Method for identifying lead sealing defect part of cable with high imaging resolution

Abstract

The invention provides a high-imaging-resolution method for identifying a lead sealing defect part of a cable, and relates to the field of nondestructive testing. Comprising the steps that a linear array probe array is utilized, a plurality of array elements emit ultrasonic waves to a to-be-detected defect part at the same time, and a received echo signal reflects the waveform shape and the waveform phase of the defect. A two-dimensional image is preprocessed through a synthetic aperture focusing technology, a combined image is reprocessed through a phase coherence technology, the quality and accuracy of the image are further improved, and a high-resolution three-dimensional image is constructed in combination with an imaging synthetic aperture focusing technology. The ultrasonic phased array technology, the SAFT technology, the phase coherence technology and the synthetic aperture focusing technology are combined, the imaging resolution is effectively improved, the method has remarkable advantages especially when defect parts in complex structures such as cable lead seals are recognized, the imaging quality is optimized, a two-dimensional image is converted into a high-resolution three-dimensional image, and finally an accurate defect recognition result is obtained.

Description

Technical Field

[0001] The invention relates to the field of nondestructive testing, and in particular to a method for identifying defective parts of cable seals with high imaging resolution. Background Art

[0002] Phased array ultrasonic technology has become a popular technology in many non-destructive testing fields. Typically, phased array ultrasonic technology uses a linear phased array to produce a two-dimensional image in the azimuth plane, and the identification of defects is often limited by their random direction or operator experience. Ways to increase the probability of detection include extending the angular range and direction of the scan, or using a complete three-dimensional representation of the defective part of the cable seal to make it easier to identify and evaluate the defect. However, the application of three-dimensional imaging in non-destructive testing is rare. The main challenge facing industrial three-dimensional imaging systems is to have many channels to handle the large number of elements of the matrix transducer and to be able to acquire, process and transmit large amounts of data in real time while scanning the entire volume. Typically, real-time visualization in the medical field requires 25 frames per second, but automated inspection of large industrial components may require higher frame rates.

[0003] In the prior art, firstly, in order to change the position, tilt angle or rotation of the linear array, a mechanical support device is usually integrated into the transducer housing, resulting in different geometric shapes of the acquired imaging of the defective part to be detected, and this built-in transducer and scanner combination can only scan a limited number of defective parts from a fixed position. Second, an external mechanism is used to track the position of the array to improve accuracy. Third, a method for free three-dimensional ultrasonic scanning is studied, in which the positioning system is based on electromagnetic or optical sensors. The sensors are attached to the probe and can realize the free movement of the transducer so that the image is collected together with its relative position and direction to reconstruct a larger volume. However, the existing alternatives require a calibration process after acquisition to determine the size and limits of the imaging of the defective part to be detected and the data scaling, resulting in low imaging resolution of cable seal defects and inaccurate identification of cable seal defective parts. Summary of the invention

[0004] The main purpose of the present invention is to provide a method for identifying cable seal defective parts with high imaging resolution, so as to solve the problem of how to improve the imaging resolution of cable seal defects and the inaccurate identification of cable seal defective parts.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for identifying defective parts of cable seals with high imaging resolution, comprising the following steps: Step S1: Collect data using a linear array probe array. The ultrasonic phased array has multiple array elements, each of which transmits ultrasonic waves to the defective part to be tested. The acquired signal becomes larger due to the beam signal width in the vertical direction, and the target echo signal of the ultrasonic wave is received to obtain a two-dimensional image of the defective part to be tested; The target echo signal includes at least: a waveform shape and a waveform phase of the target echo signal; Step S2: using synthetic aperture focusing technology to perform image preprocessing on the two-dimensional image, and combining the obtained images in the vertical direction to obtain a combined image; Step S3: for the side lobes and grating lobes in the combined image, the phase coherence technology is used for image reprocessing to obtain an imaging picture; Step S4: for the imaging picture, a high-resolution three-dimensional image is constructed using an image processor GPU to obtain a recognition result.

[0006] In the preferred embodiment, in step S1, the width of the beam signal in the vertical direction acquired by the linear array probe array becomes larger. After adding the mask, the mask opening is set to be greater than twice the wavelength. The beam spread angle is calculated as follows: ; In the formula, is the beam spread angle, is the probe aperture, is the wavelength of ultrasound; The beam width is calculated as: ; Where d is the focal depth, is the beam signal width.

[0007] In the preferred solution, step S2: in the vertical direction, synthetic aperture focusing technology is used to improve the resolution and signal-to-noise ratio of the image to obtain a combined image, and the specific steps are as follows: S201: Calculate the geometric relationship between the selected probe array element and the target, calculate the round trip time from each array element to the target point, and mark the coordinates of a pixel point as , the emission point is on the axial coordinate axis in the array, and the distance formula from the probe array element to the pixel point is: ; In the formula, is the coordinate of the launch point; The formula for calculating the round-trip time from the probe array element to the target pixel is: ; In the formula, is the round trip time, c1 is the speed of ultrasound propagating in the seal, is the distance from the probe array element to the pixel point; S202: Obtain a series of low-resolution B-scan images and save them; S203: Input the round trip time and reconstruct the output image line by line. The information of each line belongs to the sample set at the same height. Coherent enhancement is performed on each pixel in the original sample set. The intensity of each output pixel is the sum of M delayed samples. The formula is: ; In the formula, is the intensity of pixel P in the final output image, M is the number of samples involved in the superposition, is the i-th A-scan signal at index The value at, index By sampling frequency and round trip time Decide; S204: For different depths z, the maximum number of samples required can be dynamically adjusted , the corresponding formula is: ; In the formula, dy is the mechanical scanning step length in the vertical direction, z For depth, is the beam divergence angle in the vertical plane.

[0008] In a preferred embodiment, the acquired target echo signal is subjected to offline post-processing, and the post-processing further includes: time delay compensation and signal alignment, image enhancement and post-processing, and processing by an image processor.

[0009] In the preferred embodiment, the combined image is processed using phase coherence technology, and the specific steps include: Step S301: Processing the combined image using phase coherence technology, specifically including: Step S301: extracting phase information from the signal data of the combined image; The phase coherence technology PCI is used to discretize the phase of each pixel P in the input image using SCF. The formula is: ; In the formula, The signal collected for the i-th time is at position y i The signal received at a i Symbols; By sampling frequency and round trip time Decision, SCF P is the phase consistency factor; Step S302: Perform weighted processing to obtain the final output pixel value ,in is the original pixel intensity calculated in S203: ; The output signal is: a high-resolution two-dimensional image processed by PCI, that is, an imaging picture.

[0010] In the preferred solution, in step S4, the imaging picture is used as an input signal, and a three-dimensional image is generated using an image processor GPU. The specific steps are as follows: Step S401: perform volume reconstruction, the formula is: ; In the formula, is a 3D voxel grid, f is the interpolation function, is the relative coordinate of the target point relative to the lower left vertex of the smallest cube; Step S402: Obtain , perform signal superposition and synthetic aperture formation, and select samples from the corresponding M high-resolution two-dimensional images And they are superimposed according to their weights, the formula is: ; In the formula, To select samples The i-th sample, w i is the weight, determined by the distance; S403: Apply phase consistency factor weighting to perform secondary weighting on the voxel value, the formula is: ; Where SCF is the phase consistency factor; Step S403: Create a three-dimensional voxel matrix based on the fused two-dimensional image data, and render to display the final high-resolution three-dimensional image.

[0011] In the preferred embodiment, before volume reconstruction, it is necessary to ensure that all two-dimensional images are spatially aligned, including: time delay compensation and signal alignment, geometric correction, motion compensation and alignment, image resampling and interpolation, phase and intensity alignment, and three-dimensional registration and spatial alignment.

[0012] The present invention provides a method for identifying defective parts of cable seals with high imaging resolution. A linear array probe array is used, and multiple array elements simultaneously transmit ultrasonic waves to the defective parts to be tested, and the received echo signals will reflect the waveform shape and waveform phase of the defects. The two-dimensional image is preprocessed by synthetic aperture focusing technology. Synthetic aperture focusing technology can improve the spatial resolution of the image by simulating the synthesis of echo signals at multiple different angles and positions. The side lobes and grating lobes that may appear in the combined image will affect the clarity and accuracy of the image. The use of phase coherence technology for reprocessing can effectively reduce these interferences, further improve the quality and accuracy of the image, and then combine the imaging synthetic aperture focusing technology to construct a high-resolution three-dimensional image. By combining ultrasonic phased array technology, SAFT, phase coherence technology and synthetic aperture focusing technology, the imaging resolution is effectively improved, especially when identifying defective parts in complex structures such as cable seals. Each step is closely related, and the imaging quality is gradually optimized, from two-dimensional images to high-resolution three-dimensional images, and accurate defect recognition results are finally obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is a schematic flow diagram of the method of the present invention; Figure 2 is a mask schematic diagram of the beam width in the vertical direction of the present invention; Figure 3 It is a schematic diagram of calculating the intensity of a pixel P in a single medium using the synthetic aperture focusing technology of the present invention; Figure 4 is a schematic diagram of calculating the intensity of a pixel P in two media by using the synthetic aperture focusing technology of the present invention; Figure 5 This is a comparison chart of the clutter filtering effects of the present invention and the traditional algorithm. DETAILED DESCRIPTION

[0014] Example 1 The technical solution of the present application is to install a mask on the probe to broaden the beam in the y direction on the basis of using a conventional linear array probe array to collect data, such as Figure 1 As shown, the linear array probe in the traditional ultrasonic phased array technology can be used without making additional changes to the hardware equipment.

[0015] like Figure 1-5 As shown, a method for identifying defective parts of cable seals with high imaging resolution includes the following steps: Step S1: Collect data using a linear array probe array. The ultrasonic phased array has multiple array elements, each of which transmits ultrasonic waves to the defective part to be tested. The acquired signal becomes wider due to the beam signal width in the vertical direction, and the target echo signal of the ultrasonic wave is received to obtain a two-dimensional image of the defective part to be tested.

[0016] The target echo signal includes at least: a waveform shape and a waveform phase of the target echo signal.

[0017] Step S2: Using synthetic aperture focusing technology to perform image preprocessing on the two-dimensional image, and combining the obtained images in the vertical direction to obtain a combined image.

[0018] Step S3: for the side lobes and grating lobes in the combined image, phase coherence technology is used for image reprocessing to obtain an imaging picture.

[0019] Step S4: for the imaging picture, a high-resolution three-dimensional image is constructed using an image processor GPU to obtain a recognition result.

[0020] This embodiment uses a linear array probe array, and multiple array elements simultaneously transmit ultrasonic waves to the defective part to be tested, and the received echo signal will reflect the waveform shape and waveform phase of the defect. The two-dimensional image is preprocessed by synthetic aperture focusing technology. Synthetic aperture focusing technology can improve the spatial resolution of the image by simulating the synthesis of echo signals at multiple different angles and positions. The side lobes and grating lobes that may appear in the combined image will affect the clarity and accuracy of the image. The use of phase coherence technology for reprocessing can effectively reduce these interferences, further improve the quality and accuracy of the image, and then combine the imaging synthetic aperture focusing technology to construct a high-resolution three-dimensional image. This embodiment effectively improves the resolution of imaging by combining ultrasonic phased array technology, SAFT, phase coherence technology and synthetic aperture focusing technology, especially when identifying defective parts in complex structures such as cable seals. Each step is closely related, and the imaging quality is gradually optimized, from two-dimensional images to high-resolution three-dimensional images, and accurate defect recognition results are finally obtained.

[0021] In the preferred embodiment, in step S1, the width of the beam signal in the vertical direction acquired by the linear array probe array becomes larger. After adding the mask, the mask opening is set to be greater than twice the wavelength. The beam spread angle is calculated as follows: ; In the formula, is the beam spread angle, is the probe aperture, is the wavelength of ultrasound.

[0022] The beam width is calculated as: ; Where d is the focal depth, is the beam signal width.

[0023] In the preferred solution, step S2: in the vertical direction, synthetic aperture focusing technology is used to improve the resolution and signal-to-noise ratio of the image to obtain a combined image, and the specific steps are as follows: S201: Calculate the geometric relationship between the selected probe array element and the target, calculate the round trip time from each array element to the target point, and mark the coordinates of a pixel point as , the emission point is on the axial coordinate axis in the array, and the distance formula from the probe array element to the pixel point is: ; In the formula, The coordinates of the launch point.

[0024] The formula for calculating the round-trip time from the probe array element to the target pixel is: ; In the formula, is the round trip time, c1 is the speed of ultrasound propagating in the seal, is the distance from the probe array element to the pixel.

[0025] S202: Obtain a series of low-resolution B-scan images and save them.

[0026] S203: Input the round trip time and reconstruct the output image line by line. The information of each line belongs to the sample set at the same height. Coherent enhancement is performed on each pixel in the original sample set. The intensity of each output pixel is the sum of M delayed samples. The formula is: ; In the formula, is the intensity of pixel P in the final output image, M is the number of samples involved in the superposition, is the i-th A-scan signal at index The value at, index By sampling frequency and round trip time Decide.

[0027] S204: For different depths z, the maximum number of samples required can be dynamically adjusted, and the corresponding formula is: ; Where dy is the mechanical scanning step in the vertical direction, z is the depth, is the beam divergence angle in the vertical plane.

[0028] In a preferred embodiment, the acquired target echo signal is subjected to offline post-processing, and the post-processing further includes: time delay compensation and signal alignment, image enhancement and post-processing, and processing by an image processor.

[0029] Two-dimensional imaging of all the above cable seal defective parts is obtained, but due to the presence of the mask on the ultrasonic phased array probe, the signal obtained by the probe becomes larger due to the beam signal width in the y direction, resulting in a decrease in the lateral resolution of the obtained image. Therefore, it is necessary to process and combine these low-resolution images in a consistent manner using synthetic aperture focusing technology after software processing. The synthetic aperture focusing technology can improve the imaging resolution and signal-to-noise ratio of the cable seal defective parts, and the single-station method is used in the synthetic aperture focusing technology, that is, the obtained images are combined in the vertical direction, thereby ensuring that the image information can be well combined in the vertical direction, thereby ensuring the clarity and coherence of the imaging slices of the cable seal defective parts in the vertical direction, thereby significantly improving the image resolution and signal-to-noise ratio.

[0030] Specifically, all the two-dimensional imaging slices of the low-resolution cable seal defective parts in the above array elements are acquired at one time using a frequency radio device (FRD), and the target echo signals acquired by the FRD are post-processed offline, and the post-processing may at least include: synthetic aperture focusing technology, phase coherence technology, and image processor processing. FRD signal acquisition has no restrictions on the configuration of the ultrasonic phased array, so the volume data of the cable seal in the target echo signal can also be acquired, and these post-processing algorithms can be applied to three-dimensional imaging.

[0031] In the preferred embodiment, the combined image is processed using phase coherence technology, and the specific steps include: Step S301: extracting phase information from the signal data of the combined image; The phase coherence technology PCI is used to discretize the phase of each pixel P in the input image using SCF. The formula is: ; In the formula, is the symbol of the signal ai received at position yi for the i-th acquisition signal; By sampling frequency and round trip time Determine, SCFP is the phase consistency factor; Step S302: Perform weighted processing to obtain the final output pixel value , where I P is the original pixel intensity calculated in S203, and the formula is: ; The output signal is: a high-resolution two-dimensional image processed by PCI, that is, an imaging picture.

[0032] In this embodiment, the imaging of the defective part of the cable lead seal obtained in the above steps after being processed by the synthetic aperture focusing technology is obtained, but since the use of the single-station method still causes artifacts in the image, the contrast of the image is reduced. Therefore, it is necessary to use the phase coherence technology to process the image obtained in the above steps again. The phase coherence technology can reduce the side lobes and grating lobes in the image, making up for the defects of using the synthetic aperture focusing technology, and overcoming the limitations of using the synthetic aperture focusing technology, and finally enhancing the resolution of the imaging of the defective part of the cable lead seal.

[0033] In the preferred solution, in step S4, the imaging picture is used as an input signal, and a three-dimensional image is generated using an image processor GPU. The specific steps are as follows: Step S401: perform volume reconstruction, the formula is: ; In the formula, is a 3D voxel grid, f is the interpolation function, It is the relative coordinate of the target point relative to the lower left vertex of the smallest cube.

[0034] Step S402: Obtain , perform signal superposition and synthetic aperture formation, and select samples from the corresponding M high-resolution two-dimensional images And they are superimposed according to their weights, the formula is: ; In the formula, the weight w i It is determined by the distance.

[0035] S403: Apply phase consistency factor weighting to perform secondary weighting on the voxel value, the formula is: ; Where SCF is the phase consistency factor.

[0036] Step S403: Create a three-dimensional voxel matrix based on the fused two-dimensional image data, and render to display the final high-resolution three-dimensional image.

[0037] The above steps are obtained to obtain the imaging of the defective part of the cable seal after being processed by synthetic aperture focusing technology and phase coherence technology. Since the two-dimensional image generated by the phased array device has a low resolution and is buffered in real time, the output image can be processed concurrently on an external mobile terminal device to construct a high-resolution three-dimensional image. Since the phased array technology is used in the azimuth plane, the amount of data remains low, so the imaging synthetic aperture focusing technology reconstruction can be performed at high speed on the image processor platform.

[0038] The specific principles of this embodiment are: In the above optional implementation manner, in order to make the collected image slices reflect the details of the defective parts of the cable seal as accurately as possible, it is necessary to ensure that the distance between the two collected slices is not too large.

[0039] 1. The combined process of synthetic aperture focusing technology can improve the signal-to-noise ratio of the output image. In the vertical direction, the effect of the combined process of synthetic aperture focusing technology is equivalent to using a M By combining multiple samples, an effect similar to using a larger array is achieved, which improves the quality of the image and greatly improves the lateral resolution. , that is, the vertical clarity of the image, is defined according to the Rayleigh criterion, and the width is measured at a specific depth z, and the formula is: ; In the formula, is the lateral resolution, dy Refers to the mechanical scanning step length in the vertical direction, z It refers to the depth, i.e. a position inside the material.

[0040] Therefore, according to the above formula, M The larger the value, the clearer the image will be in the vertical direction.

[0041] However, the practical limit on lateral resolution is the beam divergence in the vertical plane. and wavelength λ Therefore, in order to calculate the axial distance z Pixels at P , the number of consecutive images required is the scanning step length in the vertical direction dy Specifically, it is how many consecutive images are needed to ensure that the pixels P Covered by all beams.

[0042] ; The symbol “││↓” means rounding down, that is, rounding a decimal down to the nearest integer. The maximum number of images that need to be combined depends on the axial distance. It is a dynamic aperture process to ensure the best resolution at all depths.

[0043] The amplitude (brightness or intensity) of each output pixel is obtained by summing the M input samples. In simple terms, the brightness of each pixel is calculated by adding up multiple samples. In a homogeneous medium, the speed of sound in the medium is From a position in the array To the coordinates ( , ) is fixed. In the vertical and axial directions, there is a fixed scanning step length dy. Suppose we want to calculate the pixel at ( , ) The starting position of the array is The end position is ,in: ; ; For example, Figure 3 As shown, the numbers in the above formula i ={1,…, M} represents the order of obtaining data, M is the total number of samples, i From 1 to M The index of each sample of Indicates the vertical position of the slave array Distance to pixel point P Next, you need to find the index idx of the A-scan sample corresponding to pixel point P. The index is calculated based on the round-trip time, and the formula is: ; in, is the sampling frequency, From the location To Pixel P time, then The result is rounded off. P The relevant sample values ​​are added together to get the final intensity of the pixel. The intensity of each output pixel It is obtained by calculating M delayed samples, and the formula is: ; In the formula, Ai(n) is the position The true representation of the A-Scan signal received at is obtained by applying the Hilbert transform H, as follows: ; Among them, the Hilbert transform H Used to extract the envelope of a signal.

[0044] Second, when it comes to the interface between two materials, according to Fermat's principle, sound will propagate along the shortest travel time path. In order to obtain acceptable resolution in round-trip time, the interface should be discretized in fine grain, and then find the entry point that minimizes the sound path to the pixel. The formula is as follows: ; For example, Figure 4 As shown, is the path from position yi to the entry point ei in the first material, with a sound speed of c1, is the distance from ei to the pixel P in the second material, the speed of sound is c2, and the final output brightness is obtained.

[0045] In the above optional implementation manner, the processed image obtained by the synthetic aperture focusing technology needs to be processed by the phase coherence technology. The detailed process of processing the image by the phase coherence technology is as follows: Phase coherence technology calculates a coherence factor for each point in the output image, and the range of this coherence factor is from 0 to 1. The closer the coherence factor is to 1, the more reliable the signal is; the closer the coherence factor is to 0, the less reliable the signal is.

[0046] Calculation of coherence factor: The coherence factor takes into account the phase distribution of the aperture data and measures its dispersion (such as standard deviation, variance, etc.). In order to simplify the calculation, the Sign Coherence Factor (SCF) is proposed, which uses the sign of the signal as a single-bit quantization of the phase.

[0047] The calculation formula of symbol coherence factor SCF is: ; in, is in position The sign of the signal received at the corresponding index of the round trip time indication .

[0048] The intensity of each pixel P Calculated according to the formula and weighted by its SCF, the final output pixel is obtained: ; SCF helps distinguish signals by simplifying calculations and improves image resolution and contrast. SCF helps maintain image quality even at larger scan step sizes.

[0049] In the preferred embodiment, before volume reconstruction, it is necessary to ensure that all two-dimensional images are spatially aligned, including: time delay compensation and signal alignment, geometric correction, motion compensation and alignment, image resampling and interpolation, phase and intensity alignment, and three-dimensional registration and spatial alignment.

[0050] Through the above steps, it is ensured that in the 3D reconstruction process, the 2D image can be accurately fused into a high-resolution, artifact-free 3D image, thereby improving the recognition accuracy and imaging resolution of cable seal defects. Figure 5 As shown in the figure, compared with adding the mask, the width of the input signal increases, and the clutter filtering effect after SAFT processing is better than that after only adding the mask. After PCI processing, the clutter signal is further filtered out.

[0051] This embodiment uses a conventional linear array probe to collect echo signal data generated by an ultrasonic phased array, and installs a mask on the probe to widen the beam perpendicular to the array direction, thereby solving the problem that the two-dimensional image obtained by the conventional ultrasonic phased array usually has a low resolution in the direction perpendicular to the array element. Therefore, it is achieved without introducing expensive hardware equipment. When the width of the beam becomes larger, the divergence of the beam is also greater, and the image still has a good resolution: the echo image signal collected by the phased array is input into the mobile terminal device, and the above-mentioned image with reduced lateral resolution is processed by synthetic aperture focusing technology. The synthetic aperture focusing technology combines these low-resolution images. The synthetic aperture focusing technology significantly improves the image resolution and signal-to-noise ratio from three aspects: image acquisition, reconstruction and output; then the phase coherence technology is used to reduce the side lobes and grating lobes in the final image, thereby enhancing the image resolution; finally, the computing power of the image processor is used for processing, and the synthesis of high-resolution imaging of the defective part of the cable lead seal is realized at high speed. The problem of low imaging resolution of the cable lead seal part in the prior art is solved.

[0052] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for identifying defective parts of cable seals with high imaging resolution, characterized in that: Step S1: Collect data using a linear array probe array. The ultrasonic phased array has multiple array elements, each of which transmits ultrasonic waves to the defective part to be tested. The acquired signal becomes larger due to the beam signal width in the vertical direction, and the target echo signal of the ultrasonic wave is received to obtain a two-dimensional image of the defective part to be tested; The target echo signal includes at least: a waveform shape and a waveform phase of the target echo signal; Step S2: using synthetic aperture focusing technology to perform image preprocessing on the two-dimensional image, and combining the obtained images in the vertical direction to obtain a combined image; Step S3: for the side lobes and grating lobes in the combined image, the phase coherence technology is used for image reprocessing to obtain an imaging picture; Step S4: for the imaging picture, a high-resolution three-dimensional image is constructed using an image processor GPU to obtain a recognition result.

2. The method for identifying defective parts of cable seals with high imaging resolution according to claim 1 is characterized in that: In step S1, the width of the beam signal in the vertical direction obtained by the linear array probe array becomes larger. After adding the mask, the mask opening is set to be greater than twice the wavelength. The formula for calculating the beam spread angle is: ； In the formula, is the beam spread angle, is the probe aperture, is the wavelength of ultrasound; The beam width is calculated as: ； Where d is the focal depth, is the beam signal width.

3. The method for identifying defective parts of cable seals with high imaging resolution according to claim 1, characterized in that: Step S2: In the vertical direction, synthetic aperture focusing technology is used to improve the resolution and signal-to-noise ratio of the image to obtain a combined image. The specific steps are as follows: S201: Calculate the geometric relationship between the selected probe array element and the target, calculate the round trip time from each array element to the target point, and mark the coordinates of a pixel point as , the emission point is on the axial coordinate axis in the array, and the distance formula from the probe array element to the pixel point is: ； In the formula, is the coordinate of the launch point; The formula for calculating the round-trip time from the probe array element to the target pixel is: ； In the formula, is the round trip time, c1 is the speed of ultrasound propagating in the seal, is the distance from the probe array element to the pixel point; S202: Obtain a series of low-resolution B-scan images and save them; S203: Input the round trip time and reconstruct the output image line by line. The information of each line belongs to the sample set at the same height. Coherent enhancement is performed on each pixel in the original sample set. The intensity of each output pixel is the sum of M delayed samples. The formula is: ； In the formula, is the intensity of pixel P in the final output image, M is the number of samples involved in the superposition, is the i-th A-scan signal at index The value at, index By sampling frequency and round trip time Decide; S204: For different depths z, the maximum number of samples required can be dynamically adjusted , the corresponding formula is: ; In the formula, dy is the mechanical scanning step length in the vertical direction, z For depth, is the beam divergence angle in the vertical plane.

4. The method for identifying defective parts of cable seals with high imaging resolution according to claim 1, characterized in that: The acquired target echo signal is post-processed offline, and the post-processing also includes: time delay compensation and signal alignment, image enhancement and post-processing, and image processor processing.

5. The method for identifying defective parts of cable seals with high imaging resolution according to claim 1, characterized in that: The phase coherence technique is used to process the combined image. The specific steps include: Step S301: extracting phase information from the signal data of the combined image; The phase coherence technology PCI is used to discretize the phase of each pixel P in the input image using SCF. The formula is: ； In the formula, The signal collected for the i-th time at position y i The signal received at a i Symbols; By sampling frequency and round trip time Decision, SCF P is the phase consistency factor; Step S302: Perform weighted processing to obtain the final output pixel value ,in is the original pixel intensity calculated in S203: ； The output signal is: a high-resolution two-dimensional image processed by PCI, that is, an imaging picture.

6. The method for identifying defective parts of cable seals with high imaging resolution according to claim 1, characterized in that: Step S4, the imaging picture is used as an input signal, and a three-dimensional image is generated using an image processor GPU. The specific steps are as follows: Step S401: perform volume reconstruction, the formula is: ； In the formula, is a 3D voxel grid, f is the interpolation function, is the relative coordinate of the target point relative to the lower left vertex of the smallest cube; Step S402: Obtain , perform signal superposition and synthetic aperture formation, and select samples from the corresponding M high-resolution two-dimensional images And they are superimposed according to their weights, the formula is: ； In the formula, To select samples The i-th sample, w i is the weight, determined by the distance; S403: Apply phase consistency factor weighting to perform secondary weighting on the voxel value, the formula is: ； Where SCF is the phase consistency factor; Step S403: Create a three-dimensional voxel matrix based on the fused two-dimensional image data, and render to display the final high-resolution three-dimensional image.

7. The method for identifying defective parts of cable seals with high imaging resolution according to claim 6 is characterized in that: Before volume reconstruction, it is necessary to ensure that all two-dimensional images are spatially aligned, including: time delay compensation and signal alignment, geometric correction, motion compensation and alignment, image resampling and interpolation, phase and intensity alignment, and three-dimensional registration and spatial alignment.

Images