Background
In the physical experiment research of hydrodynamics, often will measure the change of wave propagation in-process surface of water height, adopt the wave height appearance to measure surface of water height in traditional experiment wave pond. However, the wave height gauge has two disadvantages: one is that the wave height gauge is in contact with the water body, so that wave propagation is influenced to a certain extent; another is that the wave front height change can only be measured at one point, which is time consuming and laborious to perform in a large range and in an intensive measurement. If the non-contact multi-point measurement with a larger coverage area can be carried out, the height change of the free water surface during wave motion can be obtained more effectively, and more comprehensive data is provided for scientific research and production.
The prior art has the following problems:
firstly, in the prior art, the pictures taken by the camera at each moment are subjected to full-picture image analysis to obtain two-dimensional coordinates of the centroid of the identification point. Since the resolution of the picture is generally high, the amount of calculation for performing the image analysis of the full picture on each picture is large.
In the prior art, a perspective transformation algorithm is used for correcting the inclined light spot array into a rectangle. The algorithm of perspective transformation is to use the positions of four corners to perform graph transformation, so that the trapezoid is changed into a rectangle. However, when the local three-dimensional deformation of the waves is large, the inside of the light spot array cannot be regularly arranged after perspective transformation processing is carried out by utilizing the positions of the four vertex angles, and in order to solve the problem in the prior art, the number of the light spots cannot be excessively dense. Thus, when the local three-dimensional deformation of the waves is large, the light spot array after perspective transformation processing is irregular inside, but the light spots are few, and the light spot array can be sequenced according to a certain method. However, fewer spots will result in a reduced spatial resolution of the three-dimensional reconstruction technique, and a sufficiently dense spatial three-dimensional position measurement cannot be obtained.
And thirdly, when the measurement area is large, the distortion error of the camera lens is inevitably introduced, and the phenomena of large near objects and small far objects are generated. Although there is a lens distortion compensation coefficient in the calibration process, there are a variety of optical errors that may affect the measurement results in the actual measurement environment, and these optical errors including lens distortion are difficult to be handled all at once. If no other compensation correction is carried out, the high-precision measurement required by the hydrodynamic experiment cannot be met. Therefore, the three-dimensional reconstruction result of the wave surface in the prior art is difficult to be applied to scientific researches with higher precision requirements.
In conclusion, the prior art cannot perform rapid high-precision measurement on the water surface with large local deformation in a large measurement range.
Disclosure of Invention
Aiming at the problem that the prior art can not solve the problem of rapid high-precision measurement of the water surface with large local deformation in a large measurement range, the invention adopts the technical scheme that:
a non-contact three-dimensional imaging measurement method, the method comprising:
step 1, interpolating a reconstruction result of a multilayer still water plane to obtain a compensation value of lens distortion;
and 2, automatically inheriting the positions and the sequence of the identification points in the two adjacent pictures to realize the matching of the irregular light spot arrays, and obtaining the spatial three-dimensional coordinates of the identification points according to the compensation values.
Wherein, the step 1 specifically comprises:
step 1.1, calibrating a camera to obtain a mapping relation matrix R between a space three-dimensional coordinate and a two-dimensional pixel coordinate in a photo;
step 1.2, determining characteristic identification points on a static water surface;
step 1.3, calculating the centroid of the identification point based on the mapping relation matrix R, carrying out perspective transformation, changing the trapezoid image formed by the original light spot array into a rectangle, and completing automatic matching of the identification point;
step 1.4, solving three-dimensional coordinates of the identification points;
and step 1.5, subtracting the average value of the Z-direction height of all the identification points obtained in step 1.4, carrying out space surface fitting, and then calculating the error compensation value of the Z direction at the position.
The method according to claim 2, wherein the step 1 further comprises:
step 1.6, replacing the height of the water surface to obtain an error compensation value corresponding to each height;
and step 1.7, after the error compensation value is obtained, interpolating each space position (X, Y) in the Z direction by utilizing multiple planes.
Wherein, the step 1.5 specifically comprises:
deducting the average value of the Z-direction height of all the identification points obtained in the step 1.4, fitting based on a fitting formula, and fitting to obtain coefficients A-D to be determined; when the waiting coefficients A-D are determined;
the fitting formula shown is:
and solving a Z-direction value based on a fitting formula according to each point (X, Y) in a Z-direction plane of the space where the still water surface is positioned, wherein the value is an error compensation value of the lens distortion in the current plane.
Wherein the step 2 comprises:
step 2.1, acquiring a dynamic wave surface photo;
step 2.2, calculating the centroid of the first wave surface picture and carrying out perspective transformation;
2.3, searching the centroid of the next wave-front photo by adopting an iterative algorithm;
step 2.4, solving the three-dimensional coordinates of the identification points after automatic matching;
step 2.5, deducting an error compensation value;
step 2.6, repeat step repeat 2.3, 2.4 and 2.5 to process the picture at the next moment.
Wherein the step 2.3 comprises:
starting from the second picture taken at 2.1, when the picture at the new moment is processed, the centroid coordinates of the picture at the new moment, which are precisely determined for each identification point in the previous picture, are taken as the centroid coordinates of the picture at the new moment.
Wherein the step 2.4 comprises:
and solving the space three-dimensional coordinate of each identification point in the photo by combining the calibration result according to the centroid position and the automatically matched identification point.
The invention also provides a non-contact three-dimensional imaging measurement system, comprising:
the compensation value calculation module is used for interpolating the reconstruction result of the multilayer still water plane to obtain a compensation value of lens distortion;
and the coordinate calculation module is used for automatically inheriting the positions and the sequence of the identification points in the two adjacent pictures to realize the matching of the irregular light spot arrays and obtain the spatial three-dimensional coordinates of the identification points according to the compensation values.
The invention also provides a non-contact three-dimensional imaging measurement system, comprising:
a memory for storing a computer program,
a processor for invoking a computer program in memory to implement the method.
The invention also provides a computer-readable storage medium, which is characterized in that the storage medium stores a computer program; when invoked by a processor in a computing device, causes the computing device to perform the method.
The invention has the beneficial effects that:
1. the three-dimensional change of the water surface can be accurately measured in a large range. By utilizing the interpolation of the multi-layer still water surface reconstruction result as the compensation of the distortion error, various optical errors including the lens distortion influence can be further eliminated.
2. Can quickly process the larger local deformation of the water surface. For an identified point, only the centroid needs to be iteratively recalculated around the known position using the corresponding result of the previous picture, without having to reprocess the entire picture. Therefore, the calculation efficiency is greatly improved, and more importantly, the centroid position of the identification point can be automatically tracked in the face of the dynamic change of the water wave surface (when the position of the identification point is greatly deviated), so that the position of the identification point in each camera is automatically matched.
Drawings
The invention is further illustrated with reference to the following figures and examples.
FIG. 1 is a flow chart of a method of a preferred embodiment of the present invention;
Detailed Description
The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic views illustrating only the basic structure of the present invention in a schematic manner, and thus show only the constitution related to the present invention.
To facilitate the understanding of the invention, the terms involved are described:
hydrodynamics: the disciplines of water and other liquids' law of motion and its interaction with boundaries (e.g., structures) are studied.
Wave height instrument: an instrument for measuring and recording the height of waves at a given location. After the water surface is inserted into water, the feedback voltage values are different when the water surface heights are different.
Three-dimensional reconstruction of images: generally, at least 2 cameras are used for collecting the graphic features of the surface of an observed object, once the identification points are extracted from the graphic features, the three-dimensional position of each identification point on the surface of the object can be reconstructed in a computer by adopting an image processing algorithm, and therefore the three-dimensional shape of the object is obtained.
Marking points: this means that the camera captures the center position of the graphic features on the surface of the object when the image is reconstructed three-dimensionally. For the image processing algorithm, the graphic features are used to identify the surface of the object, so the center position of the graphic features is called an identification point.
Calibrating a camera: in image measurement processes and machine vision applications, in order to determine the correlation between the three-dimensional geometric position of a certain point on the surface of an object in space and the corresponding point in the image, a geometric model of camera imaging must be established, and the parameters of the geometric model are the parameters of the camera. The process of establishing the model to obtain the parameters is camera calibration.
And (3) centroid: the geometric center of the figure.
Image post-processing: the picture taken by the camera is digitized to obtain the desired graphical information. The conventional method comprises the following steps: noise reduction, binarization, opening and closing operation, centroid calculation and the like.
Perspective transformation: the method is characterized in that a perspective center, an image point and a target point are collinear, a bearing surface (perspective surface) rotates a certain angle around a trace line (perspective axis) according to a perspective rotation law, an original projection light beam is damaged, and the transformation of a projection geometric figure on the bearing surface can be still kept unchanged. Mathematically, it can be simply understood that: one graph is transformed into another graph according to a given formula.
Lens distortion: the perspective distortion inherent to the optical lens is a general term that makes it impossible to maintain the planar characteristic of the object image. For example, when an object is shot by a wide-angle lens, there is a phenomenon that "a near object is large and a far object is small", which is distortion caused by lens distortion.
Fitting: a series of points on a plane are connected by a smooth curve (surface).
Interpolation: the continuous function is interpolated on the basis of the discrete data such that the continuous curve passes through all given discrete data points. Here to fill the gaps between pixels in the image transformation.
As shown in fig. 1, a non-contact three-dimensional imaging measurement method of the present invention includes:
step 1, interpolating a reconstruction result of a multilayer still water plane to obtain a compensation value of lens distortion;
and 2, automatically inheriting the positions and the sequence of the identification points in the two adjacent pictures to realize the matching of the irregular light spot arrays, and obtaining the spatial three-dimensional coordinates of the identification points according to the compensation values.
The first step mainly aims to reduce optical errors by interpolating reconstruction results of a multilayer still water plane to obtain a compensation value of lens distortion, and the specific method comprises the following steps:
1.1 calibration. And (3) fixing at least two high-speed cameras above the wave pool to obtain a mapping relation matrix R between the space three-dimensional coordinates in the formula (1) and the two-dimensional pixel coordinates in the photo. A Zhangzhengyou calibration method is selected in the wave pool, so that calibration can be realized conveniently and quickly.
In order to carry out non-contact multi-point measurement with larger coverage area, an image three-dimensional reconstruction technology is adopted, namely, a plurality of cameras shoot the water surface from different angles, and the three-dimensional shape of the water surface can be reconstructed by carrying out image processing on the pattern with certain characteristic marks in the water surface picture.
The core technical principle of the image three-dimensional technology is shown in formula (1).
Wherein n represents the nth camera (n is not less than 2). The first step is "calibration". And calibrating the cameras to obtain a mapping relation matrix R of two-dimensional pixel coordinates (X, Y) and three-dimensional space coordinates (X, Y, Z) in the picture shot by each camera. The second step is "identify". After shooting a space object by a camera, determining an ith space identification point (X) through image processingi，Yi，Zi) Corresponding two-dimensional pixel coordinates (x) in all n camera imagesin，yin). The third step is "match". For each identification point, it is necessary to determine which of the many identification points of each camera takes a picture corresponds to. Once determined, the formula (1) can be solved in a row to obtain the three-dimensional coordinates of the ith space identification point (Xi，Yi，Zi)。
In the prior art, the first step of calibration is a normalized process, and has no special points. The performance effect of three-dimensional reconstruction is mainly determined by the second step "recognition" and the third step "matching". The stereo reconstruction of image three-dimensional technology for solid surface identification points has been widely used in many fields (such as widely used binocular stereo reconstruction technology), and the fundamental reason is that the graphic identification points of the solid surface are easy to capture (i.e. the two-dimensional pixel coordinates of the solid surface in all photos are easy to determine). However, in the field of hydrodynamics, the surface movement causes the surface identification point to have a strong dynamic course. The irregular deformation and movement of the identification points makes it difficult to efficiently capture and match them, making conventional three-dimensional techniques of images difficult to apply. Moreover, when the shooting range and the surface deformation are large, the wide-angle camera matched with the camera can introduce lens distortion which cannot be completely eliminated, so that the target at the center of the lens is closer, and the targets at the periphery of the lens are farther. Although the lens distortion error is eliminated to some extent during the "calibration" process. However, when the water wave surface is deformed greatly, the lens error will still cause the measurement accuracy to be reduced.
1.2 shoot still water surface. All cameras shoot the water level in the water surface static state synchronously. Note that, at this time, a characteristic mark point is required on the still water surface, and can be obtained by a laser or a projector transmission method.
1.3 calculate the centroid and perform perspective transformation. Obtaining the two-dimensional coordinates of the centroid of each identification point in the picture, namely (x) in formula (1), by using a traditional image post-processing algorithmin，yin). At the moment, the water surface is static, and the arrangement of the identification points is regular, so that the pictures can be subjected to perspective transformation by using the formulas (2) and (3), the trapezoid images formed by the original light spot arrays are changed into rectangles, and the automatic matching of the identification points is completed.
In the formula (2), xn，ynIs the two-dimensional coordinates of the four vertexes of the light spot array in the original picture,
four vertexes of the rectangle are designated after transformation, and undetermined global transformation coefficients a-h can be obtained. Substituting a-h into formula (3) to transform the original quadrilateral light spot array into a rectangular array. In the formula (3), xi，yiIs the two-dimensional coordinates of any pixel of the original photograph,
for the perspective transformed pixel coordinates, w is an arbitrary scale factor used to enlarge or reduce the image.
And 1.4, solving the three-dimensional coordinates of the identification points. Substituting the calibration result of 1.1 and the matching result of 1.3 into the formula (1) to obtain a solution, and obtaining the three-dimensional space position (Xi, Yi, Zi) of each identification point
1.5 fitting the space surface. Since one of the main sources of optical errors is that the distortion caused by the lens can cause the near-far situation in the depth of field direction (generally, the Z direction), and the shape of the camera negative is rectangular, the error value caused by the lens distortion can be approximately regarded as an elliptic paraboloid in space. I.e. the three-dimensional spatial position (Xi, Yi, Zi) of each identified point solved in 1.4, will not perfectly lie in the same plane, but will be distributed on an elliptic paraboloid similar to the form of equation (4). And subtracting the average value of the Z-direction height of all the identification points obtained in the step 1.4, and substituting all the identification points near a zero reference surface in the Z direction into a formula (4) for fitting to obtain undetermined coefficients A-D. After the coefficients A to D to be determined are determined, for each point (X, Y) in the Z-direction plane of the space where the current still water surface is located, the Z-direction value at the position can be obtained through a formula (4), and the value is an error compensation value of the lens distortion in the current plane. In other words, in practical use, after three-dimensional reconstruction obtains the spatial coordinates of one identification point, if the Z-direction height of the coordinates is equal to the Z-direction height of the current still water surface, an elliptic paraboloid equation should be fitted according to this step, and the corresponding Z-direction compensation value at X, Y coordinates of the elliptic paraboloid equation is subtracted, so that the optical error mainly caused by lens distortion is reduced.
1.6 changing the water level. Section 1.5 is only to compensate for optical errors in the reconstruction of one still surface. When the surface of water fluctuates greatly, need carry out optical error's compensation to a plurality of still water levels in the height range to improve the reconstruction precision when the surface of water fluctuates greatly.
1.7 multiple hydrostatic interpolation. To further reduce optical errors, a multi-plane interpolation correction method is used. Because the actually reconstructed Z-direction height does not exactly fall in the water surface height selected in 1.6, after obtaining a plurality of groups of error compensation values of the water surface, interpolating each space position (X, Y) in the Z direction by utilizing a plurality of planes, and obtaining the error compensation value at any Z-direction height.
And after the first step is completed, the reconstruction of the fluctuating three-dimensional water surface can be started. The main method of the second step is that the positions and the sequence of the identification points are automatically inherited in two adjacent pictures, so that the matching of the irregular light spot array can be quickly realized, and the space three-dimensional coordinates of the identification points are obtained. The method comprises the following concrete steps:
2.1 shooting dynamic wave surface. After the first step is completed, the dynamic wave surface with wave motion is shot by a high-speed camera without changing the position of the camera. Note that at least it is ensured that in the initial state, one picture is taken from the time the water surface is still, i.e. the state of fig. 1. The choice of using a high speed camera is 2.3 in preparation for ensuring that the dynamic course of the water wave surface does not change significantly in successive photographs.
2.2 centroid is calculated for the first sheet and perspective transformed. As in 1.3, but only one of the still water surfaces was calculated in all pictures taken at 2.1. For this sheet, the array of identification points is transformed into a rectangle by a perspective transformation. After this sheet, the array of marker points will gradually be out of alignment with the perspective transformation due to the water surface fluctuation.
2.3 search for the centroid for the next one using an iterative algorithm. Starting from the second picture taken at 2.1, when processing a new picture at a time, there is no need to search through the whole picture to retrieve the position of each identified point, but an iterative algorithm is used. The method is realized by taking the centroid coordinate of each identification point in the previous picture which is accurately determined as an estimation value, and only needing to search again by a graphical processing method within a range which is larger than the size of each identification point by taking the estimation value as the center in the new picture. The iterative algorithm can be used for processing the dynamic process because a high-speed camera is used, so that the position of the identification point does not move greatly in two continuous photos, and the shape of the identification point does not change obviously.
And 2.4, solving the three-dimensional coordinates of the identification points after automatic matching. The iterative algorithm of 2.3 not only simplifies the searching and identifying process of the centroid of the identification point, but also can realize the matching process without perspective change. Since the position of each identification point is iteratively calculated from the position of the last corresponding point, the recording order of the identification points in the computer is inherited. This allows each identification point in the new picture to automatically achieve a match in each camera picture. Therefore, even if the identification point array is locally shifted due to the dynamic motion of the waves, the use of the centroid iteration algorithm of 2.3 can still ensure that the identification points are automatically matched. And substituting the centroid position and the automatically matched identification points into a formula (1) in combination with the calibration result, so that the spatial three-dimensional coordinate of each identification point in the new sheet can be obtained through solving.
2.5 deduct the error compensation value. For the spatial three-dimensional coordinate value of each identification point obtained in 2.4, the error compensation value at the corresponding position is found in 1.7, and the final measurement value can be obtained after deduction.
2.6 processing the picture at the next moment. Repeat 2.3, 2.4 and 2.5. All processing is completed until the last picture taken in 2.1 is processed.
The invention adopts an iterative identification method to track and capture the dynamic position of each identification point. Meanwhile, before three-dimensional wave surface reconstruction, an elliptic paraboloid equation is used for fitting a reconstruction result of the still water surface, and interpolation of multiple layers of still water surfaces is used as compensation of optical errors. Therefore, a three-dimensional reconstruction algorithm can be rapidly and accurately realized, the three-dimensional position of the water surface identification point is obtained by utilizing the formula (1), and large-range and high-precision three-dimensional wave surface data are provided for a hydrodynamic experiment.
The invention also provides a non-contact three-dimensional imaging measurement system, comprising:
the compensation value calculation module is used for interpolating the reconstruction result of the multilayer still water plane to obtain a compensation value of lens distortion;
and the coordinate calculation module is used for automatically inheriting the positions and the sequence of the identification points in the two adjacent pictures to realize the matching of the irregular light spot arrays and obtain the spatial three-dimensional coordinates of the identification points according to the compensation values.
The invention also provides a non-contact three-dimensional imaging measurement system, comprising:
a memory for storing a computer program,
a processor for invoking a computer program in memory to implement the method.
The invention also provides a computer-readable storage medium, which is characterized in that the storage medium stores a computer program; when invoked by a processor in a computing device, causes the computing device to perform the method.
The invention has the beneficial effects that:
1. the three-dimensional change of the water surface can be accurately measured in a large range. By utilizing the interpolation of the multi-layer still water surface reconstruction result as the compensation of the distortion error, various optical errors including the lens distortion influence can be further eliminated.
2. Can quickly process the larger local deformation of the water surface. For an identified point, only the centroid needs to be iteratively recalculated around the known position using the corresponding result of the previous picture, without having to reprocess the entire picture. Therefore, the calculation efficiency is greatly improved, more importantly, the centroid position of the identification point can be automatically tracked in the face of the dynamic change of the water wave surface, and therefore the position of the identification point in each camera is automatically matched.
In light of the foregoing description of the preferred embodiment of the present invention, many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The technical scope of the present invention is not limited to the content of the specification, and must be determined according to the scope of the claims.
Claims (10)
1. A non-contact three-dimensional imaging measurement method, characterized in that the method comprises:
step 1, interpolating a reconstruction result of a multilayer still water surface to obtain a compensation value of lens distortion;
step 2, automatically inheriting the positions and the sequence of the identification points in two adjacent pictures to realize the matching of the irregular light spot array, and calculating the space three-dimensional coordinates of the identification points according to the compensation values;
wherein, the step 1 specifically comprises:
step 1.1, calibrating a camera to obtain a mapping relation matrix R between a space three-dimensional coordinate and a two-dimensional pixel coordinate in a photo;
step 1.2, determining characteristic identification points on a still water surface, wherein the still water surface is a horizontal plane of a water surface in a static state;
step 1.3, calculating the centroid of the identification point based on the mapping relation matrix R, carrying out perspective transformation, changing the trapezoid image formed by the original light spot array into a rectangle, and completing automatic matching of the identification point;
step 1.4, solving three-dimensional coordinates of the identification points;
step 1.5, subtracting the average value of the Z-direction height of all the identification points in the step 1.4, carrying out space surface fitting, and then calculating the error compensation value of the Z direction at the position;
step 1.6, replacing the height of the water surface to obtain an error compensation value corresponding to each height;
and step 1.7, after the error compensation value is obtained, interpolating each space position (X, Y) in the Z direction by utilizing multiple planes.
2. The method according to claim 1, wherein step 1.5 specifically comprises:
deducting the average value of the Z-direction height of all the identification points in the step 1.4, fitting based on a fitting formula, and fitting to obtain coefficients A-D to be determined; wherein,
the fitting formula is:
and after the coefficients A to D to be determined are determined, according to each point (X, Y) in a Z-direction plane of a space where the still water surface is located, a Z-direction value is calculated based on a fitting formula, and the Z-direction value is determined as an error compensation value of the lens distortion in the current plane.
3. The method of claim 1 or 2, wherein the step 2 comprises:
step 2.1, acquiring a dynamic wave surface photo;
step 2.2, calculating the centroid of the first wave surface picture and carrying out perspective transformation;
2.3, searching the centroid of the next wave-front photo by adopting an iterative algorithm;
step 2.4, solving the three-dimensional coordinates of the identification points after automatic matching;
step 2.5, deducting an error compensation value;
step 2.6, repeat step 2.3, step 2.4 and step 2.5 to process the picture at the next moment.
4. A method according to claim 3, wherein said step 2.3 comprises:
starting from the second picture taken in step 2.1, when the picture at the new time is processed, the centroid coordinates of the picture at the new time, which are precisely determined for each identification point in the previous picture, are taken as the centroid coordinates of the picture at the new time.
5. The method of claim 4, wherein the step 2.4 comprises:
and solving the space three-dimensional coordinate of each identification point in the photo by combining the calibration result according to the centroid coordinate and the automatically matched identification point.
6. The method according to claim 3, characterized in that said step 2.1 comprises in particular:
and shooting the dynamic wave surface with wave motion by using a high-speed camera to obtain a dynamic wave surface picture.
7. A non-contact three-dimensional imaging measurement system, characterized in that the system comprises:
the compensation value calculation module is used for interpolating the reconstruction result of the multilayer still water surface to obtain a compensation value of lens distortion;
the coordinate calculation module is used for automatically inheriting the positions and the sequence of the identification points in the two adjacent pictures to realize the matching of the irregular light spot arrays and obtain the spatial three-dimensional coordinates of the identification points according to the compensation values;
wherein the compensation value calculation module is specifically configured to perform:
calibrating a camera, and acquiring a mapping relation matrix R between a space three-dimensional coordinate and a two-dimensional pixel coordinate in a photo;
determining characteristic identification points on a still water surface, wherein the still water surface is a horizontal plane of a water surface in a static state;
calculating the centroid of the identification point based on the mapping relation matrix R, carrying out perspective transformation, changing the trapezoid image formed by the original light spot array into a rectangle, and completing automatic matching of the identification point;
solving the three-dimensional coordinates of the identification points;
deducting the average value of the Z-direction height of all the solved identification points, performing space surface fitting, and then solving the Z-direction error compensation value at the position;
replacing the height of the water surface, and acquiring an error compensation value corresponding to each height;
after the error compensation values are obtained, interpolation is performed for each spatial position (X, Y) in the Z direction using multiple planes.
8. The non-contact three-dimensional imaging measurement system of claim 7, wherein the coordinate calculation module is specifically configured to:
acquiring a dynamic wave surface photo;
calculating the centroid of the first wave surface picture and performing perspective transformation;
searching the centroid of the next wave-surface picture by adopting an iterative algorithm;
solving the three-dimensional coordinates of the identification points after automatic matching;
deducting an error compensation value;
and repeatedly executing the steps of searching the centroid of the next wave surface picture by adopting an iterative algorithm, automatically matching, solving the three-dimensional coordinates of the identification point and deducting the error compensation value so as to process the picture at the next moment.
9. A non-contact three-dimensional imaging measurement system, characterized in that the system comprises:
a memory for storing a computer program,
a processor for invoking a computer program in a memory to implement the method of any of claims 1-6.
10. A computer-readable storage medium, characterized in that the storage medium has stored therein a computer program; the computer program, when invoked by a processor in a computing device, causes the computing device to perform the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110289260.5A CN113012279B (en) | 2021-03-18 | 2021-03-18 | Non-contact three-dimensional imaging measurement method and system and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110289260.5A CN113012279B (en) | 2021-03-18 | 2021-03-18 | Non-contact three-dimensional imaging measurement method and system and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113012279A CN113012279A (en) | 2021-06-22 |
| CN113012279B true CN113012279B (en) | 2022-02-11 |
Family
ID=76409523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110289260.5A Active CN113012279B (en) | 2021-03-18 | 2021-03-18 | Non-contact three-dimensional imaging measurement method and system and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113012279B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114993613B (en) * | 2022-05-27 | 2023-01-24 | 上海交通大学 | Test method and test system for structure water entering test |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109345461A (en) * | 2018-09-30 | 2019-02-15 | 中国科学院长春光学精密机械与物理研究所 | A kind of image distortion correction method, apparatus, equipment and storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013113696A (en) * | 2011-11-29 | 2013-06-10 | Mitsubishi Electric Corp | Displacement measuring method and displacement measuring apparatus |
| CN104482921B (en) * | 2014-12-23 | 2016-09-07 | 河北工业大学 | A kind of measuring method of waterborne target |
| CN108957447B (en) * | 2018-04-16 | 2019-11-08 | 长江水利委员会水文局长江下游水文水资源勘测局 | A kind of ship base radar water boundaries method for automatic measurement |
| CN111006743B (en) * | 2019-12-05 | 2021-07-02 | 水利部南京水利水文自动化研究所 | Method for eliminating interference in water level measurement based on plane radar water level gauge |
| CN111708038B (en) * | 2020-06-23 | 2023-06-27 | 上海埃威航空电子有限公司 | Unmanned ship laser radar point cloud data correction method based on attitude sensor and GNSS |
| CN112488955B (en) * | 2020-12-08 | 2023-07-14 | 大连海事大学 | Underwater image restoration method based on wavelength compensation |
-
2021
- 2021-03-18 CN CN202110289260.5A patent/CN113012279B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109345461A (en) * | 2018-09-30 | 2019-02-15 | 中国科学院长春光学精密机械与物理研究所 | A kind of image distortion correction method, apparatus, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113012279A (en) | 2021-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112066879B (en) | Air floatation motion simulator pose measuring device and method based on computer vision | |
| CN110880185B (en) | High-precision dynamic real-time 360-degree all-dimensional point cloud acquisition method based on fringe projection | |
| CN106408609B (en) | A kind of parallel institution end movement position and posture detection method based on binocular vision | |
| US9928595B2 (en) | Devices, systems, and methods for high-resolution multi-view camera calibration | |
| CN111192235B (en) | Image measurement method based on monocular vision model and perspective transformation | |
| CN112465912B (en) | Stereo camera calibration method and device | |
| CN111263142B (en) | Method, device, equipment and medium for testing optical anti-shake of camera module | |
| Douxchamps et al. | High-accuracy and robust localization of large control markers for geometric camera calibration | |
| WO2018201677A1 (en) | Bundle adjustment-based calibration method and device for telecentric lens-containing three-dimensional imaging system | |
| CN112184811B (en) | Monocular space structured light system structure calibration method and device | |
| CN109325927B (en) | Image brightness compensation method for industrial camera photogrammetry | |
| CN104697463B (en) | The blanking feature constraint scaling method and device of a kind of binocular vision sensor | |
| CN113920206B (en) | Calibration method of perspective tilt-shift camera | |
| CN109974618A (en) | The overall calibration method of multisensor vision measurement system | |
| JP6061770B2 (en) | Camera posture estimation apparatus and program thereof | |
| Su et al. | Auto-calibration and real-time external parameter correction for stereo digital image correlation | |
| CN109887002A (en) | Image feature point matching method and device, computer equipment and storage medium | |
| WO2020173194A1 (en) | Image feature point tracking method and apparatus, image feature point matching method and apparatus, and coordinate obtaining method and apparatus | |
| CN113012279B (en) | Non-contact three-dimensional imaging measurement method and system and computer readable storage medium | |
| CN114170321A (en) | Camera self-calibration method and system based on distance measurement | |
| CN113012143A (en) | Test piece quality detection method based on two-dimensional digital image correlation method | |
| CN109859313B (en) | 3D point cloud data acquisition method and device, and 3D data generation method and system | |
| CN110686593A (en) | Method for measuring relative position relation of image sensors in spliced focal plane | |
| WO2023040095A1 (en) | Camera calibration method and apparatus, electronic device, and storage medium | |
| CN116051634A (en) | Visual positioning method, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |