Real-time thickness measuring method and thickness measuring device

Abstract

The invention discloses a real-time thickness measuring method and a thickness measuring device, and relates to the technical field of computer vision. The real-time thickness measurement method comprises the steps that a first image set is obtained, and the first image set is shot image frames of a first measurement plane, and the shot image frames are continuously shot by a camera with a calibration plane as the reference along an initial point; recording the shooting coordinates of each frame of image in the first image set, and calculating the definition of each frame of image in the first image set frame by frame; determining a first clear image, and recording a shooting coordinate of the first clear image as a first shooting coordinate; and determining a second clear image, and recording the shooting coordinate of the second clear image as a second shooting coordinate. And respectively calculating the coordinate difference between the first shooting coordinate and the initial point and the coordinate difference between the second shooting coordinate and the initial point to obtain the thickness value of the to-be-measured object. By adopting the technology provided by the invention, the non-contact high-precision measurement of the thickness of the object can be realized at low cost.

Description

Technical Field

[0001] The present invention relates to the technical field of computer vision, and in particular to a real-time thickness measurement method and a thickness measurement device. Background Art

[0002] The leather thickness gauge is a special device used to measure the thickness of leather materials. It usually uses a presser foot and a weight to apply constant pressure to the leather, and the thickness of the leather is determined by measuring the distance the presser foot drops from the leather sample. However, due to the looseness of leather, this crimping measurement method will cause the leather to deform, which will lead to large measurement errors.

[0003] Based on this, with the development of computer vision technology and computer hardware and software, non-contact thickness measurement devices have been proposed, such as 3D image measurement technology. However, the current 3D measurement is still in its infancy, which includes steps such as 3D data acquisition, distortion correction, point cloud processing and 3D reconstruction. The measurement rate is slow and it is difficult to meet the needs of high-speed ranging and thickness measurement. In addition, it has high requirements on the material of the surface of the object to be measured. It is only suitable for measuring the thickness of flat objects, and is relatively limited to the measurement of soft objects such as leather. Summary of the invention

[0004] The present invention provides a real-time thickness measurement method and a thickness measurement device to solve the problems in the prior art that 3D image measurement is difficult to meet the demand for high-speed thickness measurement and 3D measurement equipment is expensive.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is to provide a real-time thickness measurement method, which comprises: A first image set is obtained, wherein the first image set is a first measurement plane captured by the camera with the calibration plane as a reference, along the initial point displacement, and continuously captured. The shooting coordinates of each frame of the image in the first image set are recorded, and the clarity of each frame of the image in the first image set is calculated frame by frame. When the current frame image in the first image set reaches a first clarity threshold, the current frame image is determined to be a first clear image, and the shooting coordinates of the first clear image are recorded as first shooting coordinates.

[0006] When the current frame image in the first image set is lower than the first clarity threshold, the previous frame of the current frame image is determined to be a second clear image, and the shooting coordinates of the second clear image are recorded as second shooting coordinates. The coordinate differences between the first shooting coordinates and the second shooting coordinates and the initial point are calculated respectively to obtain the thickness value of the object to be measured.

[0007] The technical solution provided by the present invention has the following beneficial effects compared with the prior art: The camera continuously shoots the first measuring plane to obtain a set of images of the object to be measured in 2D dimension. The camera is moved based on the principle that the camera can form clear images within a specific depth of field (that is, when the aperture and focal length of the camera lens remain unchanged, the depth of field of the camera can be approximately considered to be a fixed value. When the object enters the depth of field range, it begins to form a clear image. The camera is moved until the surface of the object is clearly imaged, and the moving distance of the camera is the height of the object). When the clarity of the first image set is analyzed frame by frame and the camera shooting coordinates are recorded, the thickness of the object to be measured can be measured in real time and contactlessly. Specifically, the initial point is the clear imaging point when the camera is focused on the calibration plane. When the camera moves to make the current image clear again, the moving distance of the camera in depth compared to the initial point can be regarded as the thickness of the object to be measured.

[0008] Among them, when the first measurement plane of the object to be measured photographed by the camera is clear, the first clear image and the first shooting coordinates are obtained. The first shooting coordinate point is the point closest to the surface of the object to be measured when the camera measures the thickness; the camera is moved along the thickness direction until the first measurement plane of the object photographed by the corresponding camera for the last time is clear, and the second clear image and the second shooting coordinates are obtained. The second shooting coordinate point is the point farthest from the object to be measured by the camera when measuring the thickness. Comparing them with the initial points respectively can determine the measurement error, thereby effectively reducing the influence of surface characteristics (such as reflectivity, roughness, etc.) on the measurement data.

[0009] In addition, compared with the conventional method of measuring the object to be measured by measuring tools, the above real-time thickness measurement method will not cause deformation of the object to be measured; compared with the method of measuring thickness by three-dimensional modeling, the above real-time thickness measurement method requires less calculation, has low equipment cost, and has small measurement error. It is more suitable for thickness measurement of various materials, such as transparent objects, low reflectivity objects, sponge-like objects and other objects with complex surface shapes.

[0010] In some embodiments, the real-time thickness measurement method further includes system calibration, specifically: A standard plane is preset, wherein a marker is projected onto the standard plane; a second image set is obtained, wherein the second image set is image frames of the standard plane continuously shot by the camera; the shooting coordinates of each image frame in the second image set are recorded, and the clarity of each image frame in the second image set is calculated frame by frame.

[0011] Determine a clarity peak in the second image set; when the clarity peak reaches a second clarity threshold, record the shooting coordinates corresponding to the clarity peak as the coordinates of the initial point, and record the standard plane as the calibration plane, wherein the object to be tested is placed on the calibration plane.

[0012] By adopting the above technical solution, the starting point of the camera displacement can be determined through system calibration, that is, the initial point is the point with zero thickness. Among them, the preset standard plane is generally the placement plane of the object to be measured. In some application scenarios, the marker can also be replaced to verify the accuracy of the system calibration result, thereby ensuring the ranging accuracy of the calibrated camera.

[0013] In some application scenarios, the acquisition of the first image set also includes projecting a marker onto the first measurement plane, wherein the marker is a pattern that matches the surface features of the first measurement plane, or the marker is a preset pattern. That is, when the features on the first measurement plane of the object to be measured are obvious, the first image set captured can collect enough features to meet the definition calculation requirements; when the features on the first measurement plane of the object to be measured are insufficient, the feature extraction of the image can be enhanced by projecting the marker. The marker can also reduce the interference of environmental factors on the thickness measurement of the object to be measured.

[0014] In some embodiments, the real-time thickness measurement method further comprises: It is determined whether the first measurement plane has a captureable feature; if the captureable feature is present, the camera is displaced along the initial point with the calibration plane as a reference, and continuously photographs the first measurement plane.

[0015] Otherwise, perform the following steps: Projecting a marker onto the first measurement plane; photographing the marker using a first camera and a second camera respectively, wherein the distance between the projection centers of the first camera and the second camera is , let the imaging point of the first camera be , let the imaging point of the second camera be .

[0016] Then the imaging point of the first camera is The imaging point of the second camera The distance between them is:

[0017] in, is the distance from the imaging point of the first camera to the imaging surface, is the distance from the imaging point of the second camera to the imaging surface.

[0018] Then, calculate the height value , recorded as:

[0019] Wherein, f represents the focal length of the first camera and the second camera.

[0020] By adopting the above technical solution, when there is no obvious feature that can be captured on the surface of the object to be tested, the parallax between the first camera and the second camera can be used to simulate the stereoscopic vision of the human eye to determine the depth information of the marker. Therefore, the height value without the object to be tested can be used as the reference height, and the height value after the object to be tested is placed can be used as the current height. The difference between the current height and the reference height is the thickness value of the object to be tested.

[0021] In some embodiments, determining the first clear image of the first image set and recording the shooting coordinates of the first clear image as the first shooting coordinates also includes: recording the shooting time of the first clear image as the first shooting time; and recording the shooting time of the second clear image as the second shooting time.

[0022] A spectral confocal sensor is started to measure the distance of a second measurement plane from the first shooting time to the second shooting time, wherein the second measurement plane and the first measurement plane are opposite sides of the object to be measured; and a distance measurement result of the spectral confocal sensor is recorded, wherein the distance measurement of the spectral confocal sensor is synchronized with the camera shooting.

[0023] With the above technical solution, the spectral confocal sensor will regard the lower surface of the object to be measured as the same plane when measuring distance. However, in actual application scenarios, the lower surface of the object to be measured will have a certain curvature, especially for soft leather items. Therefore, when the camera takes a clear image and then starts the spectral confocal sensor, the closest real plane of the object to be measured is determined, thereby effectively reducing the ranging error of the spectral confocal sensor.

[0024] In some embodiments, the frame-by-frame calculation of the clarity of each frame in the first image set also includes: determining a clarity peak in the first image set, and recording the image corresponding to the clarity peak as a third clear image, and recording the shooting time of the third clear image as a third shooting time; and recording the result of the ranging of the spectral confocal sensor at the third shooting time as the final measurement result.

[0025] With the above technical solution, since objects can be clearly imaged within the depth of field of the camera, it is approximately considered that the depth of field of the camera is a fixed value, which will result in a certain error. In order to reduce this error, the spectral confocal sensor and the camera are synchronized (that is, when the camera moves once and shoots, the spectral confocal sensor measures the thickness information once at the same time) to obtain the best focus position of the camera, thereby narrowing the range of camera imaging clarity and reducing the camera measurement error.

[0026] Furthermore, starting the spectral confocal sensor to measure the distance of the second measurement plane from the first shooting time to the second shooting time also includes: projecting a marker on the second measurement plane, wherein the marker is a pattern matching the surface features of the second measurement plane, or the marker is a preset pattern; the projected marker can enhance the image features of the second measurement plane.

[0027] In some embodiments, respectively calculating the coordinate differences between the first shooting coordinate and the second shooting coordinate and the initial point to obtain the thickness value of the object to be measured further includes: Calculate the height difference between the first shooting coordinate and the second shooting coordinate, and record the height difference as the thickness value of the object to be measured; obtain multiple thickness values ​​of the object to be measured within the preset measurement period according to a preset measurement period, and generate a thickness value set; determine the maximum thickness value in the thickness value set, and record the maximum thickness value as the true thickness value of the object to be measured.

[0028] By adopting the above technical solution, when the object to be tested is leather, for example, the surface flatness of the material is low, taking the maximum measured value as the true thickness value of the object to be tested can facilitate comparison of whether the layer thickness on the leather surface meets specific quality control standards, that is, ensuring that the thickest part of the leather does not exceed the specified upper limit.

[0029] In some embodiments, when the detection accuracy of the first measurement plane within a preset measurement cycle is greater than a preset theoretical accuracy, the thickness of the first measurement plane is marked as abnormal; and / or when the detection accuracy of the second measurement plane within a preset measurement cycle is greater than the preset theoretical accuracy, the thickness of the second measurement plane is marked as abnormal.

[0030] By adopting the above technical solution and adding the thickness abnormality judgment, it is possible to judge whether the object to be tested meets the specified standards and timely identify whether there are defects on the first measuring surface and the second measuring surface of the object to be tested.

[0031] In some embodiments, the present application also provides a thickness measurement device, which applies the above-mentioned real-time thickness measurement method, including: a camera, a displacement measurement device and a placement plane.

[0032] The camera is based on a calibration plane and is used to move along an initial point and continuously photograph a first measurement plane to obtain a first image set, wherein the first measurement plane is the front or back side of the object to be measured; the displacement measurement device is used to measure the shooting coordinates of the camera; the object to be measured is located on the moving plane, wherein the camera is suspended on one side of the moving plane along a first direction.

[0033] In some embodiments, the thickness measuring device also includes a spectral confocal sensor and a laser source, wherein the spectral confocal sensor is suspended on a side of the mobile platform away from the camera and is used to measure the distance to a second measurement plane, wherein the second measurement plane and the first measurement plane are opposite sides of the object to be measured; and the laser source is used to project a marker onto the first measurement plane and / or the second measurement plane.

[0034] The above technical solution is adopted, and the clarity of the image captured by the camera is analyzed and the clarity is calculated by communicating with the computer program product. Among them, the camera can be replaced with a telecentric lens with different depth of field according to the needs of image acquisition. The camera can be fixed to the transport device to automatically adjust the position information of the camera, such as moving up or down in the thickness direction, and the current position information of the camera can be recorded in real time through the displacement measurement device and transmitted to the computer program product. In addition, it can be used with a spectral confocal sensor for extended testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work, among which: Figure 1 This is a schematic diagram of a process of an embodiment of a real-time thickness measurement method provided by the present invention. Figure 1 ; Figure 2 This is an imaging measurement diagram of an embodiment of a real-time thickness measurement method provided by the present invention. Figure 1 ; Figure 3 This is a schematic diagram of a process of an embodiment of a real-time thickness measurement method provided by the present invention. Figure 2 ; Figure 4 A first image set of a first measurement plane of an embodiment of a real-time thickness measurement method provided by the present invention; Figure 5 This is a schematic diagram of a process of an embodiment of a real-time thickness measurement method provided by the present invention. Figure 3 ; Figure 6 This is an imaging measurement diagram of an embodiment of a real-time thickness measurement method provided by the present invention. Figure 2 ; Figure 7 This is a measurement data diagram of sample No. 1 on the first measurement plane in an embodiment of a real-time thickness measurement method provided by the present invention; Figure 8This is a measurement data diagram of sample No. 2 on the first measurement plane in one embodiment of a real-time thickness measurement method provided by the present invention; Fig. 9 This is a measurement data diagram of sample No. 3 on the first measurement plane in an embodiment of a real-time thickness measurement method provided by the present invention; Fig.10 It is a schematic diagram of imaging measurement of an embodiment of a thickness measurement device provided by the present invention. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0037] In some embodiments, the real-time thickness measurement method comprises: Step S100: acquiring a first image set.

[0038] Exemplarily, the first image set is the image frames of the first measurement plane captured by the camera 10 with the calibration plane as the reference, along the displacement of the initial point, and continuously captured. For example, the distance that the camera 10 moves in the thickness direction each time is set to 5 μm, and the camera 10 captures an image each time it moves, and the collection of all captured images is the first image set. Among them, the calibration plane is the placement plane 20 of the object 40 to be measured, for example, the object 40 to be measured is leather, and the calibration plane is the transfer platform of the leather; the initial point is the starting point of the displacement shooting of the camera 10.

[0039] Step S200 , recording the shooting coordinates of each frame of the image in the first image set, and calculating the clarity of each frame of the image in the first image set frame by frame.

[0040] Exemplarily, during the shooting process, the camera 10 can be moved by a robot or a pan / tilt head or other moving devices, and then the displacement distance of the camera 10 is recorded by a displacement measuring device, for example, a displacement sensor or a counting sensor is set on the moving device of the camera 10.

[0041] When the object 40 to be measured is leather, for example, the first measurement plane may have more feature points, that is, the information content of any image in the first image set is sufficient, so the entropy value of the image can be calculated by counting the probability of occurrence of each grayscale pixel, and then the entropy value formula is used to evaluate the image clarity. Among them, the present application does not limit the calculation method of clarity, for example, the energy gradient function, the image clarity evaluation function based on two-dimensional discrete Fourier transform, etc. can also be used to evaluate the image clarity.

[0042] Step S300: When the current frame image in the first image set reaches a first clarity threshold, the current frame image is determined to be a first clear image, and the shooting coordinates of the first clear image are recorded as first shooting coordinates.

[0043] For example, in combination Figure 2 As shown, when the object 40 to be tested is not placed, the distance between the camera 10 and the calibration plane at the initial point is the clear imaging distance, and after the object 40 to be tested is placed, it has a thickness, so that the distance between the camera 10 and the calibration plane at the initial point is less than the clear imaging distance. Therefore, during the displacement process of the camera 10, the imaging rule of the first image set is blur to clear and then blur (for example Figure 4 For example, if the camera 10 moves along the thickness direction so that the distance between the camera 10 and the upper surface of the object 40 to be measured is a clear imaging distance, then the displacement distance of the camera 10 is the thickness of the object 40 to be measured.

[0044] Step S400: when the current frame image in the first image set is lower than the first clarity threshold, determine that the previous frame of the current frame image is a second clear image, and record the shooting coordinates of the second clear image as second shooting coordinates.

[0045] Since the above working principle is: when the aperture and focal length of the camera 10 lens remain unchanged, the depth of field of the camera 10 can be approximately considered to be a fixed value. When the object enters the depth of field range, a clear image begins to be formed, and the camera 10 is moved until the surface of the object is clearly imaged. The moving distance of the camera 10 is the height of the object. However, objects can be clearly imaged within the depth of field of the camera 10. Therefore, there will be a certain error in approximately considering the depth of field of the camera 10 to be a fixed value. Exemplarily, by acquiring the second clear image, the nearest end (first shooting coordinate) and the farthest end (second shooting coordinate) between the camera 10 and the surface of the object 40 to be measured within the clear imaging distance can be obtained, which is convenient for calculating the maximum and minimum values ​​of the measured thickness, thereby facilitating the judgment of the error.

[0046] Step S500 , respectively calculating the coordinate differences between the first shooting coordinate and the second shooting coordinate and the initial point to obtain the thickness value of the object 40 to be measured.

[0047] Exemplarily, the movement distance of the camera 10 in the thickness direction compared to the initial point can be regarded as the thickness of the object to be measured 40. By comparing the first shooting coordinate and the second shooting coordinate with the initial point (for example, the depth value of the first shooting coordinate minus the depth value of the initial point; the depth value of the second shooting coordinate minus the depth value of the initial point), the maximum and minimum values ​​of the thickness can be calculated, and the measurement error can be determined, thereby effectively reducing the influence of surface characteristics (such as reflectivity, roughness, etc.) on the measurement data.

[0048] Among them, the present application does not limit the selection of the actual thickness value. For example, if the diameter CC of the circle of confusion of the telecentric lens of the camera 10 is 10μm and the magnification is 1, then the depth of field is calculated as: DOF=2*10*(1+1) / 1²=2*10*2 / 1=40μm, that is, the calculation error of the camera 10 is ±40μm. Therefore, the selection of the first shooting coordinate, the second shooting coordinate, or the shooting coordinate corresponding to any clear image between the two to calculate the thickness value can meet the demand for online detection of leather thickness during the production line production process.

[0049] In the embodiment of the present application, the measurement method described above has adjustable measurement accuracy and low computing power requirements, and the real-time shooting of the camera 10 can be applied to high-speed motion measurement in the production line.

[0050] In some embodiments, in combination Figure 3 As shown, the real-time thickness measurement method further includes system calibration, specifically: step S110, presetting a standard plane.

[0051] For example, the standard plane is generally a placement plane 20 of the object 40 to be tested, and a marker 30 is projected onto the placement plane 20. Figure 2 As shown in , the marker 30 may be a cross so that the camera 10 can focus on the standard plane. The present application does not limit the selection of the marker 30, for example, a chessboard type may also be selected, and the marker 30 selected during the calibration process may be consistent with or different from the marker 30 during the actual measurement process.

[0052] Step S120: acquiring a second image set.

[0053] For example, when the camera 10 is calibrated, it is moved by the same manipulator or pan / tilt as in the actual measurement, and the shooting parameters of the camera 10 are the same as in the actual measurement to ensure the accuracy and reliability of the measurement. The second image set is the image frames of the standard plane continuously shot by the camera 10.

[0054] Step S130 , recording the shooting coordinates of each frame of the image in the second image set, and calculating the clarity of each frame of the image in the second image set frame by frame.

[0055] Exemplarily, the coordinate recording method and the clarity calculation method of the second image set are consistent with actual measurement to minimize calculation errors.

[0056] Step S140, determining a definition peak value in the second image set.

[0057] Exemplarily, since the camera 10 can clearly image within the depth of field, the corresponding shooting point position, that is, the optimal focus position of the camera 10 , can be obtained by determining the clarity peak value in the second image set.

[0058] Step S150: When the clarity peak reaches the second clarity threshold, the shooting coordinates corresponding to the clarity peak are recorded as the coordinates of the initial point, and the standard plane is recorded as the calibration plane.

[0059] For example, the second definition threshold is set to determine whether the best focus position of the standard plane photographed by the current camera 10 meets the measurement accuracy requirement. When the definition peak is lower than the second definition threshold, the standard plane can be adjusted or the camera 10 with a different depth of field can be replaced, and the calibration can be performed again. When the definition peak is higher than the second definition threshold, it can be considered that the current calibration has met the accuracy requirement and the calibration is completed.

[0060] In some embodiments, the above-mentioned real-time thickness measurement method further comprises: Step S101, determining whether the first measurement plane has a captureable feature.

[0061] For example, when the object to be tested is leather, most leather surfaces have obvious features (e.g., texture, pattern), e.g. Figure 4 As shown, Figure 4 It is a grayscale image of the first image set of the first measurement plane of the leather, which has enough feature points for the camera 10 to collect, and the first image set conforms to blur to clear to blur.

[0062] Otherwise, combine Figure 5 As shown, perform the following steps: Step S600: projecting the marker 30 onto the first measurement plane.

[0063] The surface features of the first measurement plane are enriched by projecting the marker 30, wherein the marker 30 may be a pattern matching the surface features of the first measurement plane, or a preset pattern (such as a cross, a dot, etc.). By way of example, when the features on the first measurement plane of the object 40 to be measured are insufficient, by projecting the marker 30, feature extraction of the image can be enhanced. And the interference of environmental factors on the thickness measurement of the object 40 to be measured can be reduced. By way of example, the marker 30 may be a conventional pattern (such as a Figure 2 The cross marker 30 in the image may be matched with the feature of the object 40 to be measured (for example, leather texture). That is, when the first measurement plane has enough captureable features due to the projection of the marker 30, steps S100 to S500 may also be performed.

[0064] Step S700 may also be executed to use the first camera and the second camera to photograph the landmark 30 respectively.

[0065] See also Figure 6 As shown, Figure 6 The imaging measurement diagram of an embodiment of a real-time thickness measurement method provided by the present application is shown in FIG. Figure 2 .

[0066] Exemplarily, a circular marker 30 is projected onto the first measurement plane, and the first camera and the second camera respectively photograph the marker 30 at set positions. The distance (baseline) between the projection centers of the first camera and the second camera is denoted as , the imaging point of the first camera is , the imaging point of the second camera is .

[0067] Then the imaging point of the first camera is The imaging point of the second camera The distance between them is:

[0068] in, is the distance from the imaging point of the first camera to the imaging surface, is the distance from the imaging point of the second camera to the imaging surface; Then, calculate the height value , recorded as:

[0069] Wherein, f represents the focal length of the first camera and the second camera.

[0070] For example, in combination Figure 6 As shown, when there is no obvious feature that can be captured on the surface of the object 40 to be measured, the parallax between the first camera and the second camera can be used to simulate the stereoscopic vision of the human eye to determine the depth information of the marker 30. Referring to step 200, the first camera and the second camera are moved to obtain the shooting coordinates of a clear image of a clarity threshold, and the current height value is measured with the shooting coordinates.

[0071] Specifically, the height value without the object to be measured is calculated by parallax as the reference height, and the height value with the object to be measured is the current height. The difference between the current height and the reference height is the thickness value of the object to be measured 40 .

[0072] In some implementation schemes, step S300, determining the first clear image of the first image set, and recording the shooting coordinates of the first clear image as the first shooting coordinates, also includes: recording the shooting time of the first clear image as the first shooting time; recording the shooting time of the second clear image as the second shooting time.

[0073] The spectral confocal sensor 60 is started to measure the distance of the second measurement plane from the first shooting time to the second shooting time, wherein the second measurement plane and the first measurement plane are opposite sides of the object to be measured 40; and the distance measurement result of the spectral confocal sensor 60 is recorded, wherein the distance measurement of the spectral confocal sensor 60 is synchronized with the shooting of the camera 10.

[0074] In the embodiment of the present application, the spectral confocal sensor 60 will regard the lower surface of the object to be measured 40 as the same plane when measuring the distance. However, in actual application scenarios, the lower surface of the object to be measured 40 will have a certain curvature, especially for soft leather products. Therefore, when the camera 10 takes a clear image, the spectral confocal sensor 60 is activated to determine the closest real plane of the object to be measured 40, thereby effectively reducing the distance measurement error of the spectral confocal sensor 60.

[0075] Exemplarily, the steps of using the spectral confocal sensor 60 for extended measurement are: 1) Project the marker 30 onto the second measurement plane, and regard the second measurement plane as the standard plane. 2) Pre-install the spectral confocal sensor 60, and simultaneously measure the distance from the spectral confocal sensor 60 to the standard plane. 3) When the camera 10 shoots the first measurement plane and determines that the current camera 10 shooting coordinates are the optimal focus coordinates, the real lower surface can be obtained, that is, the distance between the camera 10 and the spectral confocal sensor 60 minus the clear imaging distance of the camera 10 is the distance between the spectral confocal sensor 60 and the real lower surface; 4) Start the spectral confocal sensor 60 and the object to be measured 40 to synchronously displace and measure the distance.

[0076] In some embodiments, step S200, calculating the clarity of each frame in the first image set frame by frame, also includes: determining a clarity peak in the first image set, and recording the image corresponding to the clarity peak as a third clear image, and recording the shooting time of the third clear image as a third shooting time; recording the result of the ranging of the spectral confocal sensor 60 at the third shooting time as the final measurement result.

[0077] In the embodiment of the present application, in order to reduce the error of depth of field estimation, the spectral confocal sensor 60 is operated synchronously with the camera 10 (that is, when the camera 10 moves once and shoots, the spectral confocal sensor 60 simultaneously measures the thickness information once) to assist in obtaining the optimal focus position of the camera 10, thereby narrowing the range of imaging clarity of the camera 10 to reduce the measurement error of the camera 10.

[0078] Furthermore, starting the spectral confocal sensor 60 to measure the distance of the second measurement plane from the first shooting time to the second shooting time also includes: projecting a marker 30 on the second measurement plane, wherein the marker 30 is a pattern matching the surface features of the second measurement plane, or the marker 30 is a preset pattern; the projected marker 30 can enhance the image features of the second measurement plane.

[0079] In some embodiments, step S500, respectively calculating the coordinate differences between the first shooting coordinate and the second shooting coordinate and the initial point to obtain the thickness value of the object to be measured 40, further includes: Calculate the height difference between the first shooting coordinate and the second shooting coordinate, and record the height difference as the thickness value of the object 40 to be measured; obtain multiple thickness values ​​of the object 40 to be measured within a preset measurement period according to a preset measurement period, and generate a thickness value set; determine the maximum thickness value in the thickness value set, and record the maximum thickness value as the true thickness value of the object 40 to be measured.

[0080] In the embodiment of the present application, when the object to be tested is leather, for example, the surface flatness of the material is low, and taking the maximum measured value as the true thickness value of the object to be tested 40 can facilitate comparison of whether the layer thickness on the leather surface meets specific quality control standards, that is, ensuring that the thickest part of the leather does not exceed the specified upper limit.

[0081] For example, Figure 7 The measurement data of sample 1 (black rough leather) on the first measurement plane. A total of 3148 records were measured in a 3S measurement cycle, with a deviation of <10μm per second and a single-point repeat error of <10μm; the deviation within 2.5S was 25μm, and the error was caused by the uneven measurement plane of the leather.

[0082] Figure 8 The measurement data of sample No. 2 (brown rough leather) on the first measurement plane. A total of 2737 records were measured in a measurement cycle of 2.7 seconds, with a deviation of less than 10μm per second and a single-point repetition error of less than 10μm; the deviation within 2.5 seconds was 20μm, and the error was caused by the uneven measurement plane of the leather.

[0083] Fig. 9 The measurement data of sample No. 3 (black fine leather) on the first measurement plane. A total of 2884 records were measured in a measurement cycle of 2.8S, with a deviation of 16μm per second. When the maximum value is the true distance of the leather surface, the deviation is <10μm. The error is caused by the uneven measurement plane of the leather.

[0084] In some embodiments, when the detection accuracy of the first measurement plane within a preset measurement cycle is greater than a preset theoretical accuracy, the thickness of the first measurement plane is marked as abnormal; and / or when the detection accuracy of the second measurement plane within a preset measurement cycle is greater than a preset theoretical accuracy, the thickness of the second measurement plane is marked as abnormal.

[0085] In the embodiment of the present application, adding thickness abnormality judgment can determine whether the object to be tested 40 meets the specified standards, and timely identify whether there are defects on the first measuring surface and the second measuring surface of the object to be tested 40.

[0086] In some embodiments, in combination Fig.10As shown, the present application also provides a thickness measurement device, which applies the above-mentioned real-time thickness measurement method, including: a camera 10, a displacement measurement device and a placement plane 20. The camera 10 is used to displace along the initial point with the calibration plane as a reference, and continuously shoot the first measurement plane to obtain a first image set, wherein the first measurement plane is the front or back of the object 40 to be measured; the displacement measurement device is used to measure the shooting coordinates of the camera 10; the object 40 to be measured is located on the moving plane, wherein the camera 10 is suspended on one side of the moving plane along the first direction.

[0087] In some embodiments, the thickness measuring device also includes a spectral confocal sensor 60 and a laser source 50. The spectral confocal sensor 60 is suspended on a side of the mobile platform away from the camera 10 and is used to measure the distance to a second measurement plane, wherein the second measurement plane and the first measurement plane are opposite sides of the object to be measured 40; the laser source 50 is used to project the marker 30 onto the first measurement plane and the second measurement plane.

[0088] In the embodiment of the present application, the camera 10 is connected to the computer program product through communication, so as to analyze the clarity of the image captured by the camera 10 and calculate the clarity. Among them, the camera 10 can be replaced with a telecentric lens with different depths of field according to the needs of image acquisition. The camera 10 is fixed to a manipulator or a transfer device to automatically adjust the position information of the camera 10, such as moving up or down in the thickness direction, and the current position information of the camera 10 is recorded in real time through a displacement measuring device and transmitted to the computer program product.

[0089] Exemplarily, the laser source 50 is a projection device for the marker 30, which can select marker patterns of different shapes, colors or thicknesses according to the different material characteristics of the object 40 to be tested. In addition, it can be matched with a spectral confocal sensor 60 for extended testing. For example, the spectral confocal sensor 60 used in this application is a linear type (such as the TS-C400 spectral confocal sensor 60), with a reference distance of 35mm, a measurable range of 4mm, an accuracy of 1μm, and a linear error of <±0.12μm. Among them, when the measurement error of the camera 10 is ±40μm, the error of the above-mentioned thickness measuring device can be effectively controlled within ±60μm.

[0090] The above description is only an implementation mode of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly used in other related technical fields, should be included in the protection scope of the present invention.

Claims

1. A real-time thickness measurement method, characterized in that: include: Acquire a first image set, wherein the first image set is image frames of a first measurement plane that are continuously photographed by the camera based on a calibration plane and displaced along an initial point; Recording the shooting coordinates of each frame of the first image set, and calculating the clarity of each frame of the first image set frame by frame; When the current frame image in the first image set reaches a first clarity threshold, determining the current frame image as a first clear image, and recording the shooting coordinates of the first clear image as first shooting coordinates; When the current frame image in the first image set is lower than the first clarity threshold, determining that the previous frame of the current frame image is a second clear image, and recording the shooting coordinates of the second clear image as second shooting coordinates; The coordinate differences between the first shooting coordinate and the second shooting coordinate and the initial point are calculated respectively to obtain the thickness value of the object to be measured.

2. The real-time thickness measurement method according to claim 1, characterized in that: The real-time thickness measurement method also includes system calibration, specifically: Preset a standard plane, wherein a marker is projected onto the standard plane; Acquire a second image set, wherein the second image set is image frames of the standard plane continuously photographed by the camera; Recording the shooting coordinates of each frame of the image in the second image set, and calculating the clarity of each frame of the image in the second image set frame by frame; determining a peak sharpness in the second image set; When the clarity peak reaches a second clarity threshold, the shooting coordinates corresponding to the clarity peak are recorded as the coordinates of the initial point, and the standard plane is recorded as the calibration plane, wherein the object to be measured is placed on the calibration plane.

3. The real-time thickness measurement method according to claim 1, characterized in that: The real-time thickness measurement method further includes: Determining whether the first measurement plane has a capturable feature; If the captureable feature exists, the camera moves along the initial point with the calibration plane as a reference, and continuously photographs the first measurement plane; Otherwise, perform the following steps: Projecting a marker onto the first measurement plane; The marker is photographed using a first camera and a second camera respectively, wherein the distance between the projection centers of the first camera and the second camera is , let the imaging point of the first camera be , let the imaging point of the second camera be ; Then the imaging point of the first camera is The imaging point of the second camera The distance between them is: in, is the distance from the imaging point of the first camera to the imaging surface, is the distance from the imaging point of the second camera to the imaging surface; Then, calculate the base height , recorded as: Wherein, f represents the focal length of the first camera and the second camera.

4. The real-time thickness measurement method according to claim 1, characterized in that: The determining of the first clear image of the first image set and recording the shooting coordinates of the first clear image as the first shooting coordinates further includes: The shooting time of the first clear image is recorded as the first shooting time; The shooting time of the second clear image is recorded as the second shooting time; Starting the spectral confocal sensor to measure the distance of a second measurement plane from the first shooting time to the second shooting time, wherein the second measurement plane and the first measurement plane are opposite sides of the object to be measured; The distance measurement result of the spectral confocal sensor is recorded.

5. The real-time thickness measurement method according to claim 4, characterized in that: The step of calculating the clarity of each frame of the first image set frame by frame further includes: Determine a clarity peak value in the first image set, and record an image corresponding to the clarity peak value as a third clear image, and record a shooting time of the third clear image as a third shooting time; The distance measurement of the spectral confocal sensor is synchronized with the camera shooting, and the distance measurement result of the spectral confocal sensor at the third shooting time is recorded as the final measurement result.

6. The real-time thickness measurement method according to claim 5, characterized in that: The method of starting the spectral confocal sensor to measure the distance of the second measurement plane from the first shooting time to the second shooting time also includes: a marker is projected onto the second measurement plane, wherein the marker is a pattern matching the surface features of the second measurement plane, or the marker is a preset pattern.

7. The real-time thickness measurement method according to claim 1, characterized in that: The step of respectively calculating the coordinate differences between the first shooting coordinate and the second shooting coordinate and the initial point to obtain the thickness value of the object to be measured further includes: Calculate the height difference between the first shooting coordinate and the second shooting coordinate, and record the height difference as the thickness value of the object to be measured; Acquire a plurality of thickness values ​​of the object to be measured within the preset measurement period according to the preset measurement period, and generate a thickness value set; A maximum thickness value in the thickness value set is determined, and the maximum thickness value is recorded as the true thickness value of the object to be measured.

8. The real-time thickness measurement method according to any one of claims 1 to 7, characterized in that: The real-time thickness measurement method further comprises: When the detection accuracy of the first measurement plane within a preset measurement period is greater than a preset theoretical accuracy, marking the thickness of the first measurement plane as abnormal; and / or When the detection accuracy of the second measurement plane within the preset measurement period is greater than the preset theoretical accuracy, the thickness of the second measurement plane is marked as abnormal.

9. A thickness measuring device, characterized in that: The real-time thickness measurement method according to any one of claims 1 to 8 comprises: A camera, the camera is based on the calibration plane, and is used to move along the initial point and continuously photograph a first measurement plane to obtain a first image set, wherein the first measurement plane is the front side or the back side of the object to be measured; A displacement measuring device, wherein the displacement measuring device is used to measure the shooting coordinates of the camera; A moving plane, the object to be tested is located on the moving plane, wherein the camera is suspended on one side of the moving plane along a first direction.

10. The thickness measuring device according to claim 9, characterized in that: The thickness measuring device also includes a spectral confocal sensor and a laser source. The spectral confocal sensor is suspended on a side of the mobile platform away from the camera and is used to measure the distance to a second measurement plane, wherein the second measurement plane and the first measurement plane are opposite sides of the object to be measured; the laser source is used to project a marker onto the first measurement plane and / or the second measurement plane.

Images