Calibration method for a transmission-type ultrasonic scanning system

Abstract

The application relates to a correction method suitable for a transmission type ultrasonic scanning system. The method comprises the following steps: acquiring a roller scanning characteristic of a roller pair in the transmission type ultrasonic scanning system, and constructing a correction matrix group based on the physical length of the corresponding surface of the current roller pair, wherein the correction matrix group comprises a transmission correction matrix corresponding to a transmission roller in the current roller pair and a receiving correction matrix corresponding to a receiving roller in the current roller pair; when the transmission signal is corrected, the working rotating state of the current roller pair is determined, the correction gain coefficient corresponding to each receiving array element is acquired based on the working rotating state of the current roller pair, and the acquired correction gain coefficient is used to correct the received transmission signal. The application can effectively correct the transmission signal, can eliminate the signal distortion caused by the unevenness of the roller surface, the unevenness of the material and the angle deviation of the current roller pair during rotation, and can improve the precision and reliability of the transmission type ultrasonic scanning detection.

Description

Technical Field

[0001] This invention relates to a calibration method, and more particularly to a calibration method suitable for transmission ultrasound scanning systems. Background Technology

[0002] Ultrasound is a sound wave with a frequency higher than 20kHz. It has good directionality, strong penetrating power, and is easy to concentrate. It can propagate over long distances in both liquids and solids, making it suitable for non-destructive testing (NDT) to detect bubbles, cracks, and other defects in workpieces. In workpiece inspection based on industrial ultrasonic imaging, solid-coupled inspection schemes can be used, requiring solid materials with good sound conductivity and low attenuation to avoid workpiece contamination. In workpiece inspection based on industrial ultrasonic imaging, a transmission ultrasonic scanning system with a multi-roller coupling medium can be used for ultrasonic inspection of the workpiece.

[0003] Existing technologies disclose transmission ultrasonic scanning systems using multi-roller coupling media, which can perform ultrasonic testing on batteries produced on battery production lines. For example, patent application CN109283259A uses a combination of solid and liquid ultrasonic transmission media, placing the transducer in a liquid-filled roller with the outer wall of the roller in contact with the workpiece to be tested. The rotation of the roller enables relative movement between the transducer and the workpiece, thus achieving ultrasonic mechanical scanning. Furthermore, to improve the signal-to-noise ratio of the ultrasonic scanning signal and enhance the accuracy and reliability of ultrasonic testing, patent application CN119310176A performs efficient reconstruction of the transmission ultrasonic signal obtained through solid-coupled scanning, particularly after signal-to-noise ratio enhancement processing, thereby improving the resolution of the generated ultrasonic image.

[0004] The aforementioned transmission ultrasonic scanning system includes at least a pair of rollers, with a coupling adhesive layer cured on the outer surface of the rollers, which serves as a solid coupling medium between the rollers and the workpiece to be tested.

[0005] Further research revealed that when using the aforementioned transmission ultrasonic scanning system to inspect workpieces, there is a technical bottleneck due to significant inherent system noise interference. This results in low accuracy and automation in current workpiece inspection. Specifically, unevenness inside the roller, uneven thickness / material of the coupling adhesive layer on the roller surface, and angular deviation of the roller during rotation all introduce severe inherent system noise. This inherent noise leads to distortion of the transmission signal, resulting in low quality and confidence of the subsequently generated workpiece image, making it difficult to distinguish whether the signal anomaly originates from defects in the workpiece itself or from system errors.

[0006] Therefore, there is an urgent need for a calibration method suitable for transmission ultrasound scanning systems to solve the problem of inherent noise interference. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a correction method suitable for transmission ultrasonic scanning systems. This method can effectively correct transmission signals and eliminate signal distortion caused by factors such as unevenness inside the roller, uneven thickness / material of the coupling adhesive layer on the roller surface, and angular deviation of the roller during rotation, thereby improving the accuracy and reliability of transmission ultrasonic scanning detection.

[0008] According to the technical solution provided by the present invention, a correction method suitable for a transmission ultrasonic scanning system is characterized in that it is used to correct the transmission signal generated by any roller pair in the transmission ultrasonic scanning system during ultrasonic scanning, the correction method comprising: The scanning features of a pair of rollers within a transmission ultrasonic scanning system are acquired, and a set of correction matrices is constructed based on the physical length of the corresponding surface of the current roller pair. The correction matrix set includes the transmission correction matrix corresponding to the transmitting drum in the current drum pair and the reception correction matrix corresponding to the receiving drum in the current drum pair. The transmission correction matrix includes several physical lengths of the transmission correction surface that characterize the correction rotation angle of the transmission drum, and the transmission calibration signal corresponding to each receiving array element under each physical length of the transmission correction surface. The receiving correction matrix includes several physical lengths of receiving correction surfaces that characterize the rotation angle of the receiving drum correction, and a receiving calibration signal corresponding to each receiving array element under each physical length of the receiving correction surface. During transmission signal correction, the current operating rotation state of the roller pair is determined, and the correction gain coefficient corresponding to each receiving array element is obtained based on the current operating rotation state of the roller pair. The obtained correction gain coefficient is then used to correct the received transmission signal. The correction gain coefficient is generated based at least on the transmit reference signal and the receive calibration signal associated with the current working rotation state of the drum.

[0009] When the current working rotation state of the roller pair is synchronized with the correction rotation state used when constructing the correction matrix group, it includes: The transmission rotation angle of the transmitting drum is obtained, and the physical length of the corresponding transmitting working surface is calculated based on the drum scanning characteristics and the transmission rotation angle, so as to extract the transmission calibration signal that corresponds to the physical length of the transmitting working surface for each receiving array element in the transmission calibration matrix. The receiving rotation angle of the receiving drum is obtained, and the physical length of the corresponding receiving working surface is calculated based on the drum scanning characteristics and the receiving rotation angle, so as to extract the receiving calibration signal that corresponds to the physical length of the receiving working surface for each receiving array element in the receiving calibration matrix. For each receiving array element, the extracted transmit calibration signal is multiplied by the receive calibration signal, and the product is used as the correction gain coefficient corresponding to the current reception.

[0010] When the current working rotation state of the roller pair is not synchronized with the correction rotation state used when constructing the correction matrix group, including: Obtain the transmitting rotation angle of the transmitting drum and the receiving rotation angle of the receiving drum. The physical length of the transmitting working surface is calculated based on the drum scanning characteristics and the transmitting rotation angle. The transmitting gain of each receiving array element is then calculated based on this physical length. When calculating the transmit gain of each receiving element, the reference receiving element associated with the current receiving element and the corresponding transmit calibration signal are determined within the transmit correction matrix based on the physical length of the transmitting working surface. Based on all reference calibration array elements and their corresponding transmit calibration signals, the transmit gain of the current receive array element is calculated and generated. The method for calculating and generating the transmit gain includes at least linear interpolation. The physical length of the receiving working surface is calculated based on the drum scanning characteristics and the receiving rotation angle. The receiving gain of each receiving array element is then calculated based on this physical length. When calculating the receiving gain of each receiving array element, the reference receiving array element associated with the current receiving array element and the receiving calibration signal corresponding to the reference receiving array element are determined within the receiving calibration matrix based on the physical length of the receiving calibration surface. Based on all reference calibration array elements and their corresponding receive calibration signals, the receive gain of the current receive array element is calculated, wherein the method for calculating the receive gain includes at least linear interpolation. For each receiving array element, the transmit gain calculated above is multiplied by the receive gain, and the product is used as the correction gain coefficient for the current receiving array element.

[0011] For each receiving element, if the physical length of the transmitting working surface matches the physical length of a transmitting correction surface within the transmitting correction matrix, then the current receiving element is configured as the reference receiving element, and the corresponding transmitting correction signal is used as the transmitting gain. Otherwise, bilinear interpolation is used to calculate the transmitting gain. When using bilinear interpolation to calculate the transmit gain, the reference receive elements associated with the current receive element are determined to be at least a first reference receive element and a second reference receive element, wherein... If the current receiving array element is the first element, then the first receiving array element is the receiving array element with the current array number, and the second receiving array element is the receiving array element one position below the first element. If the current receiving array element is the last one in the arrangement sequence, then the first receiving array element is used as the reference for the current receiving array element, and the second receiving array element is used as the reference for the receiving array element whose arrangement sequence number is one position above the last one. If the current receiving array element's arrangement number is between the first and last positions, then the first receiving array element is the receiving array element one position below the current receiving array element's arrangement number, and the second receiving array element is the receiving array element one position above the current receiving array element's arrangement number. The physical length of the launch working surface is searched downwards to generate the first reference physical length for launch, wherein the first reference physical length for launch is less than or equal to the physical length of the launch working surface and is the same as the physical length of a launch correction surface within the launch correction matrix; Based on the first reference physical length for launch, a second reference physical length for launch is configured, wherein the second reference physical length for launch is the same as the physical length of a launch correction surface within the launch correction matrix, the second reference physical length for launch is greater than the first reference physical length for launch, and the second reference physical length for launch and the first reference physical length for launch are adjacent to each other within the launch correction matrix; Based on the corresponding transmission calibration signals of the first reference receiving array element at the first and second reference physical lengths of transmission, and the corresponding transmission calibration signals of the second reference receiving array element at the first and second reference physical lengths of transmission, the corresponding transmission gain is generated by bilinear interpolation.

[0012] When bilinear interpolation is used to calculate the transmit gain, then:

[0013] in, For the arrangement sequence number The current receiving array element and the transmission gain calculated based on the corresponding physical length of the transmitting working surface. The physical length of the launching working surface. This is the current array element's sequence number. For the first reference physical length of launch, For the second reference physical length of launch, For reference, the arrangement sequence number of the first receiving array element, For reference, the arrangement sequence number of the second receiving array element, To reference the transmission calibration signal corresponding to the first receiving array element at the first reference physical length, To reference the transmission calibration signal corresponding to the first receiving element at the second reference physical length, To reference the transmission calibration signal corresponding to the second receiving element at the first reference physical length, This is to reference the transmission calibration signal corresponding to the second receiving array element at the second reference physical length.

[0014] The current roller scanning features of the roller pair include at least the physical features of the rollers, wherein the physical features of the rollers include the perimeter and radius of the cross-section corresponding to the transmitting roller, and the perimeter and radius of the cross-section corresponding to the receiving roller; An array of transmitting elements is set inside the transmitting drum. The transmitting elements are arranged along the axial direction of the transmitting drum. The receiving elements are arranged in an array along the axial direction of the receiving drum. The number of transmitting elements is the same as the number of receiving elements, and the corresponding positions of the transmitting elements and receiving elements correspond to each other. The constructed set of correction matrices includes: The number of physical lengths of the transmit correction surface within the transmit correction matrix corresponds to the number of physical lengths of the receive correction surface within the receive correction matrix, and there is a one-to-one correspondence between the physical lengths of the transmit and receive correction surfaces. The rotation angle of the transmitting drum, represented by the physical length of each transmitting correction surface, is consistent with the rotation angle of the receiving drum, represented by the physical length of the corresponding receiving correction surface.

[0015] When constructing the emission correction matrix, the following are included: Both the transmitting drum equipped with the transmitting array and the receiving array are placed in the calibration coupling medium, and the receiving array is directly corresponding to the transmitting array in the transmitting drum. The drive launch drum rotates one revolution in angular steps according to the correction rotation angle. After each revolution, the physical length of the corresponding launch correction surface is calculated based on the radius of the launch drum's cross-section. After each rotation, the transmitting array is configured to transmit ultrasonic signals to the receiving array, and the intensity of the transmitted signal received by each receiving element is calculated. The intensity of the transmitted signal received by each receiving element is used as the transmission calibration reference signal under the current physical length of the transmission calibration surface. After obtaining the physical lengths of all transmit calibration surfaces and the transmit calibration reference signal for each receive element at each physical length of the transmit calibration surface, all transmit calibration reference signals are normalized so that each normalized value is configured as a transmit calibration signal.

[0016] Before the current roller pair is operational or a correction matrix group is constructed within a transmission ultrasonic scanning system, the physical state of the current roller pair is checked, including... When checking the physical status of the current roller pair, the following is included: The current roller pair is configured to be unloaded and the transmitting roller and receiving roller inside the current roller pair are brought into contact. Then, the transmitting roller and receiving roller are driven to rotate one revolution and the transmission signal under the unloaded state is collected. A background ultrasound image is reconstructed based on all transmitted signals; The uniformity characteristics of the background ultrasound images are analyzed, and the physical state of the current roller pair is determined based on the analyzed uniformity characteristics. The physical state of the current roller pair includes normal rollers, abnormal rollers, and degraded state. When the physical state is in a state degradation state, the correction matrix group of the current roller pair is reconstructed.

[0017] After the transmission ultrasonic scanning system performs an ultrasonic scan on the workpiece, it performs ultrasonic imaging based on all corrected transmission signals to generate a scanned target image of the workpiece. When performing ultrasonic imaging based on all corrected transmission signals, a basic image of the workpiece is first generated. Then, the basic image of the workpiece is subjected to at least bilinear interpolation to generate a target image of the workpiece.

[0018] During the ultrasonic scanning of the workpiece using a transmission ultrasonic scanning system, slippage detection is also performed on the contact between the workpiece and the roller pair. When slippage occurs, image frames acquired during the slippage period are identified and discarded, and ultrasonic imaging is performed based on the discarded image frames to generate a target image of the workpiece.

[0019] The advantages of this invention are as follows: For roller pairs within a transmission ultrasonic system, a correction matrix group decoupled from the roller pair's speed is constructed. When correcting the transmission signal, the current operating rotation state of the roller pair is determined, and the correction gain coefficient corresponding to each receiving array element is obtained based on the current operating rotation state of the roller pair. The obtained correction gain coefficient is then used to correct the received transmission signal. This eliminates signal distortion caused by factors such as unevenness inside the rollers, uneven thickness / material of the coupling adhesive layer on the roller surface, and angular deviation during the rotation of the current roller pair, thereby improving the accuracy and reliability of transmission ultrasonic scanning detection. Attached Figure Description

[0020] Figure 1 This is a schematic flowchart of one embodiment of the correction method of the present invention.

[0021] Figure 2 This is a schematic diagram of one embodiment of the present invention, which uses a roller to perform ultrasonic scanning on a workpiece.

[0022] Figure 3 This is a schematic diagram of one embodiment of constructing the emission correction matrix according to the present invention.

[0023] Figure 4 This is a schematic diagram of one embodiment of the present invention when constructing the receiving correction matrix.

[0024] Explanation of reference numerals in the attached drawings: 1-Transmitting roller, 2-Receiving roller, 3-Transmitting array, 4-Receiving array, 5-Workpiece, 6-Correction coupling medium. Detailed Implementation

[0025] The present invention will be further described below with reference to specific accompanying drawings and embodiments.

[0026] To effectively correct the transmission signal and eliminate signal distortion caused by factors such as unevenness inside the roller, uneven thickness / material of the coupling adhesive layer on the roller surface, and angular deviation during the rotation of the roller pair, thereby improving the accuracy and reliability of transmission ultrasonic scanning detection, this invention provides a correction method suitable for transmission ultrasonic scanning systems. Specifically, it is used to correct the transmission signal generated by any roller pair in an ultrasonic scanning system during ultrasonic scanning. The correction method includes: The scanning features of a pair of rollers within a transmission ultrasonic scanning system are acquired, and a set of correction matrices is constructed based on the physical length of the corresponding surface of the current roller pair. The correction matrix set includes the transmission correction matrix corresponding to the transmitting drum in the current drum pair and the reception correction matrix corresponding to the receiving drum in the current drum pair. The transmission correction matrix includes several physical lengths of the transmission correction surface that characterize the correction rotation angle of the transmission drum, and the transmission calibration signal corresponding to each receiving array element under each physical length of the transmission correction surface. The receiving correction matrix includes several physical lengths of receiving correction surfaces that characterize the rotation angle of the receiving drum correction, and a receiving calibration signal corresponding to each receiving array element under each physical length of the receiving correction surface. During transmission signal correction, the current operating rotation state of the roller pair is determined, and the correction gain coefficient corresponding to each receiving array element is obtained based on the current operating rotation state of the roller pair. The obtained correction gain coefficient is then used to correct the received transmission signal. The correction gain coefficient is generated based at least on the transmit reference signal and the receive calibration signal associated with the current working rotation state of the drum.

[0027] It should be noted that the transmission ultrasound scanning system is a commonly used existing ultrasound scanning system, such as the ultrasound scanning form disclosed in publication number CN119310176A. Specifically, the transmission ultrasound scanning system may include one or more roller pairs. When there are multiple roller pairs, the multiple roller pairs are arranged sequentially. Each roller pair includes a transmitting roller 1 and a receiving roller 2, with the transmitting roller 1 and receiving roller 2 placed correspondingly. A transmitting element array 3 is arranged in the transmitting roller 1, and a receiving element array 4 is arranged in the receiving roller 2. The transmitting element array 3 includes several... The transmitting array elements are arranged in sequence. The receiving array element array 4 includes a number of receiving array elements arranged in sequence. The transmitting array elements in the transmitting array element array 3 are linearly distributed along the axial direction of the transmitting roller 1. The receiving array elements in the receiving array element array 4 are linearly distributed along the axial direction of the receiving roller 2. The number of transmitting array elements is consistent with the number of receiving array elements, and the corresponding positions of the transmitting array elements and the receiving array elements correspond to each other. The corresponding situation of the transmitting roller 1, receiving roller 2, transmitting array element array 3 and receiving array element array 4 can be referred to the technical content disclosed in CN119310176A.

[0028] Figure 2 The illustration shows an embodiment of ultrasonic scanning detection of workpiece 5 using a pair of rollers. As shown in the figure, workpiece 5 rolls in contact with the transmitting roller 1 and the receiving roller 2. During the movement of workpiece 5, the transmitting array element in the transmitting roller 1 emits ultrasonic waves towards workpiece 5, and the receiving array element in the receiving roller 2 receives the transmitted signal. The form of ultrasonic scanning detection between workpiece 5 and the roller pair is consistent with the prior art and will not be described in detail here. It should be understood that the correction of the present invention specifically refers to the correction of the transmitted signal received by the receiving array element. After correction, signal distortion caused by factors such as unevenness inside the rollers (transmitting roller 1 / receiving roller 2), uneven thickness / material of the coupling adhesive layer on the roller surface, and angular deviation during the rotation of the current roller pair can be eliminated, thereby improving the accuracy and reliability of transmission ultrasonic scanning detection.

[0029] Understandably, when multiple roller pairs exist, at least one roller pair's transmission signal should be corrected, and preferably all roller pairs' transmission signals should be corrected. Figure 1 It can be seen that when it is necessary to correct the transmission signal of a roller pair, a correction matrix group for the current roller pair should be constructed. When constructing the correction matrix group, the roller scanning characteristics of the current roller pair should be obtained, and the constructed correction matrix group should be based on the physical length of the corresponding surface of the current roller pair. Specifically, The current roller scanning features of the roller pair include at least the physical features of the rollers, wherein the physical features of the rollers include the perimeter and radius of the cross-section corresponding to the transmitting roller 1, and the perimeter and radius of the cross-section corresponding to the receiving roller 2. Figure 2 and Figure 3The figure shows an embodiment of the transmitting roller 1. As can be seen from the figure, the transmitting roller 1 is generally cylindrical and the cross-section of the transmitting roller 1 is circular or annular. Similarly, the corresponding cross-section of the receiving roller 2 is also circular or annular. In specific implementation, the circumference and radius of the corresponding cross-section of the transmitting roller 1 should be equal to the circumference and radius of the corresponding cross-section of the receiving roller 2. That is, the transmitting roller 1 and the receiving roller 2 should preferably be the same roller.

[0030] When the transmitting roller 1 and the receiving roller 2 adopt the above-mentioned corresponding forms, the physical length of the corresponding surface of the roller pair specifically refers to the length corresponding to the product of the rotation angle of the transmitting roller 1 and the cross-sectional radius of the transmitting roller 1, and of course, it also includes the length corresponding to the product of the rotation angle of the receiving roller 2 and the cross-sectional radius of the receiving roller 2.

[0031] To effectively correct the transmitted signal, the correction matrix set of this invention should include a transmission correction matrix and a reception correction matrix, both in two-dimensional matrix form. The data formats within the transmission and reception correction matrices are similar, but the transmission correction matrix is ​​generated for the transmission roller 1, while the reception correction matrix is ​​mainly generated for the reception roller 2. Specifically, the transmission correction matrix includes several physical lengths of the transmission correction surface, and the reception correction matrix includes several physical lengths of the reception correction surface. Therefore, using the corresponding physical lengths of the current roller pair as a reference—specifically, using the physical lengths of the transmission and reception correction surfaces as a reference—it can be seen from the above description that the correction matrix set is decoupled from the corresponding rotational speeds of the transmission roller 1 and the reception roller 2. Thus, when correcting the transmitted signal based on the correction matrix set, signal distortion caused by factors such as unevenness inside the rollers, uneven thickness / material of the coupling adhesive layer on the roller surface, and angular deviations during the rotation of the current roller pair can be eliminated.

[0032] In practical implementation, the transmit calibration matrix should include several physical lengths of transmit calibration surfaces. The number of physical lengths of transmit calibration surfaces should be related to the calibration rotation angle used when constructing the calibration matrix group. For example, when the calibration rotation angle of transmit roller 1 is 1°, the physical lengths of transmit calibration surfaces in the transmit calibration matrix should be 360. When the calibration rotation angle is 10°, the physical lengths of transmit calibration surfaces in the transmit calibration matrix should be 36. After selecting the calibration rotation angle, the number of physical lengths of transmit calibration surfaces can be determined. Within the transmit calibration matrix, under each physical length of transmit calibration surface, there exists a transmit calibration signal corresponding to each receiver array element. The details of the transmit calibration signals can be found in the following description.

[0033] When performing ultrasonic scanning on workpiece 5, the rotational speed of the transmitting roller 1 should generally be the same as that of the receiving roller 2. Therefore, when constructing the receiving correction matrix, the correction rotation angle of the receiving roller 2 should be equal to the correction rotation angle of the transmitting roller 1. Thus, for the constructed correction matrix set, we have: For example, when the correction rotation angle is 1°, a mark is set as the zero point on the transmitting roller 1. When the mark rotates 1° from the zero point, the physical length of the first transmitting correction surface can be calculated in the above manner. When the mark rotates from the 1° position to the 2° position, the rotation angle used to calculate the physical length of the second transmitting correction surface is 2°. Other cases are deduced in the same way. It can be seen that the number of physical lengths of the transmitting correction surface in the transmitting correction matrix is ​​consistent with the corresponding number of physical lengths of the receiving correction surface in the receiving correction matrix, and the physical lengths of the transmitting correction surface and the physical lengths of the receiving correction surface are in one-to-one correspondence. In addition, the rotation angle of the transmitting roller 1 represented by each physical length of the transmitting correction surface is consistent with the rotation angle of the receiving roller 2 represented by the corresponding physical length of the receiving correction surface.

[0034] It should be understood that once a set of correction matrices for a pair of rollers is constructed, the transmission signal generated by the current roller in the ultrasonic scan can be corrected. Therefore, for each pair of rollers, there is a one-to-one correspondence between the roller pair and the constructed set of correction matrices. That is, only after the corresponding set of correction matrices is constructed can the transmission signal generated by the roller pair in the ultrasonic scan be corrected.

[0035] Depend on Figure 1 It is understood that when performing transmission signal correction, the current operating rotation state of the roller pair should be determined. The main purpose of determining the current operating rotation state of the roller pair is to identify the transmit calibration signal and receive calibration signal used to calculate the correction gain coefficient for each receiving element. Once the transmit calibration signal and receive calibration signal used to calculate the correction gain coefficient for each receiving element are obtained, the correction gain coefficient corresponding to each receiving element can be determined. After obtaining the correction gain coefficient, the transmission signal received by the current receiving element can be corrected. It is understood that in subsequent detection stages, the corrected transmission signal should be used, such as for ultrasonic imaging. The method of using the corrected transmission signal in the detection stage is consistent with existing technology and will not be elaborated here.

[0036] In specific implementation, the method for obtaining the correction gain coefficient can be selected as needed. For example, based on the current working rotation state of the roller pair, the transmit correction signal and receive calibration signal associated with each receiving array element can be obtained. Then, the corresponding correction gain coefficient can be calculated and generated based on the obtained transmit calibration signal and receive calibration signal. Alternatively, based on the transmit calibration signal and receive calibration signal of each receiving array element, a corresponding gain coefficient matrix can be constructed first. Each gain coefficient matrix includes the correction gain coefficient corresponding to each array element and the correction gain coefficient corresponding to each working rotation state. Of course, the calculation method of the correction gain coefficient in the gain coefficient matrix is ​​consistent with the above, that is, it is calculated and generated based on the corresponding transmit calibration signal and receive calibration signal. For specific calculation methods, please refer to the corresponding explanation below.

[0037] In one embodiment of the present invention, constructing the emission correction matrix includes: The transmitting drum 1, which is equipped with the transmitting array 3, and the receiving array 4 are both placed in the calibration coupling medium 6, and the receiving array 4 is directly aligned with the transmitting array 3 in the transmitting drum 1. Drive the launch roller 1 to rotate one revolution according to the angular step of the correction rotation angle. After each rotation, calculate the physical length of the corresponding launch correction surface based on the radius of the cross-section of the launch roller 1. After each rotation, the transmitting array 3 is configured to transmit ultrasonic signals to the receiving array 4, and the intensity of the transmitted signal received by each receiving array element is calculated. The intensity of the transmitted signal received by each receiving array element is used as the transmission calibration reference signal under the current physical length of the transmission calibration surface. After obtaining the physical lengths of all transmit calibration surfaces and the transmit calibration reference signal for each receive element at each physical length of the transmit calibration surface, all transmit calibration reference signals are normalized so that each normalized value is configured as a transmit calibration signal.

[0038] Figure 3The diagram illustrates one embodiment of constructing a transmission correction matrix. As shown, when constructing the transmission correction matrix, a transmitting roller 1 and a transmitting element array 3 assembled within the transmitting roller 1 should be provided, but a receiving roller 2 is not provided. Only a receiving element array 4 assembled within the receiving roller 2 is used. Specifically, the transmitting roller 1 and the receiving element array 4 should be placed within a correction coupling medium 6, which can be water or other media suitable for ultrasonic scanning detection. It should be understood that the transmitting roller 1 used here should correspond to the transmitting roller 1 in the roller pair within a transmission ultrasonic scanning system. Similarly, the transmitting element array 3 and the receiving element array 4 should also be consistent. Within the correction coupling medium 6, the receiving element array 4 corresponds directly to the transmitting element array 3. This direct correspondence should correspond to the corresponding states of the transmitting element array 3 and receiving element array 4 within the transmission ultrasonic scanning system, specifically simulating the ultrasonic transmission and reception of the roller pair within the transmission ultrasonic scanning system. Further details are omitted here.

[0039] It should be noted that both the transmitting roller 1 and the receiving roller 2 within the transmission ultrasound scanning system should be able to rotate. To facilitate the construction of the transmission correction matrix, the transmitting roller 1, placed within the correction coupling medium 6, should also be able to rotate. The method of driving the transmitting roller 1 to rotate can be consistent with the rotation drive of the corresponding transmitting roller 1 in the transmission ultrasound scanning system. Specifically, the transmitting roller 1 should be driven to rotate one revolution according to the angular step of the correction rotation angle. The correction rotation angle and angular step can be found in the corresponding description above. After each rotation, the physical length of the corresponding transmission correction surface is calculated based on the radius of the cross-section of the transmitting roller 1. Each rotation specifically refers to rotating the transmitting roller 1 to the corresponding position according to the correction rotation angle and angular step. At this point, the physical length of the corresponding transmission correction surface can be calculated.

[0040] It should be understood that after each rotation, in addition to calculating the physical length of the transmitting calibration surface, the transmitting element array 3 should be configured to transmit ultrasonic signals to the receiving element array 4. This could involve configuring each transmitting element in the transmitting element array 3 to transmit ultrasonic signals towards the receiving element array 4, and configuring each receiving element to simultaneously receive the ultrasonic signals. The received ultrasonic signals are then transmitted through the transmitting roller 1. After each receiving element 4 receives the transmitted signal, it should calculate the intensity of the received transmitted signal and use this intensity as the transmission calibration reference signal under the current physical length of the transmitting calibration surface. It should be noted that the intensity of the transmitted signal can be a characteristic that characterizes the signal intensity, such as PPV (Peak-to-Peak Value), amplitude, or envelope. The method for determining the intensity of the transmitted signal can be selected as needed, but the characteristic type selected for all transmitted signal intensities should be the same, such as PPV, amplitude, or envelope. Examples will not be provided here.

[0041] To facilitate the subsequent calculation of the correction gain coefficient, after obtaining all the transmit calibration reference signals, a normalization calculation should be performed. During the normalization calculation, the mean of all transmit calibration reference signals is first calculated. Then, each transmit calibration reference signal is divided by the calculated mean, and the quotient of the transmit calibration reference signal and the mean is used as the corresponding transmit calibration signal. Of course, other normalization calculation methods can also be used, which will not be illustrated here.

[0042] The above describes one method for constructing the transmit correction matrix; the receive correction matrix can be constructed using the same method. Unlike constructing the transmit correction matrix, when constructing the receive correction matrix, only the transmit element array 3 should be used, but the receive drum 2 and the receive element array 4 assembled within the receive drum 2 should also be used. The transmit element array 3, receive drum 2, and receive element array 4 participating in the receive correction matrix should all be placed within the correction coupling medium 6, such as... Figure 4 As shown, the distribution positions of the transmitting array 3, the receiving drum 2, and the receiving array 4 within the correction coupling medium 6 can all be referenced in constructing the transmitting correction matrix and for... Figure 3 The corresponding explanation.

[0043] Furthermore, unlike the construction of the transmit correction matrix, the receive roller 2 should be configured to rotate. During the rotation of the receive roller 2, it still rotates one full revolution according to the correction rotation angle used to construct the transmit correction matrix. The situation of the receive roller 2 rotating according to the correction rotation angle can be referred to the description of the corresponding rotation of the transmit roller 1 above, and will not be repeated here. After the receive roller 2 rotates, the receive calibration reference signal at each physical length of the receive calibration surface can be obtained in the above manner. After normalizing all the receive calibration reference signals, the corresponding receive calibration signal can be obtained.

[0044] As can be seen from the above description of the process of constructing the transmit correction matrix and the receive correction matrix, the transmit correction matrix only characterizes the non-uniformity of the transmit roller 1, and the receive correction matrix only characterizes the non-uniformity of the receive roller 2, thus eliminating the interference of the other roller on the single-sided correction.

[0045] In one embodiment of the present invention, when the current working rotation state of the roller pair is synchronized with the correction rotation state used when constructing the correction matrix group, the following is included: The transmission rotation angle of the transmitting drum 1 is obtained, and the physical length of the corresponding transmitting working surface is calculated based on the drum scanning features and the transmission rotation angle, so as to extract the transmission calibration signal that corresponds to the physical length of the transmitting working surface for each receiving array element in the transmission calibration matrix. The receiving rotation angle of the receiving roller 2 is obtained, and the physical length of the corresponding receiving working surface is calculated based on the roller scanning characteristics and the receiving rotation angle, so as to extract the receiving calibration signal that corresponds to the physical length of the receiving working surface for each receiving array element in the receiving calibration matrix. For each receiving array element, the extracted transmit calibration signal is multiplied by the receive calibration signal, and the product is used as the correction gain coefficient corresponding to the current receiving array element.

[0046] As explained above, when constructing the calibration matrix group, both the transmitting roller 1 and the receiving roller 2 should rotate one full revolution according to the calibration rotation angle. Therefore, the calibration rotation state used in constructing the calibration matrix group specifically refers to the state in which the transmitting roller 1 and the receiving roller 2 rotate according to the calibration rotation angle. During operation, the transmitting roller 1 and the receiving roller 2 are generally driven by servo motors, and the corresponding rotation angles of the transmitting roller 1 and the receiving roller 2 are collected through methods such as rotary encoders. When the collected rotation angles of the transmitting roller 1 and the receiving roller 2 are consistent with the calibration rotation angle used in constructing the calibration matrix group, it is considered that the current working rotation state of the roller pair is synchronized with the calibration rotation state used when constructing the calibration matrix group; otherwise, it can be considered asynchronous.

[0047] When the above synchronization is satisfied, the launch rotation angle of the launch roller 1 is obtained. The launch rotation angle is the rotation angle of the launch roller 1 during operation. After obtaining the launch rotation angle, multiplying the launch rotation angle by the radius of the cross-section of the launch roller 1 yields the physical length of the launch working surface. The physical length of the launch working surface can be found in the description of the physical length of the launch correction surface described above. It should be understood that, due to the synchronization state, the physical length of the launch working surface should correspond to the physical length of one launch correction surface within the launch correction matrix. Therefore, based on the physical length of the launch working surface, the launch correction signal corresponding to each array element can be extracted within the launch correction matrix.

[0048] Similarly, for the receiving roller 2, the receiving rotation angle is obtained. As can be seen from the above description, under the same working time, the receiving rotation angle should be the same as the transmitting rotation angle. The physical length of the receiving working surface can be calculated from the receiving rotation angle. After that, the receiving calibration signal corresponding to each receiving array element can be extracted from the receiving calibration matrix.

[0049] After determining the transmit calibration signal and receive calibration signal for each receiving array element using the above method, the transmit calibration signal and receive calibration signal are multiplied together, and the product is used as the correction gain coefficient corresponding to the current receiving array element.

[0050] Furthermore, if the current rotational state of the roller pair is synchronized with the rotational state used when constructing the correction matrix group, the gain coefficient matrix can be constructed first, as mentioned above. For the correction gain coefficients within the gain coefficient matrix, the transmit calibration signal corresponding to each receiving element can be multiplied by the receive calibration signal, and the product can be used as a correction gain coefficient within the gain coefficient matrix. In this case, each correction gain coefficient corresponds directly to a physical length of the receiving working surface and a physical length of the transmitting working surface, that is, it corresponds directly to a physical length of the transmitting correction surface and a physical length of the receiving correction surface. After the gain coefficient matrix is ​​constructed, the correction gain coefficient of each receiving element can be directly extracted from the gain coefficient matrix according to the current rotational state of the roller pair.

[0051] As can be seen from the above description, if the current working rotation state of the roller pair is synchronized with the correction rotation state used when constructing the correction matrix group, the correction gain coefficient of each receiving array element can be obtained by one of the above methods, and the specific method can be selected as needed.

[0052] For a transmission ultrasonic scanning system, the system's amplitude response in the frequency domain is multiplicative. The emitted ultrasonic wave passes through the transmitting element—transmitting roller 1—workpiece 5—receiving roller 2—receiving element. When the product of the transmitted calibration signal and the received calibration signal is used as the correction gain coefficient, the inherent errors of the system (errors of transmitting roller 1 and receiving roller 2) can be eliminated. It should be noted that during operation, the inner cylinders of both transmitting roller 1 and receiving roller 2 are filled with coupling fluid. Therefore, the ultrasonic path should also include the coupling fluid in the corresponding inner cylinders of transmitting roller 1 and receiving roller 2.

[0053] When correcting the transmitted signal, the transmitted signal should be divided by the correction gain coefficient, and the corresponding quotient is the corrected transmitted signal. It should be noted that the transmitted ultrasonic scanning system of this invention can be approximated as a linear time-invariant system, i.e., low sound pressure neglects the generation of nonlinear harmonics, and the workpiece 5 under test remains stationary during ultrasonic data acquisition. The responses of each component are generally combined in a product form; therefore, dividing the transmitted signal by the correction gain coefficient can effectively correct the transmitted signal, thereby extracting the signal that is only related to the physical characteristics of the workpiece 5 under test.

[0054] In one embodiment of the present invention, when the current working rotation state of the roller pair is not synchronized with the correction rotation state used when constructing the correction matrix group, the following applies: Obtain the transmitting rotation angle of transmitting roller 1 and the receiving rotation angle of receiving roller 2. The physical length of the transmitting working surface is calculated based on the drum scanning characteristics and the transmitting rotation angle. The transmitting gain of each receiving array element is then calculated based on this physical length. When calculating the transmit gain of each receiving array element, the reference receiving array element associated with the current receiving array element and the transmit calibration signal corresponding to the reference receiving array element are determined within the transmit calibration matrix based on the physical length of the transmit calibration surface. Based on all reference calibration array elements and their corresponding transmit calibration signals, the transmit gain of the current receive array element is calculated and generated. The method for calculating and generating the transmit gain includes at least linear interpolation. The physical length of the receiving working surface is calculated based on the drum scanning characteristics and the receiving rotation angle. The receiving gain of each receiving array element is then calculated based on this physical length. When calculating the receiving gain of each receiving array element, the reference receiving array element associated with the current receiving array element and the receiving calibration signal corresponding to the reference receiving array element are determined within the receiving calibration matrix based on the physical length of the receiving calibration surface. Based on all reference calibration array elements and their corresponding receive calibration signals, the receive gain of the current receive array element is calculated, wherein the method for calculating the receive gain includes at least linear interpolation. For each receiving array element, the transmit gain calculated above is multiplied by the receive gain, and the product is used as the correction gain coefficient for the current receiving array element.

[0055] It should be noted that when the current working rotation state of the roller pair is not synchronized with the correction rotation state used when constructing the correction matrix group, it may specifically include: the transmitting roller 1 and the receiving roller 2 rotate synchronously, and the transmitting rotation angle is consistent with the receiving rotation angle, but the transmitting rotation angle, the receiving rotation angle and the correction rotation angle are not equal; or there is a loss of synchronization between the transmitting roller 1 and the receiving roller 2, in which case the transmitting rotation angle and the receiving rotation angle are not equal. In this case, one of the transmitting rotation angle and the receiving rotation angle can be the same as the correction rotation angle, or it can be different.

[0056] As can be seen from the above description, the corresponding transmission rotation angle and reception rotation angle of the transmitting roller 1 and the receiving roller 2 can be obtained by the rotary encoder. Based on the transmission rotation angle and reception rotation angle, it can be determined whether the transmitting roller 1 and the receiving roller 2 are rotating synchronously and whether they are the same as the correction rotation angle. In other words, it can generally be judged directly by comparing the angles.

[0057] When determining whether the transmitting roller 1 and the receiving roller 2 are out of sync, a feasible method is as follows: if the angle difference between the transmitting rotation angle and the receiving rotation angle continuously exceeds a preset threshold (e.g., ±1°) and the duration reaches a certain time (e.g., three consecutive sampling periods), then it is determined that the two transmitting rollers 1 and the receiving roller 2 have lost angular synchronization. Therefore, it can be seen that in the above-mentioned asynchronous state, a loss of synchronization may occur.

[0058] If the current operating rotation state of the roller pair is not synchronized with the correction rotation state used when constructing the correction matrix group, the correction gain coefficient cannot be calculated in the above manner. Simultaneously, it is also impossible to construct the gain coefficient matrix as described above and directly obtain the correction gain coefficient from within the gain coefficient matrix. For specific methods of obtaining / determining the correction gain coefficient, please refer to the corresponding explanations below.

[0059] To improve the accuracy and reliability of the correction gain coefficient calculation, this invention should use a linear interpolation method to calculate the transmit gain and receive gain. Then, the transmit gain and receive gain are multiplied, and the product is used as the correction gain coefficient of the current receiving array element. After obtaining the correction gain coefficient, the transmitted signal can be corrected in the same way as described above. For example, for each receiving array element, the received transmitted signal is divided by the corresponding correction gain coefficient.

[0060] In practice, the methods for calculating transmit gain and receive gain can be the same. For example, for any receiving array element, when calculating the transmit gain, a reference receiving array element associated with the receiving array element and the corresponding transmit correction signal can be used. Then, linear interpolation is performed, and the result of the linear interpolation is the transmit gain. The following example illustrates the linear interpolation method using the calculation of transmit gain as an example: In one embodiment of the present invention, for each receiving array element, when the physical length of the transmitting working surface matches the physical length of a transmitting correction surface within the transmitting correction matrix, the current receiving array element is configured as a reference receiving array element, and the corresponding transmitting correction signal is used as the transmitting gain; otherwise, bilinear interpolation is used to calculate the transmitting gain. When using bilinear interpolation to calculate the transmit gain, the reference receive elements associated with the current receive element are determined to be at least a first reference receive element and a second reference receive element, wherein... If the current receiving array element is the first element, then the first receiving array element is the receiving array element with the current array number, and the second receiving array element is the receiving array element one position below the first element. If the current receiving array element is the last one in the arrangement sequence, then the first receiving array element is used as the reference for the current receiving array element, and the second receiving array element is used as the reference for the receiving array element whose arrangement sequence number is one position above the last one. If the current receiving array element's arrangement number is between the first and last positions, then the first receiving array element is the receiving array element one position below the current receiving array element's arrangement number, and the second receiving array element is the receiving array element one position above the current receiving array element's arrangement number. The physical length of the launch working surface is searched downwards to generate the first reference physical length for launch, wherein the first reference physical length for launch is less than or equal to the physical length of the launch working surface and is the same as the physical length of a launch correction surface within the launch correction matrix; Based on the first reference physical length for launch, a second reference physical length for launch is configured, wherein the second reference physical length for launch is the same as the physical length of a launch correction surface within the launch correction matrix, the second reference physical length for launch is greater than the first reference physical length for launch, and the second reference physical length for launch and the first reference physical length for launch are adjacent to each other within the launch correction matrix; Based on the corresponding transmission calibration signals of the first reference receiving array element at the first and second reference physical lengths of transmission, and the corresponding transmission calibration signals of the second reference receiving array element at the first and second reference physical lengths of transmission, the corresponding transmission gain is generated by bilinear interpolation.

[0061] As can be seen from the above description, both the receiving array elements and the transmitting array elements are distributed in an array. For example, if the transmitting array element array 3 includes 128 transmitting array elements, then the 128 transmitting array elements are arranged sequentially, and the 128 transmitting array elements are arranged linearly along the axis of the transmitting roller 1. At this time, the transmitting array elements can be sorted from top to bottom or from bottom to top. The corresponding arrangement sequence number of the receiving array elements in the receiving array element array 4 should be consistent with the arrangement sequence number of the transmitting array elements. This will not be elaborated further here.

[0062] As can be seen from the calculation method of the physical length of the launch working surface, the physical length of the launch working surface is generally a non-integer. When performing a downward search on the physical length of the launch working surface, specifically, the value of the physical length of the launch working surface is reduced, and it is moved closer to a physical length of the launch correction surface within the launch correction matrix. Moving closer to the physical length of the launch correction surface specifically means minimizing the difference between the physical length of the launch working surface and all other physical lengths of the launch correction surface. It can be understood that the value of the first launch reference physical length should be less than the value of the physical length of the launch working surface. Once the first launch reference physical length is determined, the corresponding second launch reference physical length can be determined. The value of the second launch reference physical length is greater than the value of the first launch reference physical length. The difference between the second launch reference physical length and the first launch reference physical length is the product of a correction rotation angle and the cross-sectional radius of launch drum 1.

[0063] In one embodiment of the present invention, when bilinear interpolation is used to calculate the transmit gain, the following applies:

[0064] in, For the arrangement sequence number The current receiving array element and the transmission gain calculated based on the corresponding physical length of the transmitting working surface. The physical length of the launching working surface. This is the current array element's sequence number. For the first reference physical length of launch, For the second reference physical length of launch, For reference, the arrangement sequence number of the first receiving array element, For reference, the arrangement sequence number of the second receiving array element, To reference the transmission calibration signal corresponding to the first receiving array element at the first reference physical length, To reference the transmission calibration signal corresponding to the first receiving element at the second reference physical length, To reference the transmission calibration signal corresponding to the second receiving element at the first reference physical length, This is to reference the transmission calibration signal corresponding to the second receiving array element at the second reference physical length.

[0065] The above illustrates one embodiment of calculating transmit gain using bilinear interpolation. It is understood that the calculated transmit gain is more accurate and smoother, thereby improving the accuracy of the calculated correction gain coefficient and thus the accuracy of the transmitted signal correction. When using the bilinear interpolation method to calculate transmit gain, the method described above can be referenced; the specific calculation method and process will not be elaborated here.

[0066] In one embodiment of the present invention, before the current roller pair operates or a correction matrix group is constructed in the transmission ultrasonic scanning system, the physical state of the current roller pair is checked, wherein, When checking the physical status of the current roller pair, the following is included: The current roller pair is configured to be unloaded, and the transmitting roller 1 and receiving roller 2 inside the current roller pair are in contact. The transmitting roller 1 and receiving roller 2 are driven to rotate one revolution, and the transmission signal under the unloaded state is collected. A background ultrasound image is reconstructed based on all transmitted signals; The uniformity characteristics of the background ultrasound images are analyzed, and the physical state of the current roller pair is determined based on the analyzed uniformity characteristics. The physical state of the current roller pair includes normal rollers, abnormal rollers, and degraded state. When the physical state is in a state degradation state, the correction matrix group of the current roller pair is reconstructed.

[0067] It is understandable that the roller pairs in a transmission ultrasonic scanning system may be damaged due to wear. Therefore, in order to improve the reliability of ultrasonic scanning, the physical condition of each roller pair should be checked before ultrasonic scanning. Of course, the physical condition of the roller pairs should also be checked before constructing the correction matrix group to avoid constructing the correction matrix group if the roller pairs are already damaged.

[0068] The "unloaded" state specifically refers to a situation where there is no workpiece 5 to be inspected between the rollers, but the transmitting roller 1 and the receiving roller 2 need to be in contact, i.e., the transmitting roller 1 and the receiving roller 2 are configured to roll in contact. In actual operation, the transmitting roller 1 and the receiving roller 2 should be in contact with the workpiece 5. At this time, the corresponding external coupling adhesive layers of the transmitting roller 1 and the receiving roller 2 will be deformed by compression, thus adhering to the surface of the workpiece 5, and expelling the air from the contact surface between the roller and the workpiece 5.

[0069] After the transmitting roller 1 and the receiving roller 2 come into contact, the transmitting roller 1 and the receiving roller 2 are driven to rotate one revolution according to the working rotation angle as described above. During the rotation, the transmitting element transmits ultrasonic signals to the receiving element, and the receiving element receives the corresponding transmission signals. Subsequently, imaging is performed based on all the transmission signals to reconstruct the background ultrasonic image. The method of reconstructing the background ultrasonic image can be consistent with the existing technology, and will not be described in detail here.

[0070] The uniformity characteristics of the background ultrasound image are determined and analyzed. These uniformity characteristics can be either histogram statistical characteristics or variance statistical characteristics. Therefore, after determining the uniformity characteristics, commonly used techniques in this technical field can be employed to obtain the uniformity characteristics of the background ultrasound image. It should be noted that the physical state of the current roller pair is determined based on the uniformity characteristics. The physical state includes normal roller operation, abnormal roller operation, and state degradation. The following example illustrates how the physical state is determined based on the uniformity characteristics: If the overall grayscale distribution of the background ultrasound image is uniform (the histogram statistical characteristics show a single-peaked concentrated distribution), and the global variance of the image is lower than the preset global threshold Threshold1, it indicates that the current roller pair is in good condition and can be directly used for the detection task. That is, the physical state of the current roller pair is that the rollers are normal.

[0071] If the background ultrasound image shows areas of sudden grayscale changes (such as abrupt dark spots or dark lines), and the local variance of this area exceeds the preset local threshold Threshold2 in the histogram statistics, then the current roller pair may have serious physical damage (such as damage or cracking of the surface coupling adhesive layer). In this case, an alarm log should be generated to prompt the operator to check and replace the rollers, and the physical condition of the current roller pair will be marked as roller abnormality.

[0072] If the background ultrasound image shows an overall gradual change in grayscale and a decrease in uniformity (in terms of histogram statistical characteristics, the distribution range becomes wider and the variance becomes larger), but the global variance is between the preset global threshold Threshold1 and the preset local threshold Threshold2, it indicates that the roller pair may be aging or the material properties have changed, that is, the current physical state is degraded. In this case, the above method should be used to reconstruct the correction matrix group to compensate for the slow drift of performance.

[0073] It should be noted that the values ​​of the preset global threshold Threshold1 and the preset local threshold Threshold2 can be selected as needed. For example, a normal roller pair can be used to analyze the uniformity characteristics of the corresponding background ultrasound image to set the preset global threshold Threshold1 and the preset local threshold Threshold2. The specific values ​​should be based on the ability to meet the above analysis and judgment. Examples will not be given here.

[0074] In one embodiment of the present invention, after the transmission ultrasonic scanning system performs ultrasonic scanning on the workpiece, it performs ultrasonic imaging based on all corrected transmission signals to generate a scanned target image of the workpiece, wherein... When performing ultrasonic imaging based on all corrected transmission signals, a basic image of the workpiece is first generated. Then, the basic image of the workpiece is subjected to at least bilinear interpolation to generate a target image of the workpiece.

[0075] Understandably, in the existing technology, the state of workpiece 5 is mostly analyzed by ultrasonic imaging. Workpiece 5 can be an object suitable for ultrasonic imaging, such as a battery. Therefore, after correcting the transmission signal, ultrasonic imaging can be performed on all corrected transmission signals. The ultrasonic imaging method can be consistent with the existing technology, and a workpiece scanning image can be generated after imaging.

[0076] In one embodiment of the present invention, during the generation of the workpiece scan image, a bilinear interpolation algorithm is used to refine the pixels within the basic scan image of the workpiece to improve image resolution and quality. The method of performing bilinear interpolation on the basic scan image of the workpiece can be consistent with existing technologies, such as performing bilinear interpolation on the corrected transmission signals obtained after two adjacent rotation positions. The specific bilinear interpolation method and process will not be described in detail here.

[0077] Since workpiece 5 is in contact with transmitting roller 1 and receiving roller 2 (the specific contact details can be found in the above description), slippage may occur during the ultrasonic scanning of workpiece 5. Slippage refers to a situation where transmitting roller 1 and receiving roller 2 are rotating while workpiece 5 remains stationary. To avoid slippage affecting the accuracy and reliability of workpiece 5 inspection, any slippage that occurs should be detected.

[0078] During slippage detection, the output torque of the servo motor should be monitored and recorded in real time. During normal operation, the torque remains relatively stable. When workpiece 5 begins to enter the detection area, the torque will increase significantly, and when workpiece 5 leaves the detection area, the torque will decrease significantly. If slippage occurs in the middle, an abnormal torque range (usually larger than the stable value) will appear. The servo motors involved include the servo motor driving the transmitting roller 1 and the servo motor driving the receiving roller 2.

[0079] When determining slippage, record the moments when the torque increases and decreases significantly, denoted as T_start and T_end, respectively, and calculate the time difference δT = T_end - T_start. Based on the current roller angular velocity ω and radius, calculate the theoretical linear velocity v = ω × r, and then calculate the theoretical workpiece length L_theoretical = v × δT. Compare L_theoretical with the known actual workpiece length L_actual: if L_theoretical > L_actual, then slippage has occurred during this period.

[0080] In the event of a loss of synchronization or slippage, the system directly removes the abnormal transmission signals by recording the duration of the torque anomaly. Specifically, once slippage is detected, the system automatically identifies the image frames acquired during the slippage period and removes these invalid frames from the image sequence. The image frames acquired during slippage specifically refer to the image region imaged based on the corrected transmission signals acquired during the slippage period. Subsequently, the system reinitializes the image reconstruction process based on the stable signal after the slippage ends, ensuring the continuity and accuracy of subsequent images. Simultaneously, the system prompts the user to check the mechanical transmission status and perform maintenance if necessary.

[0081] It should be noted that, in the current roller pair calibration process, those skilled in the art, based on the general understanding of existing technology, can typically only propose technical solutions that rely on a preset angular synchronization relationship between the rollers. However, under the actual operating conditions of a high-speed production line, the mechanical transmission chain inevitably has accumulated errors. In addition, during the transmission of workpiece 5, it may experience momentary slippage due to tension fluctuations or changes in surface condition, and even the servo motor may lose synchronization due to sudden load changes. All of these factors will cause the actual relative angle between the rollers to deviate from the theoretical preset value, rendering the calibration model based on a fixed angular relationship ineffective.

[0082] If the angle synchronization relationship is misaligned, the generated correction parameters will fail to accurately reflect the true motion state, leading to problems such as spatial distortion, phase misalignment, or pixel mismatch in the ultrasonic image. This not only significantly reduces imaging resolution and geometric fidelity but also easily results in false or missed detections of defects, making it difficult to meet the stringent requirements of modern industry for ultrasonic imaging inspection in terms of high precision, high stability, and high reliability.

[0083] In contrast, this invention constructs a correction matrix group decoupled from the roller pair's speed. When correcting the transmitted signal, it calculates the correction gain coefficient corresponding to each receiving array element based on the roller pair's operating rotation state. This calculated gain coefficient is then used to correct the received transmitted signal, enabling precise and adaptive correction based on the instantaneous state of the rollers during actual operation, rather than relying on fixed preset angle relationships. Regardless of complex interference conditions such as transmission errors, slippage, or motor step loss, this invention can effectively identify and compensate for the resulting motion deviations, thus ensuring the spatial consistency and phase accuracy of the ultrasonic image. This significantly improves the defect detection rate and detection robustness, fully meeting the practical needs of high-speed, high-precision industrial online inspection.

Claims

1. A calibration method suitable for transmission ultrasound scanning systems, characterized in that, The correction method is used to correct the transmission signal generated by any roller pair in a transmission ultrasonic scanning system during ultrasonic scanning, the correction method comprising: The scanning features of a pair of rollers within a transmission ultrasonic scanning system are acquired, and a set of correction matrices is constructed based on the physical length of the corresponding surface of the current roller pair. The correction matrix set includes the transmission correction matrix corresponding to the transmitting drum in the current drum pair and the reception correction matrix corresponding to the receiving drum in the current drum pair. The transmission correction matrix includes several physical lengths of the transmission correction surface that characterize the correction rotation angle of the transmission drum, and the transmission calibration signal corresponding to each receiving array element under each physical length of the transmission correction surface. The receiving correction matrix includes several physical lengths of receiving correction surfaces that characterize the rotation angle of the receiving drum correction, and a receiving calibration signal corresponding to each receiving array element under each physical length of the receiving correction surface. During transmission signal correction, the current operating rotation state of the roller pair is determined, and the correction gain coefficient corresponding to each receiving array element is obtained based on the current operating rotation state of the roller pair. The obtained correction gain coefficient is then used to correct the received transmission signal. The correction gain coefficient is generated based at least on the transmitted calibration signal and the received calibration signal associated with the current working rotation state of the drum; When the current working rotation state of the roller pair is synchronized with the correction rotation state used when constructing the correction matrix group, it includes: The transmission rotation angle of the transmitting drum is obtained, and the physical length of the corresponding transmitting working surface is calculated based on the drum scanning characteristics and the transmission rotation angle, so as to extract the transmission calibration signal that corresponds to the physical length of the transmitting working surface for each receiving array element in the transmission calibration matrix. The receiving rotation angle of the receiving drum is obtained, and the physical length of the corresponding receiving working surface is calculated based on the drum scanning characteristics and the receiving rotation angle, so as to extract the receiving calibration signal that corresponds to the physical length of the receiving working surface for each receiving array element in the receiving calibration matrix. For each receiving array element, the transmitted calibration signal extracted above is multiplied by the received calibration signal, and the product is used as the correction gain coefficient corresponding to the current reception. When the current working rotation state of the roller pair is not synchronized with the correction rotation state used when constructing the correction matrix group, including: Obtain the transmitting rotation angle of the transmitting drum and the receiving rotation angle of the receiving drum. The physical length of the transmitting working surface is calculated based on the drum scanning characteristics and the transmitting rotation angle. The transmitting gain of each receiving array element is then calculated based on this physical length. When calculating the transmit gain of each receiving element, the reference receiving element associated with the current receiving element and the corresponding transmit calibration signal are determined within the transmit correction matrix based on the physical length of the transmitting working surface. Based on all reference calibration array elements and their corresponding transmit calibration signals, the transmit gain of the current receive array element is calculated and generated. The method for calculating and generating the transmit gain includes at least linear interpolation. The physical length of the receiving working surface is calculated based on the drum scanning characteristics and the receiving rotation angle. The receiving gain of each receiving array element is then calculated based on this physical length. When calculating the receiving gain of each receiving array element, the reference receiving array element associated with the current receiving array element and the receiving calibration signal corresponding to the reference receiving array element are determined within the receiving calibration matrix based on the physical length of the receiving calibration surface. Based on all reference calibration array elements and their corresponding receive calibration signals, the receive gain of the current receive array element is calculated, wherein the method for calculating the receive gain includes at least linear interpolation. For each receiving array element, the transmit gain calculated above is multiplied by the receive gain, and the product is used as the correction gain coefficient for the current receiving array element.

2. The calibration method suitable for a transmission ultrasound scanning system according to claim 1, characterized in that, For each receiving element, if the physical length of the transmitting working surface matches the physical length of a transmitting correction surface within the transmitting correction matrix, then the current receiving element is configured as the reference receiving element, and the corresponding transmitting correction signal is used as the transmitting gain. Otherwise, bilinear interpolation is used to calculate the transmitting gain. When using bilinear interpolation to calculate the transmit gain, the reference receive elements associated with the current receive element are determined to be at least a first reference receive element and a second reference receive element, wherein... If the current receiving array element is the first element, then the first receiving array element is the receiving array element with the current array number, and the second receiving array element is the receiving array element one position below the first element. If the current receiving array element is the last one in the arrangement sequence, then the first receiving array element is used as the reference for the current receiving array element, and the second receiving array element is used as the reference for the receiving array element whose arrangement sequence number is one position above the last one. If the current receiving array element's arrangement number is between the first and last positions, then the first receiving array element is the receiving array element one position below the current receiving array element's arrangement number, and the second receiving array element is the receiving array element one position above the current receiving array element's arrangement number. The physical length of the launch working surface is searched downwards to generate the first reference physical length for launch, wherein the first reference physical length for launch is less than or equal to the physical length of the launch working surface and is the same as the physical length of a launch correction surface within the launch correction matrix; Based on the first reference physical length for launch, a second reference physical length for launch is configured, wherein the second reference physical length for launch is the same as the physical length of a launch correction surface within the launch correction matrix, the second reference physical length for launch is greater than the first reference physical length for launch, and the second reference physical length for launch and the first reference physical length for launch are adjacent to each other within the launch correction matrix; Based on the corresponding transmission calibration signals of the first reference receiving array element at the first and second reference physical lengths of transmission, and the corresponding transmission calibration signals of the second reference receiving array element at the first and second reference physical lengths of transmission, the corresponding transmission gain is generated by bilinear interpolation.

3. The calibration method suitable for a transmission ultrasound scanning system according to claim 2, characterized in that, When bilinear interpolation is used to calculate the transmit gain, then: in, For the arrangement sequence number The current receiving array element and the transmission gain calculated based on the corresponding physical length of the transmitting working surface. The physical length of the launching working surface. This is the current array element's sequence number. For the first reference physical length of launch, For the second reference physical length of launch, For reference, the arrangement sequence number of the first receiving array element, For reference, the arrangement sequence number of the second receiving array element, To reference the transmission calibration signal corresponding to the first receiving array element at the first reference physical length, To reference the transmission calibration signal corresponding to the first receiving element at the second reference physical length, To reference the transmission calibration signal corresponding to the second receiving element at the first reference physical length, This is to reference the transmission calibration signal corresponding to the second receiving array element at the second reference physical length.

4. The calibration method suitable for a transmission ultrasound scanning system according to claim 1, characterized in that, The current roller scanning features of the roller pair include at least the physical features of the rollers, wherein the physical features of the rollers include the perimeter and radius of the cross-section corresponding to the transmitting roller, and the perimeter and radius of the cross-section corresponding to the receiving roller; An array of transmitting elements is set inside the transmitting drum. The transmitting elements are arranged along the axial direction of the transmitting drum. The receiving elements are arranged in an array along the axial direction of the receiving drum. The number of transmitting elements is the same as the number of receiving elements, and the corresponding positions of the transmitting elements and receiving elements correspond to each other. The constructed set of correction matrices includes: The number of physical lengths of the transmit correction surface within the transmit correction matrix corresponds to the number of physical lengths of the receive correction surface within the receive correction matrix, and there is a one-to-one correspondence between the physical lengths of the transmit and receive correction surfaces. The rotation angle of the transmitting drum, represented by the physical length of each transmitting correction surface, is consistent with the rotation angle of the receiving drum, represented by the physical length of the corresponding receiving correction surface.

5. The calibration method suitable for a transmission ultrasound scanning system according to claim 4, characterized in that, When constructing the emission correction matrix, the following are included: Both the transmitting drum equipped with the transmitting array and the receiving array are placed in the calibration coupling medium, and the receiving array is directly corresponding to the transmitting array in the transmitting drum. The drive launch drum rotates one revolution in angular steps according to the correction rotation angle. After each revolution, the physical length of the corresponding launch correction surface is calculated based on the radius of the launch drum's cross-section. After each rotation, the transmitting array is configured to transmit ultrasonic signals to the receiving array, and the intensity of the transmitted signal received by each receiving element is calculated. The intensity of the transmitted signal received by each receiving element is used as the transmission calibration reference signal under the current physical length of the transmission calibration surface. After obtaining the physical lengths of all transmit calibration surfaces and the transmit calibration reference signal for each receive element at each physical length of the transmit calibration surface, all transmit calibration reference signals are normalized so that each normalized value is configured as a transmit calibration signal.

6. The calibration method suitable for a transmission ultrasound scanning system according to any one of claims 1 to 5, characterized in that, Before the current roller pair is operational or a correction matrix group is constructed within a transmission ultrasonic scanning system, the physical state of the current roller pair is checked, including... When checking the physical status of the current roller pair, the following is included: The current roller pair is configured to be unloaded and the transmitting roller and receiving roller inside the current roller pair are brought into contact. Then, the transmitting roller and receiving roller are driven to rotate one revolution and the transmission signal under the unloaded state is collected. A background ultrasound image is reconstructed based on all transmitted signals; The uniformity characteristics of the background ultrasound images are analyzed, and the physical state of the current roller pair is determined based on the analyzed uniformity characteristics. The physical state of the current roller pair includes normal rollers, abnormal rollers, and degraded state. When the physical state is in a state degradation state, the correction matrix group of the current roller pair is reconstructed.

7. The calibration method suitable for a transmission ultrasound scanning system according to claim 6, characterized in that, After the transmission ultrasonic scanning system performs an ultrasonic scan on the workpiece, it performs ultrasonic imaging based on all corrected transmission signals to generate a scanned target image of the workpiece. When performing ultrasonic imaging based on all corrected transmission signals, a basic image of the workpiece is first generated. Then, the basic image of the workpiece is subjected to at least bilinear interpolation to generate a target image of the workpiece.

8. The calibration method suitable for a transmission ultrasound scanning system according to claim 7, characterized in that, During the ultrasonic scanning of the workpiece using a transmission ultrasonic scanning system, slippage detection is also performed on the contact between the workpiece and the roller pair. When slippage occurs, image frames acquired during the slippage period are identified and discarded, and ultrasonic imaging is performed based on the discarded image frames to generate a target image of the workpiece.

Images