Columnar defect three-dimensional imaging method and device and storage medium

Abstract

The invention provides a columnar defect three-dimensional imaging method, and relates to the technical field of nondestructive testing, a plurality of receiving sensors are arranged on the edge of the surface of an object to be detected to define a detection area, and a plurality of excitation sensors are arranged in the detection area, so that the number of the sensors and the number of channel combinations are reduced while the good detection capability is maintained, and the detection efficiency is improved. The data acquisition and processing efficiency is improved, and the detection cost is reduced. The normalization processing step ensures the consistency of different channel signals in amplitude. In the sound time correction step, the propagation time of the excitation signal is accurately corrected, so that the accuracy of delay calculation is improved, and the positioning of deep defect and complex defect boundaries is more reliable. And finally, independent delay and superposition focusing are carried out on each pixel point in the detection area by adopting a full-focusing algorithm, so that dynamic focusing and global imaging are realized, the dependence of a traditional focusing method on a focus position is avoided, and the three-dimensional morphological characteristics of the columnar defect can be completely and clearly reconstructed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of non-destructive testing, in particular to a columnar defect three-dimensional imaging method, device and storage medium. BACKGROUND

[0002] Macroscopic isotropic materials such as concrete and granite have advantages of impact resistance, excellent dynamic performance, low construction cost, high technical maturity, etc., and are widely used in key shelter structures, bridge piers, nuclear waste disposal libraries and other military or civilian fields, solving the mechanical bearing problems in multiple fields. However, the complexity of the damage of these material structures poses a challenge to health monitoring, mainly including defect type combination, difficulty in evaluating material damage degree, inability to simply simulate mechanical loading experiments, and inability to establish a connection between mechanical performance parameters and detection data. Early solutions based on acceleration, speed and other sensors solved the problems of long-distance signal penetration and real-time health monitoring, but had the bottlenecks of low signal acquisition frequency, high background noise interference, and fast decay of low-frequency signals, which reduced the positioning of defects.

[0003] Based on this, as a non-destructive testing technology, ultrasonic imaging technology has begun to be applied in the health monitoring field of small buildings or geotechnical engineering. Ultrasonic imaging can capture defects inside the material with high precision and high sensitivity, and has high application value, especially in evaluating structural reliability, geological stability and damage and repair of ancient cultural relics or buildings. Traditional ultrasonic scanning imaging technology can intuitively display the position and size of defects, and can reconstruct the internal defects of the measured object. However, the signal processing accuracy of these imaging technologies is low, and the imaging effect is not ideal, which cannot provide accurate data for physical analysis.

[0004] In order to solve the above problems, synthetic aperture focused ultrasonic imaging (SAFT) technology emerged as the times require, successfully solving the two-dimensional positioning problem of columnar through-hole defects. However, it still cannot accurately position the depth and three-dimensional position of the defects. In 2005, Holmes et al. developed a full-focus imaging technology (TFM), which collects data through full-matrix capture and implements delay-and-sum processing, thereby achieving higher resolution and better imaging performance. The TFM technology has the advantages of not needing complex control circuit and being able to dynamically focus each pixel, solving the problem of focus dependence of traditional methods, and being able to accurately position three-dimensional defects. However, the algorithm based on time-domain delay-and-sum still has problems such as complex calculation model, large calculation load and complex imaging algorithm, which limits its popularization and application in macroscopic isotropic materials.

[0005] In addition, in the ultrasonic three-dimensional imaging of internal defects of macroscopic materials, some studies acquire data by a one-dimensional linear array probe and combine full focusing algorithm for layered slice processing to obtain three-dimensional imaging results. However, this method is low in efficiency and needs multiple scanning and data acquisition. The use of the linear array probe has the disadvantages of complex beam control, high detection cost and low imaging efficiency, and cannot meet the needs of large-scale detection.

[0006] Therefore, the prior art still has certain technical bottlenecks in the application of ultrasonic imaging of macroscopic isotropic materials, and a new technology or method is urgently needed to improve imaging efficiency, reduce cost and solve the problems of depth and three-dimensional positioning. SUMMARY

[0007] The purpose of the present application is to provide a columnar defect three-dimensional imaging method, device and storage medium to solve the above technical problems.

[0008] To achieve the above purpose, the technical scheme adopted by the present application is as follows: The present application provides a columnar defect three-dimensional imaging method, which comprises the following steps: Signal acquisition: arranging a plurality of excitation sensors and a plurality of receiving sensors on the surface of the object to be measured, arranging the plurality of receiving sensors around the edge of the surface of the object to be measured to form a detection area, arranging the plurality of excitation sensors in the detection area, transmitting excitation signals by the excitation sensors, propagating the excitation signals through the object to be measured and reflecting the excitation signals back to the receiving sensors to form receiving signals, removing the bottom echo signals in the receiving signals, and obtaining detection signals for characterizing the columnar defects inside the object to be measured; Normalization processing: performing amplitude normalization processing on the detection signals to obtain effective detection signals; Sound time correction: performing envelope processing on the excitation signals to obtain the time corresponding to the maximum amplitude of the upper envelope line of the excitation signals, correcting the transmission time in the propagation process of the excitation signals, and obtaining effective transmission time; Image generation: based on the effective detection signals and the effective transmission time, using a full focusing algorithm to obtain a pixel value matrix of the detection area, and generating a three-dimensional ultrasonic image of the columnar defects inside the object to be measured.

[0009] Further, in the signal acquisition, the number of receiving sensors is greater than or equal to 4.

[0010] Further, in the signal acquisition, the detection area is a rectangle, and the horizontal distance between each excitation sensor and the center of the detection area is less than or equal to 60 percent of the width of the detection area.

[0011] Further, in the normalization processing, the calculation formula for amplitude normalization processing of the detection signals is:

[0012] wherein, is the transmission time of the excitation signal, is the detection signal formed after the excitation signal emitted by the jth excitation sensor reaches the ith receiving sensor, is the maximum amplitude of the detection signal, is the effective detection signal obtained after normalization processing of the detection signal.

[0013] Further, in the acoustic time correction, the calculation formula for correcting the transmission time of the excitation signal in the propagation process is:

[0014]

[0015] wherein, is the coordinate of any point P in the detection area, is the coordinate of the ith receiving sensor, is the coordinate of the jth excitation sensor, is the propagation speed of the excitation signal in the object to be measured, is the transmission time of the detection signal formed after the excitation signal emitted by the jth excitation sensor reaches the ith receiving sensor through point P, and PPD is the time corresponding to the maximum amplitude of the upper envelope line of the excitation signal, is the effective transmission time of the detection signal formed after the excitation signal emitted by the jth excitation sensor reaches the ith receiving sensor through point P.

[0016] Further, the image generation specifically includes: dividing the detection area into a plurality of uniform grids, and taking each grid point as a pixel point; calculating the pixel value of each pixel point according to the effective detection signal and the effective transmission time of each pixel point; performing maximum and minimum value normalization processing on the pixel value of each pixel point to obtain a pixel value matrix; generating a three-dimensional ultrasonic image of the internal columnar defect of the object to be measured according to the pixel value matrix.

[0017] Further, in the image generation, there are pixel points:

[0018]

[0019]

[0020] wherein, are the maximum values in respectively, They are respectively The minimum value in, The size of each grid.

[0021] Furthermore, in image generation, the formula for calculating the pixel value of each pixel is as follows:

[0022] in, Let Q be the coordinates of any pixel. This indicates that the excitation signal emitted by the j-th excitation sensor reaches the ith receiving sensor via pixel Q, forming an effective detection signal. The effective transmission time for the excitation signal emitted by the j-th excitation sensor to travel through pixel Q to the i-th receiving sensor to form the detection signal is given. Let Q be the pixel value of pixel point Q.

[0023] Furthermore, in image generation, the formula for normalizing the pixel values ​​of each pixel by applying the maximum and minimum values ​​is as follows: .

[0024] in, pixel value The maximum value in, pixel value The minimum value in, This is a matrix of pixel values.

[0025] This application also provides a terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The computer program is loaded and executed by the processor to implement any of the methods described above.

[0026] This application also provides a storage medium storing a computer program, which is loaded and executed by a processor to implement any of the methods described above.

[0027] The beneficial effects of this application include: The application provides a columnar defect three-dimensional imaging method. A plurality of receiving sensors are arranged on the surface edge of an object to be detected to form a detection area, and a plurality of excitation sensors are arranged in the detection area. Under the premise of maintaining sufficient aperture coverage, detection capability comparable to that of a phased array technology can still be obtained, the number of sensors and the number of channel combinations are significantly reduced, the scale of data acquisition and storage is fundamentally reduced, the calculation load is reduced, and the detection efficiency is improved, so that the hardware investment and engineering implementation cost are reduced. The normalization processing step ensures the consistency of the amplitudes of different channel signals, effectively eliminates the bias caused by differences in coupling conditions or fluctuations in channel gain, and provides stable and reliable data input for subsequent imaging algorithms. The sound time correction step improves the accuracy of delay calculation by accurately correcting the propagation time of the excitation signal, so that the positioning of deep defects and complex defect boundaries is more reliable. Finally, by using a full-focusing algorithm to independently delay and superimpose focus on each pixel point in the detection area, dynamic focusing and global imaging are realized, the dependence on the focus point position of the traditional focusing method is avoided, and the three-dimensional morphological characteristics of the columnar defect can be completely and clearly reconstructed. Compared with the traditional 48-probe phased array, the detection range of the application is 240mmx200mm, and the maximum data amount is 150 groups. The detection range is 1.62 times that of the 48-probe phased array, and the data amount is only 6.51% of that of the 48-probe phased array. Moreover, the three-dimensional imaging result is more intuitive, and it is convenient to comprehensively judge the geometric characteristics, distribution range and possible expansion trend of the defect.

[0028] Overall, by reducing the number of sensor arrangements and combining with the full-focusing imaging technology, the application achieves a balance between detection efficiency, imaging accuracy and engineering implementability, and can detect internal defects of large-size isotropic material thick members such as concrete and granite in a large range and quickly. It is suitable for reliable and low-cost three-dimensional positioning and evaluation of columnar defects in various engineering scenes. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 A flowchart of a columnar defect three-dimensional imaging method provided by the application; Figure 2 A schematic diagram of the arrangement of a receiving sensor on an object to be detected provided by the application; Figure 3 A schematic diagram of the propagation process of an excitation signal; Figure 4 A three-dimensional imaging diagram of six groups of detection signals; Figure 5 A three-dimensional imaging diagram of twelve groups of detection signals.

[0030] Figure legend: 1-object to be detected; 2-receiving sensor. DETAILED DESCRIPTION The application provides a columnar defect three-dimensional imaging method, which comprises the following steps of Figure 1 As shown in the figure, the method comprises the following steps of S1, signal acquisition: arranging a plurality of excitation sensors and a plurality of receiving sensors 2 on the surface of the object to be detected 1, the plurality of receiving sensors 2 are arranged along the edge of the surface of the object to be detected 1 to form a detection area, the plurality of excitation sensors are arranged in the detection area, the excitation signals emitted by the excitation sensors propagate through the object to be detected 1 and are reflected back to the receiving sensors 2 to form receiving signals, the bottom echo signals in the receiving signals are removed, and detection signals for characterizing the columnar defects inside the object to be detected 1 are obtained.

[0031] Specifically, the object to be detected 1 is a large-size thick component, and the material of the thick component is approximately isotropic on a macroscopic scale, such as a building wall or a bridge pier, which meets the assumption premise of equivalent isotropic propagation of ultrasonic waves in the medium. A sparse ultrasonic sensor array is constructed on the top surface of the object to be detected 1 to obtain sufficient angular spectrum coverage and equivalent aperture with a limited number of channels. The sparse ultrasonic sensor array comprises a plurality of excitation sensors and a plurality of receiving sensors 2, the plurality of receiving sensors 2 are arranged along the edge of the top surface to increase the incident and receiving angle distribution of the internal reflector. The plurality of receiving sensors 2 enclose the detection area along the surface edge, and the plurality of excitation sensors are arranged in the detection area. By transmitting excitation signals through different excitation sensors, a plurality of independent transmission and receiving paths can be formed. A Cartesian coordinate system is established with the projection of the center of the detection area on the bottom surface as the origin, and the positive direction of the z-axis is perpendicular to the bottom surface of the object to be detected 1 upward. The coordinates of each excitation sensor and each receiving sensor 2 are recorded to provide accurate path lengths for subsequent delay calculation.

[0032] During each excitation, one or more excitation sensors generate excitation signals with known center frequency and bandwidth, the excitation signals are ultrasonic waves, the excitation signals propagate in the object to be detected 1, and when encountering columnar defects inside the object to be detected 1 or the bottom surface, reflections are generated. The energy returned to the surface is converted by a plurality of receiving sensors 2 to form receiving signals, and the receiving signals are electrical signals. In order to suppress strong reflections unrelated to defects, bottom echo elimination and noise suppression are performed on the receiving signals, and only detection signals for characterizing the position, size and shape of the columnar defects inside the object to be detected 1 are retained.

[0033] In order to ensure the stability of the whole imaging chain, the scale and geometry of the sparse ultrasonic sensor array need to be matched with the size of the monitoring area. When the number of receiving sensors 2 is not less than four, a basic angular spectrum coverage can be formed. For example, when the object to be detected 1 is a cube, one receiving sensor 2 can be arranged at the midpoint of each of the four edges of the top surface; in order to adapt to larger coverage requirements, the number can be expanded to six, eight, ten or more, and the expansion strategy is to improve the uniformity of coverage and the diversity of paths while maintaining the sparsity.

[0034] As Figure 2 shown in the embodiment, the object to be measured 1 is a cuboid, the top surface of which is a rectangle with a length of 2a and a width of 2b, and a > b. One receiving sensor 2 is arranged at the midpoint of each of the two short sides of the rectangle, and two receiving sensors 2 are arranged at intervals along the edge direction of each of the two long sides of the rectangle, so that the surface of the detection area formed thereby is the rectangular top surface of the object to be measured 1, that is, the length of the detection area is 2a and the width is 2b. The center coordinates of the detection area are set as (0, 0, z), so that the arrangement of the receiving sensors 2 is as uniform as possible. The coordinates of the receiving sensors 2 arranged at the two short sides of the rectangle are (-a, 0, z) and (a, 0, z) respectively, and the coordinates of the receiving sensors 2 arranged at the two long sides of the rectangle are (-c, b, z), (c, b, z), (c, -b, z) and (-c, -b, z) respectively, wherein z is the height of the cuboid, and the relationship between the coordinates of the receiving sensors 2 arranged at the long sides of the rectangle along the x-axis and the length and width of the rectangle is as follows:

[0035] In addition, the horizontal distance between each excitation sensor and the center of the detection area is less than or equal to 60% of the width of the detection area, that is, all the excitation sensors are arranged within a circular range with the center at the center (0, 0, z) of the rectangular detection area and a radius of 0.6b, and the interval between adjacent two excitation sensors needs to be greater than the diameter of the ultrasonic probe. In this way, by reasonably arranging the sensors, the effective aperture can be expanded and the sensitivity to different azimuth scattering can be improved without significantly increasing the number of channels. The number of excitation sensors is configured according to the target imaging resolution and positioning accuracy; when the positioning accuracy of the columnar defect is required to be high, the number of excitation sensors is increased to improve the angular spectrum coverage and equivalent aperture; when the accuracy requirement is low, the number of excitation sensors is appropriately reduced to reduce the acquisition and calculation load, thereby improving the detection efficiency and maintaining the array sparsity.

[0036] In summary, the sparse array realizes sufficient coverage of the monitoring area under the premise of controlled number of channels through reasonable edge and internal collaborative arrangement, the bottom echo rejection improves the signal-to-noise ratio of the effective echo, reduces the burden of subsequent algorithms, and is suitable for rapid screening and fine interpretation of engineering sites of macroscopically isotropic materials such as concrete and rock.

[0037] S2, normalization processing: amplitude normalization processing is performed on the detection signal to obtain an effective detection signal.

[0038] Specifically, to ensure the stability of the subsequent acoustic time retrieval and full focus superposition, the cross-channel original detection signals obtained need to establish a unified scale in the amplitude level. Differences exist in coupling state, electronic gain, propagation path and incident angle between different channels, and direct superposition can easily produce amplitude bias and weaken the pixel focusing effect. Based on the consistency of the maximum value of sampling, the amplitude of each detection signal is normalized to obtain an effective detection signal, so that each effective detection signal remains consistent in dimension and dynamic range. The specific calculation formula of amplitude normalization is:

[0039] Among them, is the transmission time of the excitation signal, is the detection signal formed after the excitation signal emitted by the jth excitation sensor reaches the ith receiving sensor 2, is the maximum amplitude of the detection signal, is the effective detection signal obtained after normalization of the detection signal.

[0040] By using the amplitude normalization processing method, the cross-channel amplitude difference can be reduced without increasing the number of channels and hardware complexity, the visibility of small amplitude scatterers can be enhanced, the energy contrast of the end of the columnar defect and the side wall can be stabilized, and the consistency and repeatability of three-dimensional imaging can be improved. Compared with the unnormalized delay superposition, the pixel intensity distribution is closer to the structure scattering intrinsic rather than the channel gain, and the robustness of threshold segmentation and connectivity analysis is improved. In addition, the dynamic range after normalization is more suitable for unified threshold and automated processing flow, which can shorten the acquisition and reconstruction time while ensuring the resolution and positioning accuracy, and effectively control the overall detection cost.

[0041] S3, acoustic time correction: envelope processing is performed on the excitation signal to obtain the time corresponding to the maximum amplitude of the upper envelope line of the excitation signal, so as to correct the transmission time in the propagation process of the excitation signal to obtain the effective transmission time.

[0042] Specifically, after completing the bottom echo elimination and amplitude normalization, the time reference of the cross-channel signal still has deviations. If no calibration is added, systematic differences will occur between geometric delay and real arrival time, which will directly affect the coherence of full focus superposition and the geometric accuracy of three-dimensional reconstruction. As shown in Figure 3 , by performing envelope processing on the excitation signal, the time corresponding to the maximum amplitude of the upper envelope line of the excitation signal can be obtained. This time is used as a correction parameter to correct the transmission time in the propagation process of the excitation signal to obtain the effective transmission time. The specific calculation formula is:

[0043]

[0044] wherein, is the coordinate of any point P in the detection region, is the coordinate of the ith receiving sensor 2, is the coordinate of the jth excitation sensor, is the propagation speed of the excitation signal in the object 1, is the transmission time of the detection signal formed by the excitation signal emitted by the jth excitation sensor passing through point P to reach the ith receiving sensor 2, and PPD is the time corresponding to the maximum amplitude of the upper envelope of the excitation signal, is the effective transmission time of the detection signal formed by the excitation signal emitted by the jth excitation sensor passing through point P to reach the ith receiving sensor 2.

[0045] By correcting the sound time, not only is the time reference unified across channels, but the coherence of the delay-and-sum is also significantly improved. In subsequent full-focus three-dimensional imaging, the delay retrieval of each pixel can be referenced to the true propagation time, the energy of the focal point is more concentrated, and the three-dimensional image generated ultimately can more clearly exhibit the geometric characteristics and depth information of the columnar defects, and the positioning result is more accurate. In particular, under the condition of a sparse sensor array, the corrected time consistency helps to maintain a high imaging resolution and contrast.

[0046] S4, image generation: processing the effective detection signal using a full-focus algorithm to obtain a pixel value matrix and generate a three-dimensional ultrasonic image of the columnar defects inside the object 1.

[0047] Further, the image generation specifically includes: S41: dividing the detection region into a plurality of uniform grids, and taking each grid point as a pixel point.

[0048] First, the detection region is subjected to spatial discretization processing: according to the monitoring range and the desired resolution, a three-dimensional boundary is defined in a given coordinate system, and the boundary is divided into uniform grids with a uniform step size, and each grid point is regarded as a pixel to be reconstructed. The size of the grid depends on the requirement for positioning accuracy and the size of the monitoring region, and the grid and pixel size are determined according to the center frequency, the effective aperture and the defect size, so that the spatial sampling is matched with the axial and lateral resolution of the system. For example, a detection region of 1 m x 1 m x 1 m can be divided into 1000 x 1000 grids, and the size of the grid is 1 mm, and the positioning accuracy can be further improved by subdividing the grid. Through this spatial modeling, the detection region obtains a clear mapping of pixel index and three-dimensional coordinates, providing a stable geometric reference for subsequent pixel-by-pixel delay retrieval. In this embodiment, there are pixel points:

[0049]

[0050]

[0051] wherein, is the maximum value in the set of pixel values is the minimum value in the set of pixel values is the minimum value in the set of pixel values is the minimum value in the set of pixel values is the size of each grid.

[0052] S42: Calculate the pixel value of each pixel point according to the effective detection signal and the effective transmission time of each pixel point.

[0053] The calculation of the pixel value is based on the full-focus algorithm, that is, for each pixel point, the aforementioned calibrated effective transmission time is used to perform delay retrieval and amplitude aggregation on all transmission and reception channels. Specifically, the normalized effective detection signal amplitude is read at the corresponding time position for each pair of channels, and the contributions of each channel are coherently superimposed at the pixel point to obtain the pixel value. This process is independently performed for all pixel points to form a global reconstruction of dynamic focusing, which is not constrained by the fixed focus position and can simultaneously consider shallow and deep reflection targets. The specific calculation formula is:

[0054] wherein, is the coordinate of any pixel point Q, represents the effective detection signal formed by the excitation signal emitted by the jth excitation sensor reaching the ith receiving sensor 2 through the pixel point Q, is the effective transmission time of the detection signal formed by the excitation signal emitted by the jth excitation sensor reaching the ith receiving sensor 2 through the pixel point Q, is the pixel value of the pixel point Q.

[0055] S43: Perform maximum and minimum value normalization processing on the pixel value of each pixel point to obtain a pixel value matrix.

[0056] In order to facilitate display and subsequent threshold segmentation, scale unification processing is required for the pixel value. Maximum and minimum value normalization is adopted to linearly map the intensity of the entire pixel set to a fixed interval to obtain a three-dimensional pixel value matrix. This matrix can be used as a rendering input and as basic data for quantitative analysis. The specific calculation formula of maximum and minimum value normalization is: .

[0057] wherein, is the maximum value in the set of pixel values is the minimum value in the set of pixel values is the minimum value in the set of pixel values The minimum value in, This is a pixel value matrix with a size of . .

[0058] S44: Generate a three-dimensional ultrasonic image of the columnar defect inside the object under test 1 based on the pixel value matrix.

[0059] Based on the pixel value matrix, a three-dimensional ultrasonic image of the interior of the object under test 1 is generated through rendering and spatial reconstruction. During the imaging process, high-intensity connected regions can be visualized using methods such as volume rendering, surface reconstruction, or orthogonal slicing, so that the columnar defect is presented with a clear geometric shape in the three-dimensional coordinate system, and the end coordinates of the columnar defect are located. In order to highlight the spatial shape of the columnar defect, a fixed threshold can be set on the normalized matrix or an adaptive threshold based on histogram can be used to extract high-energy connected regions; then, a volume view is generated using connected domain analysis and three-dimensional surface reconstruction, and orthogonal slicing, maximum projection, or volume rendering is used to display the axial extension and end boundary. For engineering interpretation needs, indicators such as geometric center, axial length, radial shape, and burial depth can be calculated in the connected domain to form a structured result that can be used for evaluation. In this embodiment, in order to highlight the spatial distribution and boundary characteristics of the columnar defect, an intensity threshold is set on the pixel value matrix that has been normalized to the range of [0, 1], and pixels with pixel values ​​not less than 0.9 are selected to participate in three-dimensional rendering and target extraction, thereby suppressing the interference of background and low-correlation scattering on imaging and enhancing the salience of the defect location.

[0060] In this embodiment, an isotropic material sample of 400 mm × 300 mm × 200 mm was used as the test object 1 for verification. A detection area of ​​240 mm × 120 mm × 200 mm was divided into 480 × 240 × 400 grids, each grid with a size of 0.5 mm. A columnar defect with a diameter of 30 mm penetrated vertically inward from the bottom of the test object 1 for 75 mm, with the center coordinates of the defect end being (-20, -40, 75). The coordinates of the six receiving sensors 2 were (-60, -100, 200), (-120, 0, 200), (-60, 100, 200), (60, 100, 200), (120, 0, 200), and (60, -100, 200). When the excitation sensor was placed at the coordinates (-40, -80, 200) for excitation, the imaging results of the six sets of detection signals were as follows: Figure 4 As shown, the positioning coordinates of the end of the columnar defect are (-23.5, -44.5, 45). Compared with the actual position, the absolute error is 30.54 mm and the relative error is 8.23%, which meets the requirement of accuracy not exceeding 10%.

[0061] Further, when the excitation sensors are arranged at positions with coordinates (0, 0, 200) and (-40, -80, 200) respectively for excitation, the imaging results of the twelve groups of detection signals are as shown in Figure 5 The end positioning coordinates of the columnar defect are (-32, -60.5, 60), and compared with the real position, the absolute error is 26.61 mm, and the relative error is 7.17%, which is still within the 10% precision threshold. However, the error of the imaging result is obviously reduced by data fusion of multiple excitation positions compared with single excitation.

[0062] In summary, the method of the present application can still achieve reliable imaging and end positioning of columnar defects in thick members of larger size under the condition of a sparse array using only a limited number of sensors. When single-point excitation is used, the imaging result already meets the engineering requirements, and when multiple-point excitation is used, the positioning accuracy is further improved. Compared with the traditional 48-probe phased array, the detection range is 330 mm x 90 mm, and the data amount is 48 x 48 = 2304 groups, while the detection range of the present application is 240 mm x 200 mm, and the maximum data amount is 6 x 25 = 150 groups. The detection range is 1.62 times that of the 48-probe phased array, and the data amount is only 6.51% of that of the 48-probe phased array. Thus, the present method not only greatly reduces the number of sensors and the amount of data, reducing the detection cost and computational burden, but also maintains a high level in imaging resolution and positioning accuracy. The embodiment verifies the applicability of the method in macroscopically isotropic materials such as concrete and granite, and shows that the method can balance detection efficiency and accuracy under engineering site conditions, meeting the demand for three-dimensional positioning of internal defects in thick members, and providing a feasible solution for structural health monitoring and engineering safety evaluation.

[0063] The present application also provides a terminal device, which comprises a memory, a processor, and a computer program stored on the memory and executable on the processor. The computer program is loaded and executed by the processor to implement any of the above methods.

[0064] The present application also provides a storage medium, which stores a computer program. The computer program is loaded and executed by the processor to implement any of the above methods.

[0065] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Those skilled in the art can make various modifications and changes to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A three-dimensional imaging method for columnar defects, characterized in that, The methods include: Signal acquisition: Several excitation sensors and several receiving sensors are arranged on the surface of the object to be tested. Several receiving sensors are arranged at the edge of the surface of the object to be tested to form a detection area. Several excitation sensors are arranged in the detection area. The excitation signal emitted by the excitation sensor is propagated through the object to be tested and reflected back to the receiving sensor to form a receiving signal. The bottom echo signal in the receiving signal is removed to obtain the detection signal used to characterize the columnar defects inside the object to be tested. Normalization processing: The amplitude of the detection signal is normalized to obtain an effective detection signal; Acoustic time correction: The excitation signal is enveloped to obtain the time corresponding to the maximum amplitude of the upper envelope of the excitation signal, so as to correct the transmission time during the propagation of the excitation signal and obtain the effective transmission time. Image generation: Based on the effective detection signal and effective transmission time, the pixel value matrix of the detection area is obtained by using the full focusing algorithm, and a three-dimensional ultrasonic image of the columnar defect inside the object under test is generated.

2. The method according to claim 1, characterized in that, In the signal acquisition, the detection area is rectangular, and the horizontal distance between each excitation sensor and the center of the detection area is less than or equal to 60% of the width of the detection area.

3. The method according to claim 1 or 2, characterized in that, In the normalization process, the calculation formula for amplitude normalization of the detected signal is as follows: in, To excite the signal transmission time, The detection signal is formed after the excitation signal emitted by the j-th excitation sensor reaches the i-th receiving sensor. This is the maximum amplitude of the detected signal. This is the effective detection signal obtained after normalizing the detection signal.

4. The method according to claim 3, characterized in that, In the aforementioned acoustic time correction, the calculation formula for correcting the transmission time during the propagation of the excitation signal is as follows: in, To determine the coordinates of any point P within the detection area, Let be the coordinates of the i-th receiving sensor. Let j be the coordinates of the excitation sensor. To excite the speed of signal propagation in the object under test, Let PPD be the transmission time from the excitation signal emitted by the j-th excitation sensor to the i-th receiving sensor via point P, forming the detection signal. PPD is the time corresponding to the maximum amplitude of the upper envelope of the excitation signal. The effective transmission time is the time for the excitation signal emitted by the j-th excitation sensor to travel through point P to the ith receiving sensor to form the detection signal.

5. The method according to claim 4, characterized in that, The image generation specifically includes: The detection area is divided into multiple uniform grids, and each grid point is used as a pixel. The pixel value of each pixel is calculated based on the effective detection signal and effective transmission time of each pixel. The pixel values ​​of each pixel are normalized by performing maximum and minimum value normalization to obtain a pixel value matrix; Based on the pixel value matrix, a three-dimensional ultrasonic image of the columnar defect inside the object under test is generated.

6. The method according to claim 5, characterized in that, In the image generation process, there are a total of Pixels: in, They are respectively The maximum value in, They are respectively The minimum value in, The size of each grid.

7. The method according to claim 5, characterized in that, In the image generation process, the formula for calculating the pixel value of each pixel is as follows: in, Let Q be the coordinates of any pixel. This indicates that the excitation signal emitted by the j-th excitation sensor reaches the ith receiving sensor via pixel Q, forming an effective detection signal. The effective transmission time for the excitation signal emitted by the j-th excitation sensor to travel through pixel Q to the i-th receiving sensor to form the detection signal is given. Let Q be the pixel value of pixel point Q.

8. The method according to claim 7, characterized in that, In the image generation process, the formula for normalizing the pixel values ​​of each pixel by applying the maximum and minimum values ​​is as follows: 。 in, pixel value The maximum value in, pixel value The minimum value in, This is a matrix of pixel values.

9. A terminal device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being loaded and executed by the processor to implement the method of any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium stores a computer program, which is loaded and executed by a processor to implement the method described in any one of claims 1 to 8.

Images