Acoustic coding radiation and deconvolution

Abstract

Encoded radiation (e.g., when pulse sequences are transmitted in accordance with a specified code) can be used for each acoustic transmission event for non-destructive testing. Improved A-scans can be constructed by convolving a specific finite impulse response filter profile with the time-domain representation of the received acoustic echo data. Such techniques can be used to improve dynamic range or signal-to-noise ratio, or to allow the use of lower analog gains to avoid saturation of the received signal processing chain from features such as front wall echoes.

Description

[Technical Field]

[0001] Priority Claim This patent application claims the benefit of priority of U.S. Provisional Patent Application No. 63 / 520,509, “Encoded Radiation with FIR Deconvolution,” filed on 18 August 2023 (Agent Reference No. 6409.268PRV), which is incorporated herein by reference in its entirety.

[0002] This document generally, but not limited to, concerns non-destructive evaluation, and more specifically, apparatus and techniques for providing coded radiation, as well as corresponding detection techniques for acoustic imaging and the like. [Background technology]

[0003] Non-destructive testing (NDT) can refer to the use of one or more different techniques to inspect areas on or within an object to determine whether a defect or malfunction exists, or to characterize the object being inspected. An example of a non-destructive testing technique is the use of eddy current testing, in which electromagnetic energy is applied to an object and the resulting induced current on or within the object is detected. The detected current (or associated impedance) value provides an indicator of the structure of the object under test, such as the presence of cracks, voids, porosity, or other inhomogeneities.

[0004] Another technique for NDT (Non-Defining Test) is the use of acoustic testing techniques, such as using one or more electroacoustic transducers to irradiate a region on or within the object under test with sound waves, thereby detecting and processing scattered or reflected acoustic energy. Such scattered or reflected energy can be referred to as an acoustic echo signal. Generally, such acoustic testing methods involve the use of acoustic frequencies within the ultrasonic frequency range, including, as an exemplary example, pulses having a specified range of energy, which can include values ​​from several hundred kilohertz to several tens of megahertz.

[0005] Acoustic testing, such as ultrasound-based testing, may include the use of individual transducers or arrays of such transducers, including providing focusing or beamforming techniques to assist in constructing data plots or images representing regions of interest on or within a test specimen. The use of arrays of ultrasonic transducer elements may include the use of phased array beamforming techniques and may be called phased array ultrasound testing (PAUT). For example, delayed-sum beamforming techniques may be used, including coherently summing the time-domain representations of acoustic signals received from each transducer element or aperture. Total focus (TFM) beamforming techniques may use data acquired when one or more elements in an array (or apertures defined by such elements) are used to transmit acoustic pulses and other elements are used to receive scattered or reflected acoustic energy, and a matrix is ​​constructed from time-series (e.g., A-scan) representations corresponding to the sequence of transmit-receive cycles in which the transmission originates from different elements (or corresponding apertures) in the array.

[0006] A matrix capture acquisition method in which A-scan data is obtained for each element (or each defined aperture) in the array can be called a “Full Matrix Capture” (FMC) technique. In a similar manner to TFM imaging, a phase-based technique can be used for one or more of the acquisition, storage, or subsequent analysis. Such a phase-based technique may include a coherent sum of normalized or quantized representations of the A-scan data corresponding to the phase information. Such a technique can be called a “Phase Coherence Imaging” (PCI) beamforming technique. Other imaging techniques include plane-wave imaging (PWI), and the techniques described herein are generally applicable to a variety of PAUT acquisition and beamforming techniques. [Overview of the project] [Problems that the invention aims to solve]

[0007] The inventors have recognized, among other things, that one challenge presented by acoustic imaging techniques is that the range of the detectable amplitude or "dynamic range" of the received acoustic echo signal can make it difficult to suppress strong front wall echo signals (or other strongly reflective features) while still detecting weakly reflected echo signals associated with defects, malfunctions, or other features of the subject. One technique is to adjust (e.g., increase) the transmit signal strength or receiver gain to detect weakly reflected features, but such an increase may produce front wall echo signals that saturate the received signal processing chain (either in the analog or digital domain). Such saturation can mask features near the front wall echo or other strong reflectors or scatterers. Reducing the transmit power or receive gain may help suppress such saturation, but as a result, weakly reflected echo signals below the noise floor of the received signal processing chain may be lost.

[0008] The inventors have recognized that coded emission schemes (e.g., a sequence of pulses corresponding to a specified code) can be used for each transmission event to address one or more of the aforementioned problems. An improved A-scan can be assembled by convolving a specific finite impulse response filter profile with the time-domain representation of the received acoustic echo data (or by a correlation-based detection technique that uses the original code sequence for evaluation). Using coded emission can improve the dynamic range or signal-to-noise ratio (SNR), or allow the use of lower analog gains to avoid saturation of the received signal processing chain due to features such as front wall echoes, while resolving weakly reflective features of the target. [Means for solving the problem]

[0009] In one example, a machine execution method may be used to perform an acoustic inspection or other acoustic evaluation, the method comprising generating each acoustic pulse transmission using a specified pulse sequence corresponding to a specified code, and acquiring acoustic echo data indicating scattered or reflected acoustic energy in response to each acoustic pulse transmission. The method may include applying correlation-based detection techniques or deconvolutional filters to provide filtered acoustic echo data. For example, the method may include filtering the acquired acoustic echo data using a finite impulse response (FIR) filter to provide filtered acoustic echo data. Imaging or beamforming may be performed by coherently summing the representations of the filtered acoustic echo data to provide pixel or voxel element values ​​in an image, and a series of such sums may be used to form an image for storage or presentation.

[0010] In one example, a machine execution method may include establishing or evaluating candidate codes for use in a coded emission scheme for acoustic inspection, the machine execution method including establishing a code for use in acoustic pulse transmission, wherein the code is defined by a specified pulse sequence exhibiting zero autocorrelation sidelobes at odd-value sample offsets, and establishing a finite impulse response (FIR) filter for use in filtering acquired acoustic echo data received in response to the acoustic pulse transmission, wherein the FIR filter results in the deconvolution of the specified pulse sequence. For example, according to a potential "pseudo-Barker" candidate code, the latter half of the specified pulse sequence after its midpoint contains a mirror image of the first half of the specified pulse sequence, but with alternating signs for each consecutive digit.

[0011] In one example, a system for performing acoustic testing or other acoustic evaluation comprises a pulse generation circuit, a receiving circuit, at least one processor circuit, and a memory circuit which, when executed by the at least one processor circuit, causes the system to perform one or more machine execution methods described herein, such as the machine execution methods described above.

[0012] This summary is intended to provide an overview of the subject matter of this patent application. This summary is not intended to provide an exclusive or exhaustive description of the invention. A more detailed description is included to provide further information relating to this patent application.

[0013] In drawings that are not necessarily drawn to actual size, the same reference numeral may represent the same component in different drawings. The same reference numeral with different prefixes may represent different aspects of the same component. Drawings generally illustrate various embodiments described herein, not as limitations. [Brief explanation of the drawing]

[0014] [Figure 1] An example of an acoustic inspection system that can be used to perform at least one or more of the techniques described herein is shown in general.

[0015] [Figure 2] An exemplary example is shown, including an A-scan acquisition where saturation occurs due to acoustic echoes from the front wall of the object under test.

[0016] [Figure 3A] An example of a sequence of values ​​corresponding to a 13-bit Barker code, which can be used as a pulse sequence for coded radiation, is shown.

[0017] [Figure 3B] Figure 3A shows an example of an autocorrelation function corresponding to the 13-bit Barker code.

[0018] [Figure 4A] An illustrative example including coefficients defining a finite impulse response (FIR) filter that can be used for deconvolution of acoustic echo signals induced in response to acoustic transmission generated using coded radiation is presented.

[0019] [Figure 4B] An illustrative example of a deconvolution response resulting from filtering an input time series using the FIR filter coefficients of Fig. 4A is presented, wherein the input time series includes the coded radiation pulse sequence of Fig. 3A.

[0020] [Figure 5] An overall relationship between noise averaged power and different code sequence lengths, together with analytically defined limits, is presented.

[0021] [Figure 6A] An illustrative example of a 39-bit pseudo-Barker code that can be used as a pulse sequence for coded radiation is presented.

[0022] [Figure 6B] An illustrative example of an autocorrelation function corresponding to the 39-bit code of Fig. 6A is presented.

[0023] [Figure 6C] An illustrative example of an FIR deconvolution filter that can be used for filtering acoustic echo signals induced in response to acoustic transmission generated using coded radiation corresponding to the 39-bit code of Fig. 6A is presented.

[0024] [Figure 6D] An illustrative example of a deconvolution response resulting from filtering an input time series using the FIR filter coefficients of Fig. 6C is presented, wherein the input time series includes the coded radiation pulse sequence of Fig. 6A.

[0025] [Figure 7] This document outlines a workflow that can be used to evaluate potential candidate codes (which define the pulse sequence for transmission), along with techniques for selecting a code.

[0026] [Figure 8] This paper provides an overview of technologies, including mechanical execution methods, that can be used for coded radiation and related detection in acoustic inspection systems.

[0027] [Figure 9A] Various exemplary images generated using TFM beamforming, demonstrating TFM beamforming without coded emission. [Figure 9B] Various exemplary examples of images generated using TFM beamforming, showing TFM beamforming where encoded emission is used and correlation-based methods are used for detection. [Figure 9C] Various exemplary examples of images generated using TFM beamforming, showing TFM beamforming where an FIR filter performs deconvolution.

[0028] [Figure 10A] Various exemplary examples of images generated using TFM beamforming, demonstrating TFM beamforming performed in the presence of noise added to the input A-scan data without the use of coded radiation. [Figure 10B] Various exemplary examples of images generated using TFM beamforming, demonstrating TFM beamforming performed in the presence of noise, including averaging of 13 images in which a noise signal added to the A-scan data below is random and varies between trials. [Figure 10C]Various exemplary examples of images generated using TFM beamforming, where noise is added but averaging is not performed, and where an FIR filter is used to perform deconvolution.

[0029] [Figure 11A] Various examples of B-scan images are shown, illustrating image acquisition where coded radiation is not used and a center frequency of 0.5 megahertz (MHz) is used. [Figure 11B] Various examples of B-scan images are shown, illustrating image acquisition where coded radiation is not used and a center frequency of 5.0 MHz is used. [Figure 11C] Various examples of B-scan images are shown, illustrating image acquisition using an FIR filter used for coded emission and deconvolution with a center frequency of 5.0 MHz.

[0030] [Figure 12] An example block diagram is provided illustrating a machine capable of performing any one or more of the techniques (e.g., methodologies) described herein. [Modes for carrying out the invention]

[0031] Non-destructive testing of structures can be performed using acoustic techniques, such as ultrasonic testing using a phased array transducer architecture and associated processing (e.g., beamforming and imaging). As mentioned above, the interpretation of acquired inspection data or associated images can present various challenges. Saturation of the receiver signal processing chain, such as due to strong front wall echoes associated with the object under test, can mask features of other weaker reflectors. One technique can perform multiple acquisition and imaging operations, such as using different transmit power or receive gain settings. Using multiple acquisitions in this way can negatively impact the productivity of the inspection, such as slowing down the inspection process or failing to detect weaker reflectors located near strongly reflectors. The techniques described in this document can be used to improve (e.g., increase) the dynamic range of amplitude measurements in acoustic inspection, or the signal-to-noise ratio (SNR), or one or more of both. Such improvements can suppress saturation of A-scan data (e.g., clipping in the digital domain or saturation of the analog front end) or suppress saturation of images resulting from the use of techniques such as plane-wave imaging (PWI) or total-focus beamforming (TFM).

[0032] Figure 1 provides an overall illustration of an example comprising an acoustic testing system 100 that can be used to perform at least one or more techniques as described herein. The testing system 100 may include test equipment 140, such as a handheld or portable assembly. The test equipment 140 may be electrically coupled to a probe assembly 150, for example, by using a multi-conductor interconnect 130. The probe assembly 150 may include one or more electroacoustic transducers, such as a transducer array 152 containing each of the transducers 154A to 154N. The transducer array may follow a linear or curved contour, or may include an array of elements extending along two axes, such as providing a matrix of transducer elements. The elements do not need to have a square footprint or be arranged along a linear axis. Element size and pitch can be varied according to the testing application.

[0033] The modular probe assembly 150 can be configured to allow the test apparatus 140 to be used with various different probe assemblies. Generally, the transducer array 152 includes piezoelectric transducers that can be acoustically coupled to a target 158 ​​(e.g., a test specimen or "object under test") via a bonding medium 156. The bonding medium can include a fluid or gel or solid film (e.g., an elastomer or other polymer material) or a combination of fluid, gel or solid structures. For example, an acoustic transducer assembly may include a transducer array coupled to a wedge structure containing a rigid thermosetting polymer having known acoustic propagation properties (e.g., Rexolite®, available from C-Lec Plastics Inc.), and water can be injected as the bonding medium 156 between the wedge and the structure under test, or the interface between the probe assembly 150 and the target 158 ​​can be immersed in the bonding medium for testing.

[0034] The test equipment 140 may include digital and analog circuits such as a front-end circuit 122 that includes one or more transmitter signal chains, receiver signal chains, or switching circuits (e.g., transmit / receive switching circuits). The transmitter signal chain may include amplifier and filter circuits to image or detect structural or in-structure defects 160 of the target 158 ​​by receiving scattered or reflected acoustic energy induced in response to the sound wave irradiation, for example, by providing transmit pulses to be delivered via the interconnect 130 to a probe assembly 150 for irradiating the target 158 ​​with sound waves.

[0035] Figure 1 shows a single probe assembly 150 and a single transducer array 152, but other configurations can be used, such as multiple probe assemblies connected to a single test instrument 140, or multiple transducer arrays 152 used with a single probe assembly 150 or multiple probe assemblies for pitch / catch inspection modes. Similarly, the test protocol can be performed using coordination between multiple test instruments 140, such as in response to an overall test scheme established from a master test instrument 140, or by another remote system such as a computing device 108, or by a general-purpose computing device such as a laptop 132, tablet, smartphone, or desktop computer. The test scheme may be established in accordance with published standards or regulatory requirements and may be performed repeatedly, for example, during initial manufacturing or for ongoing monitoring.

[0036] The receiver signal chain of the front-end circuit 122 may include one or more filter or amplification circuits, along with analog-to-digital conversion equipment, to digitize the echo signal received using the probe assembly 150. The digitization may be performed coherently to provide multiple channels of digitized data that are aligned or referenced to each other in time or phase. The front-end circuit may be coupled to and controlled by one or more processor circuits, such as processor circuit 102, which is included as part of the test equipment 140. The processor circuits may be coupled to memory circuit 104 to execute instructions causing the test equipment 140 to perform one or more of the following: acoustic transmission, acoustic acquisition, processing, or storage of data relating to acoustic inspection, or otherwise to perform techniques as described herein. The test equipment 140 may be communicably coupled to other parts of the system 100, for example, by using a wired or wireless communication interface 120.

[0037] For example, the execution of one or more techniques as described herein can be achieved by being mounted on the test equipment 140 or by using other processing or storage functions, such as using the computing equipment 108 or a general-purpose computing device such as a laptop 132, tablet, smartphone, or desktop computer. For example, processing tasks that become undesirably slow when performed mounted on the test equipment 140 or when exceeding the capabilities of the test equipment 140 can be performed remotely (e.g., on a separate system), such as in response to a request from the test equipment 140. Similarly, the storage of intermediate data, such as an A-scan matrix of imaging data or time-series data or other representations of such data, can be achieved, for example, using a remote device communicatively coupled to the test equipment 140. The test equipment may include a display 110 for presenting configuration information or results, and an input device 112 for receiving operator commands, configuration information, or responses to queries, including one or more of the following: a keyboard, trackball, function keys or soft keys, mouse interface, touchscreen, stylus, etc.

[0038] Various acoustic inspection techniques may impose limitations on the dynamic range or signal-to-noise ratio (SNR) for imaging the structure being inspected. For example, the use of all-focus (TFM) beamforming techniques based on full-matrix acquisition (FMC) may present challenges in terms of SNR limitations. For instance, TFM beamforming using FMC acquisition generally involves one transducer or one aperture transmitting and another transducer receiving the acoustic echo signal. Consequently, such transmissions may lack penetration power compared to plane-wave imaging (PWI) because they use fewer transducer elements (or generally a single transducer element) per transmit event. Such effects can also occur when using sparse acquisition techniques. Sparse acquisition may involve performing less matrix acquisition than full-matrix acquisition, and individual transmit events may again only involve a single element, as before, resulting in fewer transmit and receive acquisitions used for the total beamforming. In general, sparse acquisition may exhibit a lower signal-to-noise ratio compared to all-aperture acquisition. In corrosion and composite material inspection, operators may need to increase analog gain to observe multiple reflections or to locate backwall echoes in thick samples. However, increasing gain can saturate frontwall echoes, thereby complicating processing and associated imaging tasks. The techniques described herein may be used, for example, in the use cases or applications described above to improve signal-to-noise ratio (SNR) or to provide an improved dynamic range (or both).

[0039] As an example, Figure 2 shows an exemplary case including an A-scan acquisition 200 in which saturation occurs at 224 due to acoustic echo from the front wall of the object under test. The inventors have recognized that the various problems described above can be addressed by increasing the SNR of the acquired A-scan data without requiring an increase in analog gain, thereby also improving the dynamic range. Such an improvement in dynamic range can enable the detection of weak echo signals of the target while also suppressing saturation (or clipping in the digital domain) of the analog receiving channel.

[0040] The inventors have recognized that, in particular, one technique for increasing SNR or dynamic range (or both) may include the use of coded radiated pulse transmission and reception schemes. For example, the Omniscan X3-64 (available from Evident Scientific, Inc., Waltham, MA, USA and Evident Canada, Inc., Quebec, QC, Canada) has a transmit pulse generator capable of generating positive and negative pulses, such as supporting at least three output states (positive amplitude, 0, and negative amplitude). Such states can be mapped to output values ​​{+1, 0, -1} for positive, negative, and 0, and codes can be specified corresponding to pulse sequences of such states, thereby enabling the transmission of pulse sequences having envelope codes corresponding to binary codes (e.g., having two states including +1 or -1) instead of a single transmit pulse envelope. A code can be selected to exhibit high autocorrelation, thereby resulting in a code with a much lower (or even zero) correlation when evaluated with its own delayed representation, where the correlation corresponds to a peak in the autocorrelation function with perfect time alignment (e.g., no sample offset) and the correlation is offset by one or more samples in either direction (leading or lagging). Such values ​​in the autocorrelation function adjacent to the peak or main lobe can be called "side lobes".

[0041] Golay codes and Barker codes are binary codes that can be used to provide coded emission. Golay code sets exhibit zero side lobes and generally have twice the number of firing events (2×) compared to uncoded excitations, and therefore productivity (e.g., the reciprocal of the acquisition rate) is halved. Barker codes are another family of codes that exhibit autocorrelation function peaks, and the maximum side lobe to main lobe amplitude ratio is generally 1 / N, where N corresponds to the number of digits in the code (e.g., length in bits). The longest known Barker code has a length of 13. As a result, the corresponding minimum side lobe amplitude level is -22.3 dB.

[0042] Figure 3A shows an exemplary sequence of values ​​corresponding to the aforementioned 13-bit Barker code, which can be used as a pulse sequence for coded radiation, and Figure 3B shows an exemplary autocorrelation function corresponding to the 13-bit Barker code in Figure 3A, exhibiting a strong main lobe (autocorrelation peak 334).

[0043] In the frequency domain, the convolution process for coded radiation can be described as follows:

number

[0044] Here, Y(ω) represents the coded radiated response as a function of angular frequency, X(ω) represents the impulse response, and C(ω) represents the code. The inventors recognize, among other things, that the impulse response can be recovered using a deconvolution process, which can be expressed as follows:

number

[0045] Deconvolution using spectral techniques can amplify noise in the region where |C(ω)| approaches 0. However, for many binary codes, |C(ω)| is generally never 0. In the time domain, the deconvolution process can be represented as follows:

number

[0046] The time-domain impulse response function h(t) can be implemented (e.g., approximated) as a finite impulse response (FIR) filter. For example, Figure 4A shows an exemplary example including coefficients defining an FIR filter that can be used for deconvolution of an acoustic echo signal induced in response to an acoustic transmission generated using coded radiation. The FIR filter illustrated in Figure 4A corresponds to the 13-bit Barker code in Figure 3A. Figure 4B shows an exemplary example of the deconvolution response (showing a strong peak 444) resulting from filtering an input time series using the FIR filter coefficients in Figure 4A, in which case the input time series includes the coded radiation pulse sequence in Figure 3A.

[0047] Since the deconvolution process can be applied to any binary code having |C(ω)| that is never zero, a metric can be used to evaluate the suitability (e.g., optimality) of candidate codes and their corresponding FIR deconvolution filters. For practical application, the FIR deconvolution filter h(t) can be truncated to a limited number of taps (e.g., filter coefficients) to facilitate implementation, for example, by setting coefficients below a specified amplitude threshold to zero or dropping them otherwise. The count of coefficients (e.g., number of taps) that can provide a reasonably accurate representation of h(t) can be expressed by the following constraints:

number

[0048] The above constraints satisfy the amplitude criterion used when evaluating the set of candidate impulse response filter coefficients corresponding to the candidate codes. In the above formula, coefficients (e.g., taps) with amplitudes greater than a specified amplitude threshold are counted N. tap This can be established (here corresponding to 1 percent of the maximum amplitude of h(t)). Through empirical evaluation, as an example, the impulse response filter coefficient (N tap The count of the number of digits (N) in the specified code. code It can be seen that this can be constrained to a specified multiple or less of the coefficient of ). For example, N less than 5 tap / N code While a ratio (e.g., five FIR filter coefficients per digit of the code) appears to produce useful results, as illustrated below, other ratios can also be used based on empirical or numerical evaluation. As a numerical example, if the candidate code has a length of 20 digits, any candidate code with an associated deconvolution filter h(t) that can be represented exactly in fewer than 100 taps will satisfy the 5:1 criterion above.

[0049] Other constraints (e.g., metrics indicating optimality or relative merit) can be applied to help evaluate candidate codes and their associated deconvolutional filter coefficients. For example, coded radiation can improve the signal-to-noise ratio by suppressing non-coherent random noise. Thus, quantitative metrics can help in selecting the appropriate code by comparing the noise suppression power of different codes. Assuming that the random noise at each time sample is statistically independent, the effective noise-averaging power of a binary code and its FIR deconvolutional filter can be expressed as follows:

number

[0050] In equation 5 above, l represents the characteristic delay of the FIR filter h(t). Since the improvement in SNR is generally proportional to the square root of the noise-averaged power, it can be expressed as follows:

number

[0051] As an example, Figure 5 shows the relationship between noise-averaged power and different code sequence lengths, along with analytically defined limits corresponding to normalized autocorrelation. As discussed elsewhere in this specification, a code can be identified when the corresponding FIR deconvolutional filter has up to five coefficients per digit (per bit) of the code, and as shown in Figure 5, the “pseudo” Barker code established by the inventors can provide performance comparable to Barker codes up to 13 bits in length, for which the code is identified up to 49 bits in length. The noise-averaged power of non-ideal candidate codes and their associated deconvolutional filters can be achieved by normalized autocorrelation. code It falls below the analytically defined limit (e.g., the theoretical maximum) (e.g., detection using correlation instead of using an FIR filter for deconvolution). The use of autocorrelation always generates non-zero side lobes in the autocorrelation function.

[0052] Ideally, a coded radiated pulse sequence has an autocorrelation function sidelobe with zero amplitude. However, the inventors have recognized that a 13-bit Barker code exhibits a sidelobe that alternates between non-zero and zero values, as shown in Figure 3B, where sample 1 at 336 is non-zero and sample 2 at 338 is zero. The inventors have recognized that a new class of codes exists beyond 13 bits that have zero autocorrelation sidelobes for odd-numbered samples in the sequence. Codes that satisfy such criteria are referred to herein as “pseudo-Barker” codes. In the case of odd-length pseudo-Barker codes (e.g., having an odd number of digits), all bits after the midpoint digit can be deterministically determined, for example, by using the following analysis formula (where i represents the number of digits).

number

[0053] In Equation 7, the latter half of the specified pulse sequence after the midpoint of the specified pulse sequence (for example, in this case, i>n mid ) corresponds to the first half of the specified pulse sequence (in this case, i<n mid ) that comprises a mirror image with alternating signs for each successive digit. Since the pseudo-Barker code is defined according to the above formula, an exhaustive search for a 49-bit pseudo-Barker code computationally corresponds to an exhaustive search for all possible 25-bit binary codes. The present inventors have also recognized that, among other things, appropriate FIR deconvolution filter coefficients can be established that satisfy the constraints of Equation 4 above (e.g., having a suitably small number of taps including a specified multiple of the pseudo-Barker code length) while also providing acceptable noise averaging behavior according to Equation 5 above.

[0054] FIG. 6A shows an illustrative example of a 39-bit pseudo-Barker code that can be used as a pulse sequence for coded radiation. Maximum allowable N tap / N code When the ratio is 5, it can be seen that the optimal 39-bit code as shown in FIG. 6A is [1,1,1,1,1,-1,-1,1,-1,1,1,1,1,1,-1,-1,1,-1,-1,-1,1,-1,-1,-1,1,1,-1,1,-1,1,1,1,1,-1,-1,1,-1,1,-1]. As mentioned above, the portion of sequence 556 after midpoint digit 554 (corresponding to sample 20) is a mirror image of the first portion of the sequence, but with alternating signs between successive digits.

[0055] Figure 6B shows an exemplary example of the autocorrelation function corresponding to the 39-bit code in Figure 6A. The correlation peak (main lobe 534) is surrounded by side lobes where odd samples are zero. Figure 6C shows an exemplary example of an FIR deconvolutional filter that can be used to filter acoustic echo signals induced in response to acoustic transmissions generated using coded radiation corresponding to the 39-bit code in Figure 6A. There are 157 non-zero filter coefficients (corresponding to 157 "taps") shown in Figure 6C, providing 36 noise-averaged powers corresponding to a 15.6 dB SNR improvement.

[0056] Figure 6D shows an exemplary example of the deconvolution response resulting from filtering an input time series using the FIR filter coefficients of Figure 6C, in which case the input time series includes the coded radiant pulse sequence of Figure 6A, showing the expected peak at 544 when the FIR deconvolution filter matches the signal corresponding to the coded radiant pulse sequence of Figure 6A. In general, the sample count (e.g., length) of the convolution response can be expressed as N + M - 1, where N is the length of the input signal (in this example, the coded sequence) and M is the length of the filter. In the example case of Figure 6D (corresponding to the 39-bit code in Figure 6A), the input has a length of 39, and the FIR filter has a length of approximately 240 samples.

[0057] Figure 7 shows a workflow 700 that can generally be used to evaluate potential candidate codes (which define the pulse sequence for transmission), along with techniques for selecting codes at runtime or when establishing software or firmware for imaging. In 705, candidate or "trial" binary codes 715 can be generated during evaluation (randomly or deterministically, for example, by using equation 7 for evaluating pseudo-Barker code candidates), and in 720, the corresponding set of FIR filter coefficients can be determined. In 725, N less than equations 4 and 5 above are used. tap / N codeA first constraint can be applied, such as corresponding to a ratio. In 730, a second constraint can be applied to determine the improvement in SNR due to noise averaging. In 735, the results can be stored in a table or other data structure. In 740, the optimal code for one or more code sequence bit lengths can be determined, for example, by using the table formed in 735.

[0058] During operations for A-scan acquisition or imaging, such as runtime in 710, a code can be selected in 745 based on user-selectable criteria (e.g., selection of gain or dynamic range) or other constraints, such as being selected for a specific measurement configuration, frequency, or application. In 750, coded excitations can be generated, such as by generating acoustic pulses having an envelope defined by a specified code sequence. In 755, an acoustic echo signal induced in response to a transmitted event in 750 can be deconvolved using an FIR filter. Alternatively, such an acoustic echo signal can be detected using correlation-based detection techniques, although such techniques are more likely to introduce artifacts related to sidelobes in the autocorrelation function compared to deconvolution in the digital domain using an FIR filter. In 760, an A-scan output equivalent to a single-pulse transmission but with improved SNR can be generated. Such an A-scan output can be stored or presented to the user as an exemplary example, or used for other imaging applications such as PWI or TFM beamforming.

[0059] Figure 8 provides an overview of technique 800, including machine execution methods that can be used for coded emission and associated detection in an acoustic inspection system. 805 can generate acoustic pulse transmissions (e.g., corresponding to transmit / receive acquisition as part of FMC acquisition). Acoustic pulse transmissions can be defined by a pulse sequence corresponding to a specified code, such as a Barker code or pseudo-Barker code, as described elsewhere in this specification. Generally, the acoustic pulser generates pulses having a specified center frequency (e.g., within the ultrasonic frequency range), and for this purpose, the code defines the envelope of the pulses in the sequence. 810 can acquire acoustic echo data induced in response to the transmission in 805. 815 can filter the acoustic echo data using an FIR filter to result in deconvolution based on the specified code used for generating the acoustic pulse transmission in 805. Alternatively, correlation-based methods can be used for detection. 820 can generate an improved A-scan representation, such as having an improved SNR after deconvolution compared to an A-scan acquired in the absence of coded emission and deconvolution. In the 825, groups of enhanced A-scan representations can be coherently summed for purposes such as delayed-sum beamforming techniques to form pixel or voxel values ​​within an acoustic inspection image. As described below, a matrix of such enhanced A-scans can be formed using a series of coded radiated transmit events and corresponding received events for purposes such as TFM or other beamforming and associated imaging.

[0060] Figures 9A, 9B, and 9C show various exemplary examples of images generated using TFM beamforming, where Figure 9A shows TFM beamforming without coded emission, Figure 9B shows TFM beamforming where coded emission is used and correlation-based methods are used for detection, and Figure 9C shows TFM beamforming where an FIR filter is used to perform deconvolution. For Figures 9B and 9C, coded emission was simulated by convolving each A scan in the FMC with a 13-bit Barker code (as shown in Figure 3A). The FMC data was then reconstructed using both autocorrelation (Figure 9B) and FIR deconvolution (Figure 9C). The reconstructed FMC using FIR deconvolution (the resulting TFM image in Figure 9C) shows no change compared to the original (as represented by the TFM image in Figure 9A), but the autocorrelated reconstructed FMC (associated with the TFM image in Figure 9B) has several artifacts, as shown in region 346, due to non-zero side lobes.

[0061] Figures 10A, 10B, and 10C show various exemplary examples of images generated using TFM beamforming. In this case, Figure 10A shows TFM beamforming performed in the presence of noise added to the input A-scan data without coded radiation; Figure 10B shows TFM beamforming performed in the presence of noise, including averaging of 13 images where the noise signal added to the underlying FMC A-scan data is random and varies between trials; and Figure 10C shows TFM beamforming where noise is added but averaging is not performed, and an FIR filter is used to perform deconvolution. The simulation results in Figures 9A, 9B, 9C, 10A, 10B, and 10C suggest that FIR deconvolution does not introduce side lobes when the SNR is high, and that FIR deconvolution reduces the level of random noise when the SNR is low.

[0062] Figures 11A, 11B, and 11C show various examples of B-scan images, where Figure 11A shows image acquisition without coded radiation and using a center frequency of 0.5 megahertz (MHz), Figure 11B shows image acquisition without coded radiation and using a center frequency of 5.0 MHz, and Figure 11C shows image acquisition using coded radiation with a center frequency of 5.0 MHz and an FIR filter used for deconvolution. In general, the use of coded radiation as described in this document can increase probe penetration, and therefore, in the process of improving temporal resolution, higher frequency probes can be used for highly attenuating materials such as composite materials. Figure 11A shows the default inspection setup (using a 0.5 MHz probe), which provides good penetration but has relatively low spatial-temporal resolution. In contrast, in Figure 11B, a 5 MHz probe without coded radiation gives improved resolution compared to Figure 11A, but the front wall region is highly saturated (due to high gain) and deeper regions have a lower SNR. In contrast, using coded emission, as shown in Figure 11C, allows for the use of lower gains that reduce saturation at the front wall of region 1161. Furthermore, better SNR performance can be obtained in deeper regions. As a result, such a higher dynamic range allows for simultaneous examination of both shallow and deep regions of the sample using a single acquisition and imaging operation.

[0063] Figure 12 illustrates an example block diagram of a machine 1200 in which one or more of the techniques (e.g., methodologies) described herein can be performed. The machine 1200 (e.g., a computer system) may include a hardware processor 1202 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), a hardware processor core, or any combination thereof), main memory 1204, and static memory 1206 connected via an interlink 1230 (e.g., a link or bus), some or all of these components may constitute hardware for the aforementioned system or related implementation.

[0064] Generally, the hardware processor 1202 may include, for example, at least one of the following: a central processing unit (CPU), a reduced instruction set computing (RISC) processor, a composite instruction set computing (CISC) processor, a graphics processing unit (GPU), a digital signal processor (DSP), a tensor processing unit (TPU), a neural processing unit (NPU), a vision processing unit (VPU), a machine learning accelerator, an artificial intelligence accelerator, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a radio frequency integrated circuit (RFIC), a neuromorphic processor, a quantum processor, or any combination thereof. The processor circuit may further be a multicore processor having two or more independent processor "cores" capable of executing instructions simultaneously. A multicore processor includes multiple computing cores on a single integrated circuit die, each capable of independently executing program instructions in parallel. Parallel processing in a multicore processor may be implemented through architectures such as superscalar, VLIW, vector processing, or SIMD, which allow each core to execute separate instruction streams simultaneously. A processor circuit may be emulated as a virtual processor or virtual circuit by software running on a physical processor. A virtual processor can operate like an independent processor, but is implemented in software rather than hardware.

[0065] Specific examples of main memory 1204 include semiconductor memory devices that may include random access memory (RAM) and storage locations within semiconductors such as registers. Specific examples of static memory 1206 include semiconductor memory elements (e.g., electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM)), non-volatile memory such as flash memory devices, magnetic disks such as internal hard disks and removable disks, magneto-optical disks, RAM, or optical media such as CD-ROMs and DVD-ROMs.

[0066] The machine 1200 may further include a display device 1210, an input device 1212 (e.g., a keyboard), and a user interface (UI) navigation device 1214 (e.g., a mouse). In one example, the display device 1210, the input device 1212, and the UI navigation device 1214 may be touchscreen displays. The machine 1200 may also include a mass storage device 1208 (e.g., a drive unit), a signal generation device 1218 (e.g., a speaker), a network interface device 1220, and one or more sensors 1216 such as a Global Positioning System (GPS) sensor, a compass, an accelerometer, or any other sensor. The machine 1200 may also include an output controller 1228, such as a serial (e.g., Universal Serial Bus (USB), parallel, or other wired or wireless (e.g., infrared (IR), near field communication (NFC)) connection) for communicating with or controlling one or more peripheral devices (e.g., a printer, a card reader, etc.).

[0067] The mass storage device 1208 may include a machine-readable medium 1222 that stores one or more sets of data structures or instructions 1224 (e.g., software) that embody or utilize any one or more of the techniques or functions described herein. The instructions 1224 may also reside, all or at least partially, in the main memory 1204, static memory 1206, or hardware processor 1202 during their execution by the machine 1200. In one example, one or any combination of the hardware processor 1202, main memory 1204, static memory 1206, or mass storage device 1208 includes a machine-readable medium.

[0068] Specific examples of machine-readable media include one or more non-volatile memory such as semiconductor memory devices (e.g., EPROM or EEPROM) and flash memory devices, magnetic disks such as internal hard disks and removable disks, magneto-optical disks, RAM, or optical media such as CD-ROMs and DVD-ROMs. Although machine-readable media are shown as a single medium, the term “machine-readable media” may also include a single or multiple mediums configured to store one or more instructions 1224 (e.g., a centralized or distributed database or associated caches and servers).

[0069] The apparatus of machine 1200 may include one or more of the following: a hardware processor 1202 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), a hardware processor core, or any combination thereof), main memory 1204 and static memory 1206, a sensor 1216, a network interface device 1220, an antenna, a display device 1210, an input device 1212, a UI navigation device 1214, a mass storage device 1208, an instruction 1224, a signal generation device 1218, or an output controller 1228. The apparatus may be configured to perform one or more of the methods or operations disclosed herein.

[0070] The term “machine-readable medium” includes, for example, any medium that can store, encode, or carry instructions for execution by machine 1200, cause machine 1200 to execute one or more of the technologies of this disclosure, or cause another device or system to execute one or more of the technologies, or store, encode, or carry data structures used or associated with such instructions. Non-limiting examples of machine-readable mediums include solid-state memory, optical media, or magnetic media. Specific examples of machine-readable mediums include non-volatile memory such as semiconductor memory devices (e.g., electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM)) and flash memory devices, magnetic disks such as internal hard disks and removable disks, magneto-optical disks, random access memory (RAM), or optical media such as CD-ROMs and DVD-ROMs. In some examples, machine-readable mediums include non-temporary machine-readable mediums. In some examples, machine-readable mediums include machine-readable mediums that are not temporary propagated signals.

[0071] Instruction 1224 may be transmitted or received over a communication network 1226 using a transmission medium via a network interface device 1220 that utilizes one of several transport protocols (e.g., Frame Relay, Internet Protocol (IP), Transmission Control Protocol (TCP), User Datagram Protocol (UDP), Hypertext Transfer Protocol (HTTP), etc.). Examples of communication networks include, among others, local area networks (LANs), wide area networks (WANs), packet data networks (e.g., the Internet), mobile phone networks (e.g., cellular networks), conventional telephone (POTS) networks and wireless data networks (e.g., the IEEE 802.11 standard family, also known as Wi-Fi®), the IEEE 802.15.4 standard family, the Long-Term Evolution (LTE) 4G or 5G standard family, the Universal Mobile Telecommunications System (UMTS) standard family, peer-to-peer (P2P) networks, and satellite communication networks.

[0072] In one example, the network interface device 1220 includes one or more physical jacks (e.g., Ethernet, coaxial, or other interconnects) or one or more antennas for accessing the communication network 1226. In one example, the network interface device 1220 includes one or more antennas for wireless communication using at least one of the following techniques: single-input multiple-output (SIMO), multiple-input multiple-output (MIMO), or multiple-input single-output (MISO). In some examples, the network interface device 1220 wirelessly communicates using multi-user MIMO technique. The term “transmission medium” should be interpreted to include any intangible medium capable of storing, encoding, or carrying instructions for execution by the machine 1200, and including digital or analog communication signals or other intangible medium for facilitating communication of such software.

[0073] Various Notes Each of the non-limiting aspects in this document can stand alone or be combined in various permutations or combinations with one or more of the other aspects or other subjects described in this document.

[0074] The above detailed description includes references to the accompanying drawings, which form part of the detailed description. The drawings illustrate, for illustrative purposes, specific embodiments in which the present invention can be carried out. These embodiments are also generally referred to as “examples.” Such examples may include elements in addition to those shown or described. However, the inventors also intend examples in which only the shown or described elements are provided. Furthermore, the inventors also intend examples in which, with respect to a particular example (or one or more aspects thereof) or with respect to other examples (or one or more aspects thereof) shown or described herein, any combination or permutation of the shown or described elements (or one or more aspects thereof) is used.

[0075] In the event of any conflicting use between this document and any document incorporated in this manner by reference, the use in this document shall prevail.

[0076] In this specification, the terms “a” or “an” are used to mean one or more, independently of any other instances or uses of “at least one” or “one or more,” as is common in patent literature. In this specification, the term “or” is used to mean non-exclusive “or,” such that “A or B” includes “A but not B,” “B but not A,” and “A and B.” In this specification, the terms “including” and “in which” are used as plain English synonyms for the terms “comprising” and “wherein,” respectively. Furthermore, in the following claims, the terms “including” and “comprising” are open-ended, meaning that any system, device, article, composition, formulation or process that includes elements in addition to those enumerated after such terms in the claims is still considered to be within the scope of those claims. Furthermore, in the following claims, terms such as “first,” “second,” and “third” are used merely as labels and are not intended to impose numerical requirements on those subjects.

[0077] Examples of the methods described herein can be executed at least partially by a machine or computer. Some examples may include computer-readable or machine-readable media encoded with instructions that can be operated to configure an electronic device to perform the methods described in the examples above. Implementations of such methods may include code such as microcode, assembly language code, or high-level language code. Such code may include computer-readable instructions for performing various methods. The code may form part of a computer program product. Such instructions may be read and executed by one or more processors, for example, to enable the execution of an operation including a method. The instructions may be, but are not limited to, source code, compiled code, interpreted code, executable code, static code, or dynamic code, or any other suitable form. Furthermore, in one example, the code may be tangibly stored in one or more volatile, non-temporary, or non-volatile tangible computer-readable media, either during execution or at some other point in time. Examples of these tangible computer-readable media may include, but are not limited to, hard disks, removable magnetic disks, removable optical disks (e.g., compact discs and digital video discs), magnetic cassettes, memory cards or sticks, random access memory (RAM), and read-only memory (ROM).

[0078] The above description is illustrative and not limiting. For example, the above examples (or one or more of their embodiments) can be used in combination with each other. Considering the above description, other embodiments can be used by those skilled in the art. The abstract is provided to enable readers to quickly confirm the nature of the technical disclosure. The abstract is submitted with the understanding that it is not to be used to interpret or limit the scope of the claims. Furthermore, in the above detailed description, various features may be grouped together to streamline the disclosure. This should not be interpreted as meaning that any disclosed features not claimed are essential to any claim. Rather, the subject matter of the invention may lie in fewer features than all the features of a particular disclosed embodiment. Therefore, the following claims are incorporated into the detailed description as examples or embodiments, and each claim stands independently as a separate embodiment, and such embodiments can be combined with each other in various combinations or permutations. The scope of the invention should be determined by reference to the appended claims, together with the entire scope of equivalents to which such claims are granted.

Claims

1. A machine-driven method for performing acoustic testing (machine-driven method), To generate each acoustic pulse transmission using a specified pulse sequence corresponding to a specified code, Acquisition of acoustic echo data indicating scattered or reflected acoustic energy in response to each of the aforementioned acoustic pulse transmissions, The acquired acoustic echo data is filtered using a finite impulse response (FIR) filter to provide filtered acoustic echo data. The representations of the filtered acoustic echo data are coherently summed to provide pixel or voxel element values. A machine execution method including

2. The machine execution method according to claim 1, comprising a code having an autocorrelation peak corresponding to an autocorrelation main lobe when the specified pulse sequence is consistent with a representation of the specified pulse sequence itself having a zero sample offset, and a lower autocorrelation value corresponding to an autocorrelation side lobe when the specified pulse sequence is consistent with a representation of the specified pulse sequence itself having a non-zero sample offset.

3. The machine execution method according to claim 1 or 2, wherein the specified pulse sequence includes a Barker code.

4. The machine execution method according to claim 1 or claim 2, wherein the specified pulse sequence includes a code having an odd number of digits.

5. The machine execution method according to any one of claims 1 to 4, wherein the latter half of the designated pulse sequence after the midpoint of the designated pulse sequence includes a mirror image of the first half of the designated pulse sequence, but the symbols alternate for each consecutive digit.

6. A machine execution method according to any one of claims 1 to 5, wherein the designated pulse sequence includes a code that exhibits zero-value autocorrelation sidelobes at odd-value sample offsets when the designated pulse sequence is consistent with the representation of the designated pulse sequence itself having a non-zero sample offset.

7. The FIR filter includes coefficients established to result in the deconvolution of the specified pulse sequence, The machine execution method according to any one of claims 1 to 6, wherein the machine execution method includes establishing the coefficients by evaluating a set of candidate impulse response filter coefficients according to an amplitude criterion.

8. The machine execution method according to claim 7, wherein applying the amplitude criterion includes determining the count of candidate impulse response filter coefficients corresponding to candidate FIR filters having amplitudes greater than a specified amplitude threshold as the proportion of filter coefficients having the maximum amplitude.

9. The machine execution method according to claim 8, wherein the specified amplitude threshold is 1 percent of the maximum amplitude.

10. The machine execution method according to claim 8 or 9, wherein the candidate FIR filter is truncated by dropping candidate impulse response filter coefficients below the specified amplitude threshold in order to provide the FIR filter to be used to bring about deconvolution.

11. The machine execution method according to any one of claims 1 to 10, wherein the designated code or the FIR filter is established using the criterion that the count of the impulse response filter coefficients is constrained to a specified multiple of the number of digits in the designated code.

12. The machine execution method according to claim 11, wherein the specified multiple is 5.

13. The machine execution method according to any one of claims 1 to 12, wherein the FIR filter is established by evaluating a noise averaging metric for a candidate FIR filter.

14. The machine execution method according to any one of claims 1 to 13, wherein the filtering acoustic echo data includes A scan data.

15. Coherently summing the representations of the filtered acoustic echo data includes performing full-focus (TFM) beamforming by using the specified pulse sequence for each transmit / receive acquisition in a matrix capture / acquisition scheme. The machine execution method according to any one of claims 1 to 14, wherein the pixel or voxel element values ​​include data for an image generated using the TFM beamforming.

16. The specified pulse sequence is defined by a series of digits having values ​​selected from the set {1, -1}. The machine execution method according to any one of claims 1 to 15, wherein the values ​​{1} and {-1} correspond to specified positive and negative output amplitude values ​​from a pulse generation circuit used to generate the respective acoustic pulse transmissions.

17. A system for performing acoustic inspections, pulse generation circuit, Receiving circuit and At least one processor circuit, A memory circuit that, when executed by at least one processor circuit, includes an instruction that causes the system to execute the machine execution method described in any one of claims 1 to 14, A system that is equipped with [the following].

18. A machine-driven method for coded radiation for acoustic inspection (machine-driven method), To establish a code for use in acoustic pulse transmission, which is defined by a specified pulse sequence exhibiting zero-value autocorrelation sidelobes at odd-value sample offsets, To establish a finite impulse response (FIR) filter for use in filtering acquired acoustic echo data received in response to the aforementioned acoustic pulse transmission, wherein the FIR filter results in the deconvolution of the specified pulse sequence, A machine execution method including

19. The machine execution method according to claim 18, wherein the latter half of the designated pulse sequence after the midpoint of the designated pulse sequence includes a mirror image of the first half of the designated pulse sequence, but the symbols alternate for each consecutive digit.

20. The machine execution method according to claim 18 or 19, wherein the FIR filter is established by evaluating a set of candidate impulse response filter coefficients according to an amplitude criterion, which includes determining the count of candidate impulse response filter coefficients corresponding to candidate FIR filters having an amplitude greater than a specified amplitude threshold as the ratio of the filter coefficients having the maximum amplitude.

21. The machine execution method according to claim 20, wherein the candidate FIR filter is truncated by dropping candidate impulse response filter coefficients below the specified amplitude threshold in order to provide the FIR filter to be used to bring about deconvolution.

22. The machine execution method according to any one of claims 18 to 21, wherein the designated pulse sequence or the FIR filter is established using the criterion that the count of the impulse response filter coefficients is constrained to a designated multiple of the number of digits in the designated code.

23. The machine execution method according to claim 22, wherein the specified multiple is 5.

24. The specified pulse sequence is defined by a series of digits having values ​​selected from the set {1, -1}. The machine execution method according to any one of claims 18 to 23, wherein the values ​​{1} and {-1} correspond to specified positive and negative output amplitude values ​​from a pulse generation circuit used to generate each acoustic pulse transmission.

25. The machine execution method according to any one of claims 18 to 24, wherein the FIR filter is established by evaluating the noise averaging metric in the candidate FIR filter.

Images