Measurement system and method for detecting bending degree of steel plate based on projection fringes

Abstract

The invention relates to a measurement system and method for detecting the bending degree of a steel plate based on projection fringes, and belongs to the field of mechanical vision measurement. The system comprises an industrial camera which is used for collecting projection stripes on the surface of a steel plate to be detected and transmitting collected signals to a mechanical vision image processing unit; the stripe projector projects the single-density stripes to the surface of the steel plate to be measured through the spectroscope; the stripe projector comprises an LED light source and a single-density stripe pattern, the LED light source irradiates a single-density stripe pattern sheet to project single-density stripes with a preset period to the surface of the steel plate to be detected, and the size, the position and the stripe style of the projected stripe pattern can be adjusted; and the mechanical vision processing unit is used for analyzing and processing the acquired signals to obtain the bending degree information of the steel plate. A transportation system and position sensor combined structure and a mechanical vision projection stripe and image processing system combined structure are adopted, steel plate detection automation can be achieved, and high-precision detection of the bending degree of the steel plate is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of machine vision measurement, and particularly relates to a measurement system and method for detecting the bending degree of a steel plate based on projection stripes. Background Art

[0002] One of the important signs of the development level of the national steel industry is the quality level of steel plates. At present, steel plates are widely used in fields such as construction, automobiles, ships, aerospace, etc. However, with the rapid development of these fields, the requirements for the surface quality of steel plates are getting higher and higher. Therefore, it is of great significance to monitor the surface quality of steel plates.

[0003] At present, in the direct measurement method, which is a commonly used method for detecting the bending degree of steel plates, manual measurement is required and the accuracy is relatively low; while for the interference measurement method, the stylus measurement method, and the laser scanning method, although the accuracy is improved compared with the direct measurement method, the cost is too high. Summary of the Invention

[0004] The purpose of the present invention is to provide a measurement system and method for detecting the bending degree of a steel plate based on projection stripes, which can greatly reduce the cost and achieve high-efficiency and high-precision measurement of the bending degree of the steel plate.

[0005] To achieve the above purpose, the technical solution of the present invention is: A measurement system for detecting the bending degree of a steel plate based on projection stripes, comprising:

[0006] An industrial camera, which collects the projection stripes on the surface of the steel plate to be measured and transmits the collected signal to the machine vision image processing unit;

[0007] A stripe projector, which projects single-density stripes onto the surface of the steel plate to be measured through a beam splitter; the stripe projector includes an LED light source and a single-density stripe pattern. The LED light source irradiates on the single-density stripe pattern thin sheet to project single-density stripes with a predetermined period onto the surface of the steel plate to be measured, and the size, position, and stripe pattern of the projection stripe pattern can all be adjusted;

[0008] A machine vision processing unit, which analyzes and processes the collected signal to obtain the information on the bending degree of the steel plate.

[0009] In an embodiment of the present invention, it further includes:

[0010] A transportation system, which transports the steel plate to be measured;

[0011] A position sensor, which detects the position of the steel plate to be measured to send a signal to the transportation system to control whether the transportation system stops or not.

[0012] In an embodiment of the present invention, the machine vision image processing unit includes an image processing system and a data line. The image processing system processes the captured projection fringe information using image processing algorithms to obtain the bending degree information of the steel plate to be measured, and the data line is used to transmit the acquisition signal.

[0013] In an embodiment of the present invention, the image processing system processes the captured projection fringe information using image processing algorithms to obtain the bending degree information of the steel plate to be measured, and the specific implementation is as follows:

[0014] Select the stripe intensity signal of the middle row of the stripe image in the theoretically collected state as the reference stripe intensity;

[0015] Successively and at intervals, calculate the correlation coefficient between the stripe intensities of each row of the steel plate to be measured and the reference stripe intensity to obtain each correlation coefficient curve. The mathematical relationship of the correlation coefficient is:

[0016]

[0017] Among them, ρ(f r , f j ) is the correlation coefficient value, f r and f j respectively represent the reference stripe intensity and the stripe intensity of the j-th row. cov(f r , f j ) is the covariance of f r and f j , σ fr and respectively represent the standard deviations of f r and f j , and E is the variance;

[0018] Iteratively calculate the power of each row of the correlation coefficient curve multiple times and then perform normalization processing to be between 0 and 1;

[0019] Integrate each row of the correlation coefficient curve into a correlation coefficient curve sequence, and perform a preliminary qualitative analysis to judge whether there is bending; perform Fourier transform on each row of pixels of the stripe projection image in the theoretically collected state and the stripe projection image on the surface of the steel plate to be measured to obtain the density information of each row, and then calculate the maximum displacement of the steel plate to be measured in the X direction according to the relationship between the single-density stripe density and the displacement, and obtain the bending degree of the steel plate to be measured according to the bending degree calculation formula.

[0020] In an embodiment of the present invention, the beam splitter can change the projection stripe direction of the stripe projector so that the direction of the projection stripe projected onto the steel plate to be measured coincides with the center position of the industrial camera shooting, ensuring the accuracy of the industrial camera to obtain the projection stripe.

[0021] In an embodiment of the present invention, the projection stripe can be projected horizontally or vertically onto the surface of the steel plate to be measured.

[0022] The present invention also provides a measurement method for detecting the bending degree of a steel plate based on projection stripes by using the system described above, including the following steps:

[0023] Step S1: Set the parameters of the industrial camera, and adjust the imaging range of the industrial camera to just completely capture the surface of the steel plate to be measured; adjust the position of the single-density stripe film in the stripe projector, and make the single-density stripe clearly projected on the measurement position of the steel plate to be measured through the beam splitter;

[0024] Step S2: Start the transportation system. After the transportation system transports the steel plate to be measured to the position where the position sensor detects the steel plate to be measured and sends a signal to the transportation system, the transportation system stops;

[0025] Step S3: The industrial camera collects the projection stripes on the surface of the steel plate to be measured, and transmits them to the machine vision processing unit through the data line. The captured projection stripe information is processed by using the image processing algorithm to obtain the bending degree information of the steel plate to be measured;

[0026] Step S4: After the machine vision processing unit obtains the information, it sends an instruction to the transportation system, and the transportation system operates normally.

[0027] In an embodiment of the present invention, the specific content of step S3 is as follows:

[0028] Step S31: Select the stripe intensity signal of the middle row of the stripe image in the theoretically collected state as the reference stripe intensity;

[0029] Step S32: Calculate the correlation coefficients by successively and intermittently selecting the stripe intensities of each row of the steel plate to be measured and the reference stripe intensity, and obtain each correlation coefficient curve. The mathematical relationship of the correlation coefficient is:

[0030]

[0031] Among them, ρ(f r, f j ) is the correlation coefficient value, f r and f j respectively represent the reference stripe intensity and the stripe intensity of the jth row, cov(f r , f j ) is the covariance of f r and f j , σ fr and respectively represent the standard deviations of f r and f j , and E is the variance of the data.

[0032] Step S33: Iteratively calculate the power of each row of the correlation coefficient curve multiple times and then perform normalization processing to be between 0 and 1;

[0033] Step S34: Integrate the correlation coefficient curves of each row into a sequence of correlation coefficient curves, and conduct a preliminary qualitative analysis to determine whether there is any bending. Perform Fourier transform on each row of pixels of the fringe projection images collected under the theoretical state and the fringe projection images on the surface of the steel plate to be measured to obtain the density information of each row. Then, according to the relationship between the single-density fringe density and displacement, calculate the maximum displacement of the steel plate to be measured in the X direction, and obtain the degree of bending of the steel plate to be measured according to the bending degree calculation formula.

[0034] In an embodiment of the present invention, the relationship between the single-density fringe density and displacement is:

[0035]

[0036] where, ΔX(t) is the maximum displacement in the X direction of the time-domain signal on the detection surface of the bent steel plate, f is the focal length of the lens, M a is the imaging scale factor, d t is the fringe density of the single-density fringe at the point with the maximum deviation of the bent steel plate on the image sensor, d 0 is the reference fringe density of the single-density fringe on the non-bent steel plate on the image sensor.

[0037] In an embodiment of the present invention, the bending degree calculation formula is:

[0038]

[0039] where, K is the bending degree, Δ is the maximum deviation, and L is the length of the steel plate to be detected.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention adopts the combined structure of a transportation system and a position sensor, which can realize the automation of steel plate detection and save labor costs.

[0042] The present invention adopts the combined structure of mechanical vision projection fringes and an image processing system, which improves the detection efficiency and stability and realizes the high-precision detection of the bending degree of the steel plate.

[0043] Regarding the system fringe projection, the present invention uses a beam splitter to change the projection angle of the fringe projector, which solves the influence of uneven distribution of the projection fringes on the surface of the steel plate to be measured on information processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a schematic diagram of the device structure of an embodiment of the present invention; in the figure, 1 - industrial camera, 2 - fringe projector, 3 - single-density fringe, 4 - transportation system, 5 - steel plate to be measured, 6 - beam splitter, 7 - position sensor, mechanical vision image processing unit: 8 - image processing system, 9 - data line;

[0045] Figure 2 For the present invention Figure 1 The stripe projection effect diagram at position A in the present invention; where A1 is the stripe projection effect diagram of the steel plate without bending; A2 is the stripe projection effect diagram of the bent steel plate;

[0046] Figure 3 It is the system detection flow chart of the embodiment of the present invention;

[0047] Figure 4 It is the image processing flow chart of the embodiment of the present invention; where (a) is the stripe projection in the ideal state; (b) is the stripe projection in the bent state; (c) is the stripe density in each state; (d) is the difference in stripe density in different states. Detailed implementation manners

[0048] Next, the technical solutions and processes in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0049] The present invention provides a measurement system for detecting the bending degree of a steel plate based on projected stripes, including:

[0050] An industrial camera, which collects the projected stripes on the surface of the steel plate to be measured and transmits the collected signal to the machine vision image processing unit;

[0051] A stripe projector, which projects single-density stripes onto the surface of the steel plate to be measured through a beam splitter; the stripe projector includes an LED light source and a single-density stripe pattern. The LED light source irradiates on the single-density stripe pattern sheet to project single-density stripes with a predetermined period onto the surface of the steel plate to be measured, and the size, position, and stripe pattern of the projected stripe pattern can all be adjusted;

[0052] A machine vision processing unit, including an image processing system and a data line. The image processing system uses image processing algorithms to process the captured projected stripe information to obtain the bending degree information of the steel plate to be measured, and the data line is used to transmit the collected signal.

[0053] The system further includes:

[0054] A transportation system, which transports the steel plate to be measured;

[0055] A position sensor, which detects the position of the steel plate to be measured to send a signal to the transportation system to control whether the transportation system stops or not.

[0056] The present invention also provides a measurement method for detecting the bending degree of a steel plate based on projected stripes using the system described above, including the following steps:

[0057] Step S1: Set the parameters of the industrial camera and adjust the imaging range of the industrial camera to just fully capture the surface of the steel plate to be measured; adjust the position of the single-density stripe film in the stripe projector, and make the single-density stripe clearly projected on the measurement position of the steel plate to be measured through the beam splitter.

[0058] Step S2: Start the transportation system. After the transportation system transports the steel plate to be measured until the position sensor detects the position of the steel plate to be measured and sends a signal to the transportation system, the transportation system stops.

[0059] Step S3: The industrial camera collects the projected stripes on the surface of the steel plate to be measured and transmits them to the machine vision processing unit through the data line. Use the image processing algorithm to process the captured projected stripe information to obtain the bending degree information of the steel plate to be measured.

[0060] Step S4: After the machine vision processing unit obtains the information, it sends an instruction to the transportation system, and the transportation system runs normally.

[0061] The following is the specific implementation process of an example of the present invention.

[0062] Please refer to Figures 1-3 , the present invention provides a measurement system for detecting the bending degree of a steel plate based on projected stripes. The system includes an industrial camera 1, a stripe projector 2, a single-density stripe 3, a transportation system 4, a steel plate to be measured 5, a beam splitter 6, a position sensor 7, a machine vision image processing unit: an image processing system 8, and a data line 9. When the system is started, the transportation system 4 transports the steel plate to be measured 5, and the position sensor 7 controls the transportation system to stop when detecting that the steel plate to be measured 5 reaches the detection position. The stripe projector 2 projects the single-density stripe 3 pattern in the direction parallel to the Z axis onto the beam splitter 6, and after being refracted by 90 degrees by the beam splitter 6, it is parallel to the X axis and reaches the surface of the steel plate to be measured 5, as Figure 2 , which is used to sense the surface bending degree information of the steel plate to be measured 5. The industrial camera 1 collects the projected stripe image and records the stripe signal sequence, and transmits the signal sequence to the image processing system (8) through the data line 9 for processing. At the same time, the image processing system 8 will send an instruction to start the transportation system 4 to continue running until the detection of the steel plate to be measured 5 is completed and then stop.

[0063] In this example, refer to Figure 4 , in the case where the steel plate has no bending, that is, in the ideal state, collect the stripe projection diagram in the ideal state, that is Figure 4 (a) Projected stripes in the ideal state: Assume that the stripe intensities of each row in the projected stripe image on the steel plate surface are the same, and select the stripe intensity of one row as the reference stripe intensity; extract the single-density stripes of each row in the projected stripe image on the steel plate surface, and perform Fourier transform on its density information to obtain the frequency spectrum diagram in the ideal state, that is Figure 4 (c) Period density of the unbent stripes:

[0064] Image processing process for judging whether a steel plate is bent: When the steel plate 5 to be measured is bent, the projected fringe image on the surface of the steel plate 5 to be measured is collected, that is Figure 4 (b) Projected fringes in the bent state. The intensity of each row of fringes is calculated with the intensity of the reference fringe to obtain their respective correlation coefficient curves, and then all the correlation coefficient curves are integrated to obtain a sequence diagram of correlation curves. By observing the sequence diagram of correlation curves, it can be clearly known whether the steel plate 5 to be measured is bent.

[0065] Image processing process for calculating the bending degree of a steel plate: When the steel plate 5 to be measured is bent, the projected fringe image on the surface of the steel plate 5 to be measured is collected, and the intensity of each row of fringes in the projected fringe image is extracted and Fourier-transformed to obtain the frequency spectrum diagram of the bent steel plate, that is Figure 4 (c) The period density of the bent fringes. By comparing and calculating the frequency spectrum diagrams of the ideal state and the bent steel plate, the density difference is obtained, and its image is as Figure 4 (d) Shown by the fringe density difference. Substitute its maximum density difference into the formula to calculate the maximum displacement of the time-domain signal on the detection surface of the bent steel plate in the X direction, and then calculate the bending degree of the bent steel plate.

[0066] The above are the preferred embodiments of the present invention. Any changes made according to the technical solutions of the present invention that do not exceed the scope of the technical solutions of the present invention in terms of the functions and effects produced belong to the protection scope of the present invention.

Claims

1. A measurement system for detecting the bending degree of a steel plate based on projection fringes, characterized in that: include: The industrial camera collects the projected stripes on the surface of the steel plate to be tested and transmits the collected signals to the machine vision image processing unit; A fringe projector projects single-density fringe onto the surface of the steel plate to be tested via a beam splitter; the fringe projector comprises an LED light source and a single-density fringe pattern, the LED light source irradiates the single-density fringe pattern sheet to project single-density fringe with a predetermined period onto the surface of the steel plate to be tested, and the size, position and fringe pattern of the projected fringe can be adjusted; The mechanical vision processing unit analyzes and processes the collected signals to obtain information on the bending degree of the steel plate.

2. The measurement system for detecting the bending degree of a steel plate based on projection fringes according to claim 1 is characterized in that: Also includes: Transportation system, to transport the steel plates to be tested; The position sensor detects the position of the steel plate to be tested and sends a signal to the transportation system to control whether the transportation system stops or not.

3. The measurement system for detecting the bending degree of a steel plate based on projection fringes according to claim 1 is characterized in that: The mechanical vision image processing unit includes an image processing system and a data line. The image processing system processes the captured projection fringe information using an image processing algorithm to obtain the bending degree information of the steel plate to be tested. The data line is used to transmit the collected signal.

4. The measurement system for detecting the bending degree of a steel plate based on projection fringes according to claim 3 is characterized in that: The image processing system uses an image processing algorithm to process the captured projection fringe information to obtain the bending degree information of the steel plate to be tested, which is specifically implemented as follows: Select the fringe intensity signal of the middle row of the fringe image under the theoretical state as the reference fringe intensity; The intensity of each row of the steel plate to be tested is selected in turn and the intensity of the reference stripe is used to calculate the correlation coefficient to obtain each correlation coefficient curve. The mathematical relationship of the correlation coefficient is: Among them, ρ(f r ,f j ) is the correlation coefficient value, f r and f j Represent the reference fringe intensity and the j-th row fringe intensity, cov(f r ,f j ) is f r and f j The covariance of fr and Represents f r and f j The standard deviation of , E is the variance; Iteratively calculate the correlation coefficient curve of each row multiple times and then normalize it to between 0 and 1; The correlation coefficient curves of each row are integrated into a correlation coefficient curve sequence, and a preliminary qualitative analysis is performed to determine whether there is bending; Fourier transform is performed on each row of pixels of the fringe projection image under the acquisition theoretical state and the fringe projection image on the surface of the steel plate to be tested to obtain the density information of each row, and then the maximum displacement of the steel plate to be tested along the X direction is calculated according to the relationship between the single-density fringe density and the displacement, and the curvature of the steel plate to be tested is obtained according to the curvature calculation formula.

5. The measurement system for detecting the bending degree of a steel plate based on projection fringes according to claim 1 is characterized in that: The beam splitter can change the direction of the projection stripes of the stripe projector so that the direction in which the projection stripes are projected on the steel plate to be measured coincides with the center position of the industrial camera, thereby ensuring the accuracy of the projection stripes obtained by the industrial camera.

6. The measurement system for detecting the bending degree of a steel plate based on projection fringes according to claim 5 is characterized in that: The projection fringes can be projected horizontally or vertically onto the surface of the steel plate to be tested.

7. A method for measuring the bending degree of a steel plate based on projection fringes using the system as claimed in any one of claims 1 to 6, characterized in that: The steps include: Step S1, setting the parameters of the industrial camera, adjusting the imaging range of the industrial camera to just completely capture the surface of the steel plate to be tested; adjusting the position of the single-density stripe light sheet in the stripe projector, and projecting the single-density stripe clearly on the measuring position of the steel plate to be tested through the beam splitter; Step S2, start the transport system, transport the steel plate to be tested until the position sensor detects the position of the steel plate to be tested and sends a signal to the transport system, and then the transport system stops; Step S3, the industrial camera collects the projected fringes on the surface of the steel plate to be tested, and transmits them to the mechanical vision processing unit via the data line, and uses the image processing algorithm to process the captured projected fringes information to obtain the bending degree information of the steel plate to be tested; Step S4: After obtaining the information, the machine vision processing unit sends instructions to the transportation system, and the transportation system operates normally.

8. The method for measuring the bending degree of a steel plate based on projection fringes according to claim 7 is characterized in that: The specific steps of step S3 are as follows: Step S31, selecting a fringe intensity signal in the middle row of the fringe image under the theoretical state of acquisition as a reference fringe intensity; Step S32, sequentially and intervally select the stripe intensity of each row of the steel plate to be tested and the reference stripe intensity to calculate the correlation coefficient, and obtain each correlation coefficient curve. The mathematical relationship of the correlation coefficient is: Among them, ρ(f r ,f j ) is the correlation coefficient value, f r and f j Represent the reference fringe intensity and the j-th row fringe intensity, cov(f r ,f j ) is f r and f j The covariance of fr and Represents f r and f j E is the standard deviation of the data. Step S33, iteratively calculate the correlation coefficient curve of each row multiple times and then normalize it to between 0 and 1; Step S34, integrate the correlation coefficient curves of each row into a correlation coefficient curve sequence, and make a preliminary qualitative analysis to determine whether there is bending; perform Fourier transform on each row of pixels of the fringe projection image under the theoretical state and the fringe projection image on the surface of the steel plate to be tested to obtain the density information of each row, and then calculate the maximum displacement of the steel plate to be tested along the X direction according to the relationship between the single-density fringe density and the displacement, and obtain the curvature of the steel plate to be tested according to the curvature calculation formula.

9. The method for measuring the bending degree of a steel plate based on projection fringes according to claim 8, characterized in that: The relationship between the single density fringe density and displacement is: Where ΔX(t) is the maximum displacement of the time domain signal of the bending steel plate detection surface in the X direction, f is the focal length of the lens, M a is the imaging scale factor, d t is the stripe density of the single-density stripes at the maximum deviation point of the bent steel plate on the image sensor, and d0 is the reference stripe density of the single-density stripes on the steel plate without bending on the image sensor.

10. The method for measuring the bending degree of a steel plate based on projection fringes according to claim 9, characterized in that: The curvature calculation formula is: Where K is the curvature, Δ is the maximum deviation, and L is the length of the tested steel plate.

Images