Vision-based high-precision hydrophone calibration positioning mechanism and method

Abstract

The invention discloses a vision-based high-precision hydrophone calibration positioning mechanism and method, belongs to the technical field of underwater acoustic measurement, and mainly solves the problems of low precision, poor repeatability and great influence of pressure deformation caused by dependence on manual vision and mechanical positioning in existing laser method hydrophone calibration. The system comprises a visual imaging unit arranged outside the pressure water tank, a precision motion unit driving the visual imaging unit to move, and an image processing and control unit, squint imaging is carried out through a high-resolution industrial camera, the optimal visual angle is adjusted by combining a six-axis displacement platform, distortion correction, edge feature extraction and coordinate transformation are carried out by using an image processing algorithm, and the precision of the pressure water tank is improved. According to the invention, non-contact and automatic alignment is realized, pressure deformation interference is overcome, especially, the problem that visual positioning of the film sheet is difficult is effectively solved through strabismus imaging, and the positioning precision of the calibration device is improved.

Description

Technical Field

[0001] This invention relates to the field of underwater acoustic measurement and metrology technology, specifically to the positioning mechanism in a hydrophone sensitivity calibration device, and particularly to a non-contact, high-precision positioning device based on machine vision suitable for laser-based sound pressure calibration systems. Background Technology

[0002] Hydrophones are key sensors in underwater acoustic measurements, and their sensitivity calibration accuracy directly affects the accuracy and reliability of the underwater acoustic measurement system. Laser interferometry is currently one of the main methods for high-precision hydrophone sound pressure calibration. A typical implementation is the substitution method: an auxiliary transducer is used to generate a sound field in a sealed pressurized water tank. A standard diaphragm is used as the sound pressure transmission medium, and a laser vibrometer is used to measure the vibration velocity of the diaphragm under the influence of the sound field. This allows for the calculation of the sound pressure acting on the surface of the hydrophone under test, ultimately achieving sensitivity calibration. During calibration, the standard diaphragm and the hydrophone under test must be precisely aligned to ensure the accuracy of sound pressure transmission and the validity of measurement results. Traditional positioning methods mainly rely on precision motion mechanisms combined with manual visual adjustment. However, under high hydrostatic pressure test conditions, the pressure tank and internal installation structure may undergo slight deformation. Traditional rigid motion mechanisms may also produce unpredictable elastic deformation or displacement when subjected to pressure loads. These factors will directly cause the pre-set mechanical positioning coordinates to drift, thereby introducing additional alignment errors. In addition, manual visual adjustment is significantly affected by the operator's experience, visual fatigue, and subjective judgment, making it difficult to achieve stable, repeatable sub-millimeter level or even higher precision positioning requirements. This has become one of the bottlenecks restricting the further improvement of the overall uncertainty of laser hydrophone calibration devices.

[0003] With the development of machine vision technology, high-resolution industrial cameras and advanced image processing algorithms have been widely used in industrial inspection, precision alignment, and other fields. However, in the field of underwater acoustic metrology, especially in specific application scenarios involving high pressure, underwater, and non-contact applications, there is currently no publicly available mature technical solution for effectively integrating visual positioning technology into existing calibration devices to overcome the effects of pressure deformation and achieve high-precision, automated alignment of the diaphragm and hydrophone. Existing technologies lack dedicated visual inspection methods for diaphragm target characteristics in hydrophone calibration, such as thin edges, low contrast, and potential optical interference, and also lack a systematic design for collaborative operation with a six-degree-of-freedom precision motion platform under pressure. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a vision-based hydrophone calibration and positioning mechanism. This mechanism offers a non-contact optical positioning method that completely eliminates alignment errors caused by mechanical contact or deformation due to pressure. It enables automated, high-precision detection and positioning of the diaphragm and hydrophone during calibration, reduces the influence of human factors, improves the objectivity and repeatability of the calibration process, and enhances the positioning accuracy and efficiency of the entire hydrophone laser sound pressure calibration system.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a vision-based hydrophone calibration and positioning mechanism, applied in a laser-based hydrophone sound pressure calibration system, the calibration system comprising a pressure water tank and a standard hydrophone and diaphragm disposed therein, the pressure water tank having an observation window, characterized in that the positioning mechanism comprises:

[0006] A visual imaging unit is located outside the pressure tank, with its optical axis passing through the observation window and aligned with the target area inside the tank, for acquiring images including the end face of the standard hydrophone and the diaphragm.

[0007] A precision motion unit is driven and connected to the visual imaging unit to adjust the spatial position and attitude of the visual imaging unit.

[0008] An image processing and control unit is signal-connected to the visual imaging unit and the precision motion unit. The image processing and control unit is configured to: perform distortion correction on the image acquired by the visual imaging unit; identify and extract the edge contour features of the standard hydrophone end face and the diaphragm from the corrected image; calculate the relative positional deviation between the standard hydrophone and the diaphragm based on the edge contour features; and output a control signal based on the relative positional deviation to drive the actuator in the calibration system for adjusting the position of the standard hydrophone or the diaphragm until the relative positional deviation is less than a preset threshold.

[0009] Preferably, the visual imaging unit includes a high-resolution industrial area array camera and an optical lens; the industrial camera has a pixel count of not less than 100 million pixels; and the optical lens has an adjustable focal length or is a fixed focal length lens.

[0010] Preferably, the visual imaging unit adopts an oblique imaging layout, such that the optical axis of the industrial camera forms an angle with the normal of the standard hydrophone end face and the plane where the diaphragm is located, and the angle is an acute angle.

[0011] Preferably, the precision motion unit includes an electric six-axis displacement platform, and the industrial camera is mounted on the actuator of the six-axis displacement platform to drive the industrial camera to perform translation and rotation in three-dimensional space, so as to realize attitude transformation and autofocus from a wide-angle search position to a high-precision oblique imaging position.

[0012] Preferably, the image processing and control unit performs distortion correction on the image, specifically including:

[0013] Based on the lens distortion coefficients of the visual imaging unit obtained through pre-calibration, the image pixels are corrected; the distortion coefficients include radial distortion coefficients k1, k2, k3 and tangential distortion coefficients p1, p2;

[0014] For normalized image coordinates (x, y), the radial distortion correction formula is:

[0015] ;

[0016] The formula for tangential distortion correction is:

[0017] ; where r 2 =x 2 +y 2 .

[0018] Preferably, the image processing and control unit identifies and extracts edge contour features from the image, and uses an edge detection algorithm that combines visual contour detection with grayscale or color information; for the membrane, the center position is determined by a linear feature extraction algorithm using the strip-shaped region features formed by its oblique imaging.

[0019] Preferably, the image processing and control unit calculates the relative positional deviation, specifically including:

[0020] The planar homography matrix obtained through calibration is applied to the pixel coordinates (u, v) of the center of the standard hydrophone end face and the center of the diaphragm in the image, and converted into physical coordinates (X, v) on the target plane inside the pressure tank. w ,Y w );

[0021] The relative position deviation is obtained by calculating the difference in physical coordinates between the standard hydrophone and the diaphragm.

[0022] A hydrophone calibration and positioning method, applied to a positioning mechanism, the method comprising the following steps:

[0023] S1: Adjust the visual imaging unit to the initial imaging position using the precision motion unit;

[0024] S2: Control the visual imaging unit to acquire a target image containing a standard hydrophone and a diaphragm;

[0025] S3: Perform distortion correction on the target image;

[0026] S4: Extract the edge contour features of the standard hydrophone and the diaphragm from the corrected image, and calculate the relative positional deviation between them;

[0027] S5: Determine whether the relative position deviation is less than a preset threshold;

[0028] If not, a control command is generated based on the deviation to drive the actuator to adjust the position of the standard hydrophone or the diaphragm, and the process returns to step S2; if yes, the positioning is completed.

[0029] Compared with existing technologies, the advantages of this invention are: by using a high-resolution industrial camera and a precision image measurement algorithm, high positioning accuracy is achieved; the non-contact measurement method eliminates the influence of mechanical deformation on positioning accuracy, making it suitable for high-pressure testing environments; the entire positioning process is automatically completed by the vision system for image acquisition, processing, and position calculation, and the actuator is driven by closed-loop control, reducing reliance on operator experience and improving the automation, repeatability, and objectivity of the calibration process. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the overall structure of the laser hydrophone calibration device according to an embodiment of the present invention;

[0031] Figure 2 This is a schematic diagram of the camera imaging optical principle of the present invention;

[0032] Figure 3 This is a schematic diagram illustrating the change in viewing angle of the camera of the present invention when adjusted from a frontal view to an oblique view.

[0033] In the diagram: 1. Pressure tank, 2. Auxiliary transducer, 3. Industrial camera, 4. Electric six-axis displacement platform, 5. Vibration damping base, 6. Laser vibrometer, 7. Light-transmitting window, 8. Observation window, 9. Standard hydrophone, 10. Top cover, 11. Diaphragm. Detailed Implementation

[0034] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings, so that those skilled in the art can more clearly understand how to practice the present invention. Although the present invention has been described in conjunction with its preferred embodiments, these embodiments are merely illustrative and not intended to limit the scope of the invention.

[0035] See Figure 1-3An embodiment of the present invention provides a vision-based hydrophone calibration and positioning mechanism, which is applied to a laser-based hydrophone sound pressure calibration system. The core components of the calibration system include a pressure water tank 1, which has an openable / closable top cover 10 on its top. The tank wall is provided with a light-transmitting window 7 and an observation window 8. An auxiliary transducer 2 for generating a calibration sound field, a standard hydrophone 9 as a measurement reference, and a diaphragm 11 as a sound pressure transmission medium are installed in the internal cavity of the pressure water tank 1. A laser vibrometer 6 is set on the outside of the pressure water tank 1 at a position corresponding to the light-transmitting window 7. The vibrometer is mounted on a vibration damping base 5 to ensure that its measurement beam can be accurately incident on the surface of the diaphragm 11 inside the tank through the light-transmitting window 7. The positioning mechanism described in this application is set outside the pressure water tank 1 and corresponds to the position of the observation window 8 so as to perform optical observation of the standard hydrophone 9 and diaphragm 11 inside the tank through the window; the positioning mechanism is mainly composed of three functional units: a visual imaging unit for image acquisition, a precision motion unit for adjusting the observation posture, and an image processing and control unit for image processing and system control.

[0036] The visual imaging unit includes a high-resolution industrial camera 3 and a matching fixed-focus or zoom optical lens. The lens focal length is adjustable from 50mm to 80mm, and the object distance is 300mm. According to the optical imaging principle, its optical path diagram is as follows: Figure 2 As shown.

[0037] The industrial camera 3 preferably has 151 million pixels to obtain images with sufficiently high spatial resolution. The industrial camera 3 is fixed to the outside of the observation window 8 of the pressure tank by a separate sealed mounting base. Its optical axis can pass through the observation window 8 and be aimed at the target area inside the tank, including the end face of the standard hydrophone and the diaphragm. In order to enhance the imaging effect of the edge of the diaphragm 11, the industrial camera 3 is not installed in the traditional way facing the line of sight, but is designed with a certain angle of view.

[0038] The 151-megapixel industrial area array camera has the following resolutions: 14208 pixels × 10640 pixels, sensor size (60.03mm * 47.90mm), distance U = 300mm between the industrial camera and the standard hydrophone and diaphragm, the standard hydrophone and diaphragm being on the same plane, and a variable focal length range (50mm ~ 80mm).

[0039] Calculations show that:

[0040] Let the focal length F = 50mm, then the magnification N is:

[0041] (1)

[0042] According to formula (1), we can calculate N = 50mm / 300mm = 0.167 times.

[0043] The half-width of the sensor is h = 60.03 / 2mm = 30.015mm. According to the principle of similar triangles, formula (2) can be obtained:

[0044] (2)

[0045] Where V is the image distance, the half width of the field of view H is calculated according to formula (2) = 31.015 / 0.167mm = 185.7 mm.

[0046] The single-pixel precision Q can be calculated based on the camera resolution (14208 pixels × 10640 pixels):

[0047] (3)

[0048] Where p is the number of pixels in half the width of the sensor. According to formula (3), p = 14208 / 2 = 7104 pixels, then Q = 185.7 / 7104 = 0.026mm.

[0049] Let F = 80 mm. Based on the known conditions and formulas (1), (2), and (3), we can calculate Q = 0.016 mm.

[0050] In summary, when the object distance is 300 mm and the focal length is 50 mm to 80 mm, the single-pixel accuracy Q ranges from 0.016 mm to 0.026 mm. Based on the principles of visual imaging, its resolution is generally greater than 2 pixels. Taking 3 pixels as an example, the minimum resolution when using a 151-megapixel industrial camera is 0.016 mm * 3 = 0.048 mm.

[0051] The precision motion unit is used to drive the vision imaging unit to perform multi-degree-of-freedom fine adjustments to achieve optimal imaging state and alignment feedback control. This unit is preferably an electrically driven six-axis displacement platform 4, equipped with three linear motion axes (X, Y, Z) and three rotary motion axes (R). x , R y , R z The industrial camera 3 is securely mounted at the end of a six-axis displacement platform; the six-axis displacement platform is mechanically connected to the vision imaging unit, allowing for precise adjustment of the camera's position and orientation; through this platform, minute structural displacements caused by pressure can be compensated, precisely controlling the camera to achieve an orientation change from "position 1" to "position 2" and completing automatic focusing. The change process and field of view change are as follows: Figure 3 As shown.

[0052] Based on the data provided by the electric six-axis displacement platform, the mapping relationship from a three-dimensional space point P(X,Y,Z) to an image pixel point p(u,v) is as follows:

[0053] (4)

[0054] in:

[0055] (f x , f y ): The focal length (in pixels) along the x / y axis.

[0056] (c x , c y ): Principal point coordinates (image center);

[0057] R (3×3): Rotation matrix;

[0058] t (3×1): Translation vector, collectively called extrinsic parameter;

[0059] s: scale factor.

[0060] The image processing and control unit includes an image acquisition card, an industrial control computer, and a dedicated software system. Its core functions include: image acquisition and preprocessing, feature extraction and position detection, coordinate transformation and positioning calculation, and closed-loop control.

[0061] Image acquisition and preprocessing includes controlling the camera to capture images and performing preprocessing such as filtering and enhancement.

[0062] Camera calibration and distortion correction specifically involve pre-calibrating the camera-lens system with high precision to obtain camera intrinsic parameters (focal length f). x f y Principal point coordinates c x c y The image is analyzed using lens distortion coefficients (radial distortion coefficients k1, k2, k3; tangential distortion coefficients p1, p2). For each acquired image, distortion correction is performed according to the calibration parameters to eliminate the influence of lens optical defects on measurement accuracy. The correction model employs a complete model incorporating both radial and tangential distortion, and the correction formula is as follows:

[0063] 1) Radial distortion (main):

[0064] (5)

[0065] in

[0066] r 2 =x 2 +y 2 (6)

[0067] k1, k2, k3 are radial distortion coefficients;

[0068] 2) Tangential distortion:

[0069] (7)

[0070] Where p1 and p2 are tangential distortion coefficients.

[0071] For feature extraction and location detection, an edge detection algorithm based on visual contour detection combined with color information is used to accurately identify and extract the contour edges of the standard hydrophone end face and the diaphragm from the corrected image. Considering the characteristics of the diaphragm being thin and having potentially low edge contrast, the algorithm can combine its known geometric prior information to perform template matching or Hough transform detection to improve robustness.

[0072] The coordinate transformation and positioning calculation specifically involves converting the pixel coordinates of the feature points identified in the image to the physical coordinate system within the pressure tank. Since the target, i.e., the end face of the hydrophone and the diaphragm, can be considered to be on the same plane, this transformation can be simplified to a planar homography transformation. By pre-calibrating a calibration plate of known size, such as a checkerboard, at the target plane position through imaging, the pixel coordinates (u, v) can be fitted to obtain the physical coordinates (X). w , Y w The transformation relationship of ) can be expressed in matrix form as follows:

[0073] (8)

[0074] Simplified to:

[0075] X w =a·u+b·v+c (9)

[0076] Y w =d·u+e·v+f (10)

[0077] Where a, b, c, d, e, and f are the conversion coefficients obtained after calibration; through this conversion, the coordinates of the hydrophone center and the diaphragm center on the physical plane, as well as the relative positional deviation between them, can be accurately calculated.

[0078] The closed-loop control specifically involves sending the calculated position deviation to the main control system of the calibration device. The main control system can drive the fine-tuning mechanism within the calibration device, which is used to adjust the position of the hydrophone or diaphragm. Based on the deviation signal from the visual feedback, closed-loop adjustment is performed until the position deviation is less than the set threshold, thus completing the precise positioning. The six-axis displacement platform in this application is mainly used to optimize the camera's field of view and imaging quality, while the final alignment of the target object is completed by a dedicated fine-tuning mechanism.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A vision-based hydrophone calibration and positioning mechanism, applied in a laser-based hydrophone sound pressure calibration system, the calibration system comprising a pressure tank (1) and a standard hydrophone (9) and a diaphragm (11) disposed therein, the pressure tank (1) having an observation window, characterized in that, The positioning mechanism includes: A visual imaging unit is located outside the pressure tank (1), with its optical axis passing through the observation window and aimed at the target area inside the tank, for acquiring images including the end face of the standard hydrophone (9) and the diaphragm (11); A precision motion unit is driven and connected to the visual imaging unit to adjust the spatial position and attitude of the visual imaging unit. An image processing and control unit is signal-connected to the visual imaging unit and the precision motion unit. The image processing and control unit is configured to: perform distortion correction on the image acquired by the visual imaging unit; identify and extract the edge contour features of the end face of the standard hydrophone (9) and the diaphragm (11) from the corrected image; calculate the relative position deviation between the standard hydrophone (9) and the diaphragm (11) based on the edge contour features; and output a control signal based on the relative position deviation to drive the actuator in the calibration system for adjusting the position of the standard hydrophone (9) or the diaphragm (11) until the relative position deviation is less than a preset threshold.

2. The vision-based hydrophone calibration and positioning mechanism according to claim 1, characterized in that: The visual imaging unit includes a high-resolution industrial area array camera (3) and an optical lens; the industrial camera (3) has a pixel count of not less than 100 million pixels; the optical lens has an adjustable focal length or is a fixed focal length lens.

3. The vision-based hydrophone calibration and positioning mechanism according to claim 2, characterized in that: The visual imaging unit adopts an oblique imaging layout, such that the optical axis of the industrial camera (3) forms an angle with the normal of the plane containing the end face of the standard hydrophone (9) and the diaphragm (11), and the angle is acute.

4. The vision-based hydrophone calibration and positioning mechanism according to claim 3, characterized in that: The precision motion unit includes an electric six-axis displacement platform (4), and the industrial camera (3) is installed on the execution end of the six-axis displacement platform (4) to drive the industrial camera (3) to perform translation and rotation in three-dimensional space, so as to realize attitude transformation and autofocus from wide-angle search position to high-precision oblique imaging position.

5. The vision-based hydrophone calibration and positioning mechanism according to claim 1, characterized in that: The image processing and control unit performs distortion correction on the image, specifically including: Based on the lens distortion coefficients of the visual imaging unit obtained through pre-calibration, the image pixels are corrected; the distortion coefficients include radial distortion coefficients k1, k2, k3 and tangential distortion coefficients p1, p2; For normalized image coordinates (x, y), the radial distortion correction formula is: ； The formula for tangential distortion correction is: ; where r 2 =x 2 +y 2 .

6. The vision-based hydrophone calibration and positioning mechanism according to claim 1, characterized in that: The image processing and control unit identifies and extracts edge contour features from the image, and uses an edge detection algorithm that combines visual contour detection with grayscale or color information; for the membrane (11), it uses the strip-shaped region features formed by its oblique imaging to determine its center position through a linear feature extraction algorithm.

7. The vision-based hydrophone calibration and positioning mechanism according to claim 1, characterized in that: The image processing and control unit calculates the relative position deviation, specifically including: The planar homography matrix obtained through calibration is applied to the pixel coordinates (u, v) of the center of the end face of the standard hydrophone (9) and the center of the diaphragm (11) in the image, and converted into physical coordinates (X, v) on the target plane inside the pressure tank (1). w , Y w ); The physical coordinate difference between the standard hydrophone (9) and the diaphragm (11) is calculated to obtain the relative position deviation.

8. A hydrophone calibration and positioning method, applied to the positioning mechanism as described in any one of claims 1-7, characterized in that, The method includes the following steps: S1: Adjust the visual imaging unit to the initial imaging position using the precision motion unit (4); S2: Control the visual imaging unit to acquire a target image containing a standard hydrophone (9) and a diaphragm (11); S3: Perform distortion correction on the target image; S4: Extract the edge contour features of the standard hydrophone (9) and the diaphragm (11) from the corrected image, and calculate the relative positional deviation between them; S5: Determine whether the relative position deviation is less than a preset threshold; If not, a control command is generated based on the deviation to drive the actuator to adjust the position of the standard hydrophone (9) or the diaphragm (11) and return to step S2; if yes, the positioning is completed.

Images