A three-dimensional sound velocity imaging method using a linear array ultrasonic sensor array

Abstract

The application discloses a three-dimensional sound velocity imaging method using a linear array ultrasonic sensor array, and comprises the following steps: calibrating a calibration plate; calibrating a reflection plate, that is, arranging the reflection plate directly below the linear array ultrasonic sensor, and calibrating the positional relationship between the reflection plate and the linear array ultrasonic sensor; collecting data and simultaneously recording the positional change of the linear array ultrasonic sensor by using a camera; transforming the collected data to the same coordinate system according to the transformation relationship between the calibration plates, obtaining the collected data in the same coordinate system; mapping the collected data in the same coordinate system to the three-dimensional space through coordinate transformation, performing three-dimensional sound velocity imaging, and completing the three-dimensional sound velocity imaging using the linear array ultrasonic sensor array. The application realizes a three-dimensional sound velocity imaging method by using a low-cost and easily-obtained linear array ultrasonic sensor array, and solves the problem that the current three-dimensional sound velocity imaging is high in cost and difficult to use due to the use of a ring-shaped sensor array.

Description

Technical Field

[0001] This invention relates to a three-dimensional sound velocity imaging method, and more particularly to a three-dimensional sound velocity imaging method using a linear array of ultrasonic sensors. Background Technology

[0002] Ultrasonic velocity imaging is an imaging technique that uses sound velocity as a marker. It can distinguish targets that are difficult to distinguish using ultrasound imaging due to their similar acoustic impedance to the background. Ultrasonic velocity imaging requires large-angle data. Traditional ultrasonic velocity imaging methods use ring sensor arrays to acquire signals from multiple angles; however, ring sensor arrays are relatively expensive and have limited applicability, resulting in low adoption rates. Using common handheld linear array probes, the reconstructed two-dimensional ultrasonic velocity images are relatively poor because they cannot acquire large-angle data. Three-dimensional velocity imaging is a technique that maps the spatial location of the acquired data into three-dimensional space to achieve a three-dimensional image of the target. This process also requires large-angle data, making it difficult to use linear array probes. Summary of the Invention

[0003] Purpose of the invention: The technical problem to be solved by the present invention is to provide a three-dimensional sound velocity imaging method using a linear array of ultrasonic sensors, which addresses the shortcomings of the prior art.

[0004] To address the aforementioned technical problems, this invention discloses a three-dimensional sound velocity imaging method employing a linear array of ultrasonic sensors, comprising the following steps:

[0005] Step 1: Calibrate the calibration plate;

[0006] Step 2: Calibrate the reflector, that is, set the reflector directly below the linear ultrasonic sensor and calibrate the positional relationship between the reflector and the linear ultrasonic sensor.

[0007] Step 3: Collect data and simultaneously record the positional changes of the linear ultrasonic sensor using a camera;

[0008] Step 4: Based on the transformation relationship between the calibration plates, transform the collected data to the same coordinate system to obtain the collected data in the same coordinate system;

[0009] Step 5: Map the acquired data in the same coordinate system to three-dimensional space through coordinate transformation to perform three-dimensional sound velocity imaging, thus completing the three-dimensional sound velocity imaging using a linear array of ultrasonic sensors.

[0010] Beneficial effects:

[0011] This invention proposes a method for three-dimensional sound velocity imaging of a target using a portable and readily available linear ultrasonic sensor array. Furthermore, through an adjustable reflector bracket, the data acquisition is not limited by the size of the target being imaged, which is more conducive to the promotion of sound velocity imaging and fills the gap in three-dimensional sound velocity imaging methods using linear ultrasonic sensor arrays. Attached Figure Description

[0012] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0013] Figure 1 This is a flowchart illustrating the present invention.

[0014] Figure 2 It is a coordinate system reference diagram for calibrating a multi-faceted calibration plate.

[0015] Figure 3 This is a geometric schematic diagram of the calibration reflector and linear ultrasonic sensor.

[0016] Figure 4 This is a diagram illustrating data acquisition.

[0017] Figure 5 This is a schematic diagram of three-dimensional sound velocity reconstruction. Detailed Implementation

[0018] A method for reconstructing three-dimensional sound velocity using a linear array of ultrasonic sensors includes the following steps:

[0019] Step 1: Calibrate the multi-faceted calibration plate;

[0020] The method for calibrating the multi-faceted calibration plate includes:

[0021] Step 1-1: Calibration of the reference plane. The multi-faceted calibration plate consists of three planar calibration plates joined at a 120° angle and fixed to the linear array ultrasonic sensor. The lower right calibration plate is selected as the reference plane. A 3D-printed N-line model is used (reference: Carbajal, G., Lasso, A., Gómez). (Reference: Carbajal, G., Lasso, A., Gómez, et al. Improving N-wire phantom-based freehand ultrasound calibration. Int J CARS 8, 1063–1072 (2013)) This model is used to calibrate the positional relationship between a reference plane and a linear array ultrasound sensor. The N-wire model consists of two or more layers of N-shaped wires and a calibration plate fixed to the model (Reference: Carbajal, G., Lasso, A., Gómez, et al. Improving N-wire phantom-based freehand ultrasound calibration. Int J CARS 8, 1063–1072 (2013)). et al. Improving N-wirephantom-based freehand ultrasound calibration. Int J CARS 8, 1063–1072 (2013)). The linear array ultrasonic sensor mentioned in the paper refers to a linear array ultrasonic sensor array, which is a linear array ultrasonic sensor arranged in an array manner.

[0022] Let the camera coordinate system be denoted as {C}; the world coordinate system corresponding to the reference plane be denoted as {W}; let the N-line model take its bottom point as the origin of the coordinate system, and denote it as the coordinate system {P}; establish a coordinate system with the top left corner of the ultrasound image as the origin, the horizontal axis as the x-axis, and the vertical axis as the y-axis, and then extend it to three dimensions, and denote it as the ultrasound image coordinate system {T}.

[0023] The transformation matrix from the world coordinate system {W} to the camera coordinate system {C} is obtained using Zhang Zhengyou's chessboard calibration method (reference: Z. Zhang, "A flexible new technique for camera calibration," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 11, pp. 1330-1334, Nov. 2000). Then, the transformation matrix between the camera coordinate system {C} and the N-line model coordinate system {P} is determined using a calibration plate on the N-line model. In coordinate system {P}, given the precise coordinates of each N-line in {P}, scanning the N-line model along a direction perpendicular to the N-lines will result in several bright spots appearing as N-lines in the ultrasound image. The positions of these points in the ultrasound image coordinate system {T} are...

[0024]

[0025] Where s u and s v These represent the actual physical dimensions of each pixel along the X and Y axes in the ultrasound image, respectively; i represents the layer number of the bright spot, and j is the index of that layer of bright spots, (u ij ,v ij Let be the pixel coordinates of the bright spot in the ultrasound image. Therefore, the positions of these bright spots in the coordinate system {P} are represented as:

[0026]

[0027] Where x ij ,y ij and z ijThese are the coordinates of a point along the X, Y, and Z axes in coordinate system {P}, respectively. The transformation matrix between {W} and {T} can be obtained using the gradient descent method. This completes the calibration of the reference surface.

[0028] Steps 1-2 involve calibrating the remaining surfaces. Let the world coordinate system of the reference surface be {W0}, and the world coordinate system corresponding to the I-th calibration surface be {W... I},have:

[0029]

[0030] in, This represents the ultrasound image coordinate system {T} and the world coordinate system {W} of the i-th calibration plane. I The transformation matrix between} Represents the world coordinate system {W0} of the datum plane and the world coordinate system {W} of the i-th calibration plane. I The transformation matrix between} The transformation matrix between the ultrasound image coordinate system {T} and the reference plane world coordinate system {W0} is represented.

[0031] Adjust the position of the multi-faceted calibration board relative to the camera so that the reference plane and the other planes to be calibrated appear together in the camera's field of view. Then, call the OpenCV-encapsulated function for detecting the calibration board to calculate the... This completes the calibration of the remaining calibration surfaces.

[0032] Step 2: Place a reflector directly below the linear array ultrasonic sensor and calibrate the positional relationship between the reflector and the linear array probe;

[0033] The method of placing a reflector directly below the linear array ultrasonic sensor includes:

[0034] An L-shaped bracket is fixed to the linear ultrasonic sensor, and the bracket has a sliding reflector slider. The reflector slider is adjusted to a suitable position according to the size of the imaging target, that is, adjusted according to the size of the target, so that the target is exactly in the middle of the ultrasonic sensor and the reflector, approximately equal to the height / length of the target, and then fixed with screws.

[0035] The method for determining the positional relationship between the calibrated reflector and the linear ultrasonic sensor includes:

[0036] The linear ultrasonic sensor and reflector are placed together in a temperature-controlled water tank. Each sensor in the linear ultrasonic sensor array emits an ultrasonic signal individually and sequentially, which is then received by all other sensors. The self-emitted and self-received signals from the first and last sensors are collected, and the time points when the reflected signals are received are calculated by taking the envelope. This allows the propagation times t1 and t2 of the ultrasonic signals to be determined, and subsequently, the distances d1 and d2 between these two sensors and the reflector are calculated. The distance between two adjacent sensors of the linear array probe is used as the grid length, and the vertical plane between the linear ultrasonic sensor array and the reflector is divided into equidistant grids. To address the potential issue of the linear ultrasonic sensor array and the reflector array not being parallel, the position of the linear ultrasonic sensor array is corrected.

[0037] Taking the imaging plane of the linear ultrasonic sensor, the reflector and the linear ultrasonic sensor can each be represented by a line segment. Assume the linear ultrasonic sensor was originally parallel to the reflector, and then rotated by an angle θ around the position of the first sensor, denoted as line segment AB. Establish a plane coordinate system XOY, with point B as the origin, the x-axis parallel to the reflector, and the y-axis perpendicular to the reflector. Place a sensor C on line segment AB. The acoustic signal emitted by C is reflected by the reflector and received by sensor D on line segment AB. The reflection point on the reflector is denoted as point G. The propagation path of this acoustic signal intersects the original linear ultrasonic sensor at points E and F, respectively. Draw a perpendicular line CP from point C to the reflector, intersecting the original linear ultrasonic sensor at point M; draw a perpendicular line DQ from point D to the reflector, intersecting the original linear ultrasonic sensor at point N. Let AD = L1 and AC = L2. From plane geometry, we can calculate:

[0038]

[0039]

[0040] Where GQ is the line segment connecting points G and Q; from this, the coordinates of point C are ((AB-L2)sinθ,(AB-L2)cosθ), the coordinates of point D are ((AB-L1)sinθ,(AB-L1)cosθ), and the coordinates of point G are (GQ,d2), which are the actual coordinates of the positions of each sensor and the reflection point.

[0041] Step 3: While collecting data, use a camera to record the positional changes of the linear array probe;

[0042] During data acquisition, the positions of the camera and the target remain fixed. The target is positioned between the linear array probe and the reflector. Then, the linear array probe is slowly moved and rotated in various directions to acquire data from as many angles as possible. The camera calibration program first calculates the rotation matrix R of each calibration plate in the first frame of the captured image, and then...

[0043] zproj =(R*[0 0 1]) T )·[0 0 1] T

[0044] The projection matrix of each calibration plate along the z-axis at the current position can be calculated, and the calibration plate with the smallest tilt along the z-axis is selected as the current actual calibration plate, until the calibration plate can no longer be captured by the camera. The camera calibration program calibrates each calibration plate image acquired in real time and outputs the rotation and translation matrix of the corresponding position. Let the rotation and translation matrix of the calibration plate relative to the camera for the m-th image be denoted as... and When collecting data, the camera emits a trigger signal for each photo it takes. The data acquisition program receives the signal and begins a data acquisition cycle, thus achieving synchronization between camera shooting and data acquisition.

[0045] Step 4: Based on the transformation relationship between the calibration plates, transform the collected data to the same coordinate system;

[0046] The world coordinate system {W0} of the reference surface in step 1 and the world coordinate system {W} of the I-th calibration surface I Transformation matrix between} Decompose into rotation matrix Translation matrix This yields the rotation matrix of the ultrasound image coordinate system relative to the world coordinate system of the I-th calibration plate. Translation matrix They are respectively:

[0047]

[0048]

[0049] in and These represent the rotation and translation matrices of the ultrasound image coordinate system relative to the world coordinate system of the reference calibration plate, respectively. For a point [p, q, 0] in the ultrasound image coordinate system... T Based on the calibration plate number selected during calibration, the formula is used.

[0050]

[0051] They can be uniformly transformed to the camera coordinate system, where and Let be the rotation and translation matrix of the calibration board relative to the camera for the nth acquired image.

[0052] Step 5: Map the collected data into three-dimensional space to perform three-dimensional sound velocity imaging.

[0053] Methods for mapping the collected data into three-dimensional space include:

[0054] The three-dimensional space is discretized into a three-dimensional grid with equal side lengths. Based on the grid of the linear ultrasonic sensor and reflector calibrated in step 2, and the position of the acquired data in the camera coordinate system obtained in step 4, each planar grid point in the two-dimensional data acquisition is mapped to a three-dimensional space grid point.

[0055] The three-dimensional sound velocity imaging method includes:

[0056] For each acquired signal, the time of flight (ToF) is calculated using the envelope method. Each time of flight corresponds to a spatial propagation path. The speed of sound is then iteratively solved using the SART algorithm (reference: Andersen AH, Kak AC. Simultaneous algebraic reconstruction technique (SART): a superior implementation of the SART algorithm. Ultrason Imaging. 1984 Jan).

[0057]

[0058] Where, σ k This is the three-dimensional spatial sound velocity distribution after the k-th iteration, t i h is the flight time of the i-th data set. r It is the spatial propagation path corresponding to the r-th group of flight times, a k λ is the speed of sound in the grid through which the propagation path passes after the k-th iteration. k It is the weight coefficient for the k-th iteration. The reciprocal of the sound velocity of constant-temperature water is taken as the initial value for iteration. When the difference in sound velocity in each iteration is less than the set threshold, it is considered that the iteration has converged. Finally, different sound velocity values ​​are mapped to different colors and displayed in three-dimensional space.

[0059] Example:

[0060] This embodiment describes a three-dimensional sound velocity reconstruction method using a linear array of ultrasonic sensors, such as... Figure 1 As shown, it includes the following steps:

[0061] Step 1: Calibrate the multi-faceted calibration plate;

[0062] The method for calibrating the multi-faceted calibration plate includes:

[0063] Step 1-1: Calibration of the reference plane. The multi-faceted calibration plate consists of three planar calibration plates joined at a 120° angle and fixed to the linear array ultrasonic sensor. The calibration plate at the lower right is selected as the reference plane. A 3D-printed N-line model is used to calibrate the positional relationship between the reference plane and the linear array ultrasonic sensor. The N-line model consists of two or more layers of N-shaped lines and calibration plates fixed to the model (Reference: Carbajal, G., Lasso, A., Gómez). et al. Improving N-wire phantom-based freehand ultrasound calibration. Int J CARS8,1063–1072(2013)).

[0064] like Figure 2 As shown, the camera coordinate system is denoted as {C}; the world coordinate system corresponding to the reference plane is denoted as {W}; let the N-line model take its bottom point as the origin of the coordinate system, and denote it as the coordinate system {P}; take the point at the upper left corner of the ultrasound image as the origin of the coordinate system, the horizontal axis as the x-axis, and the vertical axis as the y-axis to establish a coordinate system, and then extend it to three dimensions, which is denoted as the ultrasound image coordinate system {T}.

[0065] The transformation matrix from the world coordinate system {W} to the camera coordinate system {C} is obtained using Zhang Zhengyou's chessboard calibration method (reference: Z. Zhang, "A flexible new technique for camera calibration," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 11, pp. 1330-1334, Nov. 2000). Then, the transformation matrix between the camera coordinate system {C} and the N-line model coordinate system {P} is determined using a calibration plate on the N-line model. In coordinate system {P}, given the precise coordinates of each N-line in {P}, scanning the N-line model along a direction perpendicular to the N-lines will result in several bright spots appearing as N-lines in the ultrasound image. The positions of these points in the ultrasound image coordinate system {T} are...

[0066]

[0067] Where s u and s v These represent the actual physical dimensions of each pixel along the X and Y axes in the ultrasound image, respectively; i represents the layer number of the bright spot, and j is the index of that layer of bright spots, (u ij ,v ijLet be the pixel coordinates of the bright spot in the ultrasound image. Therefore, the positions of these bright spots in the coordinate system {P} are represented as:

[0068]

[0069] Where x ij ,y ij and z ij These are the coordinates of the point along the X-axis, Y-axis, and Z-axis in the coordinate system {P}, respectively.

[0070] The transformation matrix between {W} and {T} can be obtained using the gradient descent method. This completes the calibration of the reference surface.

[0071] Steps 1-2 involve calibrating the remaining surfaces. Let the world coordinate system of the reference surface be {W0}, and the world coordinate system corresponding to the I-th calibration surface be {W... I},have:

[0072]

[0073] in, This represents the ultrasound image coordinate system {T} and the world coordinate system {W} of the i-th calibration plane. I The transformation matrix between} Represents the world coordinate system {W0} of the datum plane and the world coordinate system {W} of the i-th calibration plane. I The transformation matrix between} The transformation matrix between the ultrasound image coordinate system {T} and the reference plane world coordinate system {W0} is represented.

[0074] Adjust the position of the multi-faceted calibration board relative to the camera so that the reference plane and the other planes to be calibrated appear together in the camera's field of view. Then, call the OpenCV-encapsulated function for detecting the calibration board to calculate the... This completes the calibration of the remaining calibration surfaces.

[0075] Step 2: Place a reflector directly below the linear array ultrasonic sensor and calibrate the positional relationship between the reflector and the linear array probe;

[0076] The method of placing a reflector directly below the linear array ultrasonic sensor includes:

[0077] An L-shaped bracket is fixed to the linear ultrasonic sensor, and the bracket has a sliding reflector slider made of a specific material. The reflector slider is adjusted to a suitable position according to the size of the imaging target and then secured with screws.

[0078] The method for determining the positional relationship between the calibrated reflector and the linear ultrasonic sensor includes:

[0079] The linear ultrasonic sensor and reflector are placed together in a temperature-controlled water tank. Each sensor in the linear ultrasonic sensor array emits an ultrasonic signal sequentially and individually, which is then received by all other sensors. The signals emitted and received by the first and last sensors are collected, and the time points when the reflected signals are received are calculated by taking the envelope. This allows the propagation times t1 and t2 of the ultrasonic signals to be determined, and subsequently, the distances d1 and d2 between these two sensors and the reflector are calculated. The distance between two adjacent sensors of the linear array probe is used as the grid length, and the vertical plane between the linear ultrasonic sensor array and the reflector is divided into equidistant grids. To address the potential issue of the linear ultrasonic sensor array and the reflector array not being parallel, the position of the linear ultrasonic sensor array is corrected. Figure 3 As shown, taking the imaging plane of the linear ultrasonic sensor, the reflector and the linear ultrasonic sensor can each be represented by a line segment. Assume the linear ultrasonic sensor was originally parallel to the reflector, and then rotated by an angle θ around the position of the first sensor, denoted as line segment AB. Take point B as the origin, the x-axis parallel to the reflector, and the y-axis perpendicular to the reflector, establishing a plane coordinate system XOY. Take a sensor C on line segment AB. The acoustic signal emitted by C is reflected by the reflector and received by sensor D on line segment AB. The reflection point on the reflector is denoted as point G. The propagation path of this acoustic signal intersects the original linear ultrasonic sensor at points E and F, respectively. Draw a perpendicular line CP from point C to the reflector, intersecting the original linear ultrasonic sensor at point M; draw a perpendicular line DQ from point D to the reflector, intersecting the original linear ultrasonic sensor at point N. Let AD = L1, AC = L2. From plane geometry, we can calculate:

[0080]

[0081]

[0082] Where GQ is the line segment connecting points G and Q; from this, the coordinates of point C are ((AB-L2)sinθ,(AB-L2)cosθ), the coordinates of point D are ((AB-L1)sinθ,(AB-L1)cosθ), and the coordinates of point G are (GQ,d2), which are the actual coordinates of the positions of each sensor and the reflection point.

[0083] Step 3: While collecting data, use a camera to record the positional changes of the linear array probe;

[0084] like Figure 4 As shown, during data acquisition, the positions of the fixed camera and the target remain unchanged. The target is placed between the linear array probe and the reflector. Then, the linear array probe is slowly moved and rotated in various directions to acquire data from as many angles as possible. The camera calibration program first calculates the rotation matrix R of each calibration plate in the first frame of the captured image, and then...

[0085] zproj =(R*[0 0 1]) T )·[0 0 1] T

[0086] The projection matrix of each calibration plate along the z-axis at the current position can be calculated, and the calibration plate with the smallest tilt along the z-axis is selected as the current actual calibration plate, until the calibration plate can no longer be captured by the camera. The camera calibration program calibrates each calibration plate image acquired in real time and outputs the rotation and translation matrix of the corresponding position. Let the rotation and translation matrix of the calibration plate relative to the camera for the m-th image be denoted as... and When collecting data, the camera emits a trigger signal for each photo it takes. The data acquisition program receives the signal and begins a data acquisition cycle, thus achieving synchronization between camera shooting and data acquisition.

[0087] Step 4: Based on the transformation relationship between the calibration plates, transform the collected data to the same coordinate system;

[0088] The world coordinate system {W0} of the reference surface in step 1 and the world coordinate system {W} of the I-th calibration surface I Transformation matrix between} Decompose into rotation matrix Translation matrix This yields the rotation matrix of the ultrasound image coordinate system relative to the world coordinate system of the I-th calibration plate. Translation matrix They are respectively:

[0089]

[0090]

[0091] in and These represent the rotation and translation matrices of the ultrasound image coordinate system relative to the world coordinate system of the reference calibration plate, respectively. For a point [p, q, 0] in the ultrasound image coordinate system... T Based on the calibration plate number selected during calibration, the formula is used.

[0092]

[0093] They can be uniformly transformed to the camera coordinate system, where and Let be the rotation and translation matrix of the calibration board relative to the camera for the nth acquired image.

[0094] Step 5: Map the collected data into three-dimensional space to perform three-dimensional sound velocity imaging.

[0095] Methods for mapping the collected data into three-dimensional space include:

[0096] The three-dimensional space is discretized into a three-dimensional grid with equal side length. Based on the grid of the linear ultrasonic sensor and the reflector calibrated in step 2, and the position of the acquired data in the camera coordinate system obtained in step 4, each planar grid point in the two-dimensional data acquisition is mapped to a three-dimensional space grid point.

[0097] The three-dimensional sound velocity imaging method includes:

[0098] like Figure 5 As shown, for each acquired signal, the time of flight (ToF) is calculated using the envelope method. Each time of flight corresponds to a spatial propagation path. The speed of sound is iteratively solved using the SART algorithm (reference: Andersen AH, Kak AC. Simultaneous algebraic reconstruction technique (SART): a superior implementation of the art algorithm. Ultrason Imaging. 1984 Jan).

[0099]

[0100] Where, σ k This is the three-dimensional spatial sound velocity distribution after the k-th iteration, t i h is the flight time of the i-th data set. r It is the spatial propagation path corresponding to the r-th group of flight times, a k λ is the speed of sound in the grid through which the propagation path passes after the k-th iteration. k It is the weight coefficient for the k-th iteration. The reciprocal of the sound velocity of constant-temperature water is taken as the initial value for iteration. When the difference in sound velocity in each iteration is less than the set threshold, it is considered that the iteration has converged. Finally, different sound velocity values ​​are mapped to different colors and displayed in three-dimensional space.

[0101] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding a three-dimensional acoustic velocity imaging method using a linear array of ultrasonic sensors, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0102] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MUU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0103] This invention provides a concept and method for three-dimensional sound velocity imaging using a linear array of ultrasonic sensors. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment of the invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A three-dimensional acoustic velocity imaging method employing a linear array of ultrasonic sensors, characterized in that, Includes the following steps: Step 1: Calibrate the calibration plate; Step 2: Calibrate the reflector, that is, set the reflector directly below the linear ultrasonic sensor and calibrate the positional relationship between the reflector and the linear ultrasonic sensor. Step 3: Collect data and simultaneously record the positional changes of the linear ultrasonic sensor using a camera; Step 4: Based on the transformation relationship between the calibration plates, transform the collected data to the same coordinate system to obtain the collected data in the same coordinate system; Step 5: Map the acquired data in the same coordinate system to three-dimensional space through coordinate transformation to perform three-dimensional sound velocity imaging, thus completing the three-dimensional sound velocity imaging using a linear array of ultrasonic sensors. The calibration plate mentioned in step 1 is a multi-faceted calibration plate, which is composed of three planar calibration plates joined together at a 120° angle and fixed on the linear ultrasonic sensor. One calibration plate is tilted upwards, and the other two calibration plates are horizontal to the left and horizontal to the right, respectively. The calibration of the calibration plate described in step 1 includes: Step 1-1, calibrate the reference plane, specifically including: The calibration plate in the lower right corner is selected as the reference plane. The N-line model is used to calibrate the positional relationship between the reference plane and the linear ultrasound sensor. The N-line model consists of two or more layers of N-shaped lines and a calibration plate fixed on the model. The camera coordinate system is denoted as {C}. The world coordinate system corresponding to the reference plane is denoted as {W}. The N-line model is set with its bottom point as the origin, which is denoted as the N-line model coordinate system {P}. The upper left corner of the ultrasound image acquired by the linear ultrasound sensor is used as the origin, with the horizontal axis as the X-axis and the vertical axis as the Y-axis to establish a coordinate system. This system is then extended to three dimensions and denoted as the ultrasound image coordinate system {T}. The transformation matrix from the world coordinate system {W} to the camera coordinate system {C} is obtained using Zhang Zhengyou's chessboard calibration method. Then, the transformation matrix between the camera coordinate system {C} and the N-line model coordinate system {P} is determined using a calibration plate on the N-line model. In the N-line model coordinate system {P}, given the coordinates of each N-line in the N-line model coordinate system {P}, scanning the N-line model along a direction perpendicular to the N-lines will cause the N-lines to appear as bright spots in the ultrasound image. The positions of these points in the ultrasound image coordinate system {T} are determined. for: Among them, s u and s v These represent the actual physical dimensions of each pixel along the X and Y axes in the ultrasound image, respectively; i represents the layer number of the bright spot, and j is the index of that layer of bright spots, (u ij ,v ij ) represents the pixel coordinates of the bright spot in the ultrasound image; the position of the bright spot in the N-line model coordinate system {P} is... Represented as: Where, x ij y ij and z ij These are the coordinates of the bright spot along the X, Y, and Z axes in the N-line model coordinate system {P}; the transformation matrix between the world coordinate system {W} and the ultrasound image coordinate system {T} is obtained using the gradient descent method. This completes the calibration of the reference surface; Steps 1-2 involve calibrating the remaining faces, specifically including: Let the world coordinate system of the reference surface be {W0}, and the world coordinate system corresponding to the I-th calibration surface be {W I },but: in, This represents the ultrasound image coordinate system {T} and the world coordinate system {W} of the i-th calibration plane. I The transformation matrix between} Represents the world coordinate system {W0} of the datum plane and the world coordinate system {W} of the i-th calibration plane. I The transformation matrix between} The transformation matrix between the ultrasound image coordinate system {T} and the reference plane world coordinate system {W0} is represented. Adjust the position of the multi-faceted calibration plate relative to the camera so that the reference plane and the other planes appear together in the camera's field of view, and then calculate... That is, complete the calibration of the remaining calibration surfaces; The method for placing a reflector directly below the linear array ultrasonic sensor in step 2 includes: An L-shaped bracket is fixedly installed below the linear array ultrasonic sensor, and a reflector slider is installed on the bracket; the position of the reflector slider is adjusted according to the size of the imaging target so that the target is located between the linear array ultrasonic sensor array and the reflector, and then fixed. The method for calibrating the positional relationship between the reflector and the linear ultrasonic sensor in step 2 includes: The linear ultrasonic sensor and the reflector are placed together in a constant-temperature water tank and immersed in water. Each sensor of the linear ultrasonic sensor emits an ultrasonic signal sequentially and is received by all sensors. The ultrasonic signals emitted and received by the first and last sensors are collected, and the time points when the reflected signals are received are calculated by taking the envelope. The propagation times t1 and t2 of the ultrasonic signals are calculated, and the distances d1 and d2 between these two sensors and the reflector are further calculated. The distance between two adjacent sensors of the linear ultrasonic sensor is taken as the grid length, and the vertical plane between the linear ultrasonic sensor and the reflector is divided into equidistant grids, i.e., the linear ultrasonic sensor grid and the reflector grid. To address the parallelism issue between the linear ultrasonic sensor and the reflector, the position of the linear ultrasonic sensor is corrected. Taking the imaging plane of the linear ultrasonic sensor, the reflector and the linear ultrasonic sensor are represented by line segments. Assume the linear ultrasonic sensor is initially parallel to the reflector, and then rotates by an angle θ around the position of the first sensor. Let the linear ultrasonic sensor be line segment AB. Establish a plane coordinate system XOY, with point B as the origin, the x-axis parallel to the reflector, and the y-axis perpendicular to the reflector. Take any sensor, the first sensor C, on line segment AB. The acoustic signal emitted by the first sensor C is reflected by the reflector and received by the second sensor D on line segment AB. The reflection point on the reflector is denoted as point G. The propagation path of this acoustic signal intersects the original linear ultrasonic sensor at points E and F, respectively. Draw a perpendicular line CP from point C to the reflector, intersecting the original linear ultrasonic sensor at point M. Draw a perpendicular line DQ from point D to the reflector, intersecting the original linear ultrasonic sensor at point N. Let AD = L1 and AC = L2. From plane geometry, we can calculate: Where GQ is the length of the line segment connecting points G and Q; from this, the coordinates of point C are ((AB-L2)sinθ, (AB-L2)cosθ), which is the position of the first sensor C in the linear array ultrasonic sensor; the coordinates of point D are ((AB-L1)sinθ, (AB-L1)cosθ), which is the position of the second sensor D in the linear array ultrasonic sensor; and the coordinates of point G are (GQ, d2), which is the actual coordinates of the reflection point.

2. The three-dimensional sound velocity imaging method using a linear array of ultrasonic sensors according to claim 1, characterized in that, The data collected in step 3 includes: With the positions of the camera and the target unchanged, the target is placed between the linear array probe and the reflector in the linear array ultrasonic sensor. The linear array probe is then slowly moved and rotated in various directions to acquire data. The rotation matrix R of each calibration plate is calculated from the first frame captured by the camera. Then, the projection matrix z of each calibration plate along the z-axis at the current position is calculated. proj The method is as follows: z proj =(R*[0 0 1] T )·[0 0 1] T The calibration board with the smallest tilt along the z-axis is selected as the current calibration board for actual use, until the calibration board can no longer be captured by the camera; the camera calibrates each captured image of the calibration board and outputs the rotation and translation matrix of the corresponding position. Let the rotation matrix of the calibration board relative to the camera for the m-th captured image be denoted as... The translation matrix is 3. The three-dimensional sound velocity imaging method using a linear array of ultrasonic sensors according to claim 2, characterized in that, Step 3, data collection, also includes synchronizing camera shooting and data acquisition. The specific method is as follows: While collecting data, the camera emits a trigger signal when taking each photo. Upon receiving the trigger signal, a data acquisition cycle begins, thus synchronizing camera shooting and data acquisition.

4. The three-dimensional sound velocity imaging method using a linear array of ultrasonic sensors according to claim 3, characterized in that, Step 4, which involves transforming the collected data to the same coordinate system, includes: The world coordinate system {W0} of the reference surface in step 1 and the world coordinate system {W} of the I-th calibration surface I Transformation matrix between} Decompose into rotation matrix Translation matrix This yields the rotation matrix of the ultrasound image coordinate system relative to the world coordinate system of the I-th calibration plate. Translation matrix They are respectively: in, and These represent the rotation and translation matrices of the ultrasound image coordinate system relative to the world coordinate system of the reference calibration plate, respectively; and the coordinates of a point [p, q, 0] in the ultrasound image coordinate system. T Based on the calibration plate number selected during calibration, the data is uniformly transformed to the camera coordinate system to obtain the acquired data in the same coordinate system. Specifically, it includes: in, and Let be the rotation and translation matrix of the calibration board relative to the camera for the nth acquired image.

5. A three-dimensional sound velocity imaging method using a linear array of ultrasonic sensors according to claim 4, characterized in that, Step 5 describes mapping the acquired data from the same coordinate system to three-dimensional space through coordinate transformation. The method includes: The three-dimensional space is discretized into a three-dimensional grid with equal side length. Based on the grid of the linear ultrasonic sensor and the reflector calibrated in step 2, and the position of the acquired data in the same coordinate system obtained in step 4 in the camera coordinate system, each planar grid point in the acquired data in the same coordinate system is mapped to a grid point in the three-dimensional space.

6. The three-dimensional sound velocity imaging method using a linear array of ultrasonic sensors according to claim 5, characterized in that, The method for performing three-dimensional sound velocity imaging as described in step 5 includes: For each signal acquired by the linear array ultrasonic sensor, the time of flight (ToF) of signal propagation is calculated using the envelope method. Each time of flight corresponds to a spatial propagation path, and the speed of sound is iteratively solved using the SART algorithm. Where, σ k+1 This is the three-dimensional spatial sound speed distribution after the (k+1)th iteration, σ k This is the three-dimensional spatial sound speed distribution after the k-th iteration, t r h is the flight time of the r-th data set. r It is the spatial propagation path corresponding to the r-th group of flight times, a k λ is the speed of sound in the grid through which the propagation path passes after the k-th iteration. k It is the weight coefficient for the k-th iteration; the reciprocal of the sound velocity of constant-temperature water is taken as the initial value of the iteration. When the difference in sound velocity in each iteration is less than the set threshold, it is considered that the iteration has converged. Finally, different sound velocity values ​​are mapped to different colors and displayed in three-dimensional space.

Images