Signal processing device and signal processing method

Abstract

To perform imaging processing on a test object at high speed.SOLUTION: A signal processing device comprises a processing part performing: reading processing for reading a propagation model representing a propagation path and propagation time of a plane wave when the plane wave dispatched from a plurality of transmission / reception wave elements in an inclination direction inclined to a normal direction of a boundary face of a test object propagates inside of the test object, reflects at a focal point and reaches the transmission / reception wave element; acquisition processing for acquiring reception data of the plane wave received by the plurality of transmission / reception wave elements when the plurality of transmission / reception wave elements emit the plane wave to the test object in the inclination direction; and imaging processing for generating imaging data of the test object using compression sensing on the basis of the read propagation model and the acquired reception data. The processing part partitions an imaging range of the test object into a plurality of partial images in the imaging processing, and processes in parallel the reception data for each partial image using unit time data which is reception data of the plane wave for each receipt time in the compression sensing.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a signal processing apparatus and a signal processing method.

Background Art

[0002] Compressive sensing is used in various inspection technology fields such as, for example, magnetic resonance imaging. In an imaging method using compressive sensing, it is assumed that a high-intensity distribution in a visualization region is sparse, and it is possible to restore the entire image of an inspection target using a small number of observation results or to perform imaging of the inspection target with high resolution (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In an imaging method using compressive sensing as described above, it is required to perform imaging processing of an inspection target at high speed.

[0005] The present disclosure has been made in view of the above, and an object thereof is to provide a signal processing apparatus and a signal processing method capable of performing imaging processing of an inspection target at high speed.

Means for Solving the Problems

[0006] The signal processing apparatus according to the present disclosure includes a processing unit that performs a reading process of reading a propagation model indicating a propagation path and a propagation time of a plane wave when the plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface of an inspection target from a plurality of transmitting and receiving elements propagates inside the inspection target, is reflected at a focal point, and reaches the transmitting and receiving elements, an acquisition process of acquiring reception data of the plane wave received by the plurality of transmitting and receiving elements when the plurality of transmitting and receiving elements transmit the plane wave in the inclined direction with respect to the inspection target, and an imaging process of generating imaging data of the inspection target by compressive sensing based on the read propagation model and the acquired reception data. In the imaging process, the processing unit divides an imaging range of the inspection target into a plurality of partial images, and in the compressive sensing, processes the reception data in parallel for each of the partial images using unit time data which is the reception data at each reception time of the plane wave.

[0007] The signal processing method according to the present disclosure includes a reading step of reading a propagation model indicating a propagation path and a propagation time of a plane wave when the plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface of an inspection target from a plurality of transmitting and receiving elements propagates inside the inspection target, is reflected at a focal point, and reaches the transmitting and receiving elements, an acquisition step of acquiring reception data of the plane wave received by the plurality of transmitting and receiving elements when the plurality of transmitting and receiving elements transmit the plane wave in the inclined direction with respect to the inspection target, and an imaging step of generating imaging data of the inspection target by compressive sensing based on the read propagation model and the acquired reception data. In the imaging step, the imaging range of the inspection target is divided into a plurality of partial images, and in the compressive sensing, the reception data is processed in parallel for each of the partial images using unit time data which is the reception data at each reception time of the plane wave.

Advantages of the Invention

[0008] According to the present disclosure, it is possible to provide a signal processing apparatus and a signal processing method capable of performing imaging processing of an inspection target at high speed.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0010] Hereinafter, embodiments of a signal processing apparatus and a signal processing method according to the present disclosure will be described with reference to the drawings. Note that the present invention is not limited by these embodiments. In addition, the components in the following embodiments include those that can be replaced and are easy for those skilled in the art, or those that are substantially the same.

[0011] FIG. 1 is a schematic diagram showing an example of a measurement system SYS including a signal processing apparatus 100 according to the present embodiment. As shown in FIG. 1, the measurement system SYS includes a sensor 10 and a signal processing apparatus 100.

[0012] Sensor 10 has a plurality of transmitting and receiving elements 11 that output detection transmission waves, receive the reflected waves of the output transmission waves, and perform detection of the inspection target 40. The sensor 10 is arranged in a state where a plurality (M) of transmitting and receiving elements 11 are arranged in an array. M is a real number. In the present embodiment, the transmission wave is a plane wave, which is an acoustic signal such as ultrasonic waves. Note that in this embodiment, ultrasonic waves are used for detection, but radio waves or the like may also be used. In this embodiment, the case of performing PWI (Plane Wave Imaging) using a plane wave will be described as an example.

[0013] In the present embodiment, the sensor 10 is a wedge-shaped sensor that causes the beam of the transmission wave to be incident in the inclination direction θ i (i = 1, 2,..., N B ) with respect to the normal direction (y direction in FIGS. 3 and 4) of the interface 41 of the inspection target 40. Here, N B is a real number representing the number of beams. Also, let the received signal vector (real number) of one ping be

Number

Number

[0014] The signal processing device 100 is connected to the sensor 10 via, for example, a pulsar receiver 50. The signal processing device 100 processes the signal received by the transmitting and receiving element 11 to detect the surroundings. The signal processing device 100 includes an arithmetic unit 20 and a storage unit 30. The arithmetic unit 20 includes a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit) as, for example, a processor. The arithmetic unit 20 performs various operations. The storage unit 30 includes, for example, a memory such as a RAM (Random Access Memory) and a ROM (Read Only Memory), and a storage such as an HDD (Hard Disk Drive).

[0015] The arithmetic unit 20 reads a predetermined propagation model indicating the propagation path of a plane wave when a plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface 41 of the inspection target 40 propagates inside the inspection target 40, reflects at the focal point p, and reaches the transmitting and receiving element 11 from a plurality of transmitting and receiving elements 11. The arithmetic unit 20 acquires reception data received by the plurality of transmitting and receiving elements 11 when the plurality of transmitting and receiving elements 11 transmit a plane wave in the inclined direction with respect to the inspection target 40. The arithmetic unit 20 generates imaging data of the inspection target 40 by compressive sensing based on the read propagation model and the acquired reception data. The detailed operation of the arithmetic unit 20 will be described later.

[0016] In the propagation model, a range of the interface 41 corresponding to the array aperture by the plurality of transmitting and receiving elements 11 is set inside the inspection target 40. In this case, when performing, for example, compressive sensing, the arithmetic unit 20 can invalidate plane waves that do not pass through the set range of the interface 41.

[0017] The arithmetic unit 20 executes each of the above processes by reading and executing a program (software) from the storage unit 30. The storage unit 30 stores various information such as the calculation content and program of the arithmetic unit 20. The storage unit 30 may store the processing result detected by the sensor 10, that is, the result of the detection.

[0018] The memory unit 30 causes a computer to execute a reading process of reading a predetermined propagation model indicating the propagation path and propagation time of a plane wave when the plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface 41 of the inspection target 40 from a plurality of transmission / reception elements 11 propagates inside the inspection target 40, reflects at a focal point, and reaches the transmission / reception element 11, an acquisition process of acquiring reception data received by the plurality of transmission / reception elements 11 when the plurality of transmission / reception elements 11 transmit a plane wave in the inclined direction with respect to the inspection target 40, and an imaging process of generating imaging data of the inspection target 40 by compressive sensing based on the read propagation model and the acquired reception data. In the imaging process, an imaging range of the inspection target is partitioned into a plurality of partial images, and a signal processing program for processing the reception data in parallel for each partial image using unit time data, which is the reception data for each reception time of the plane wave by compressive sensing, is stored.

[0019] Hereinafter, the measurement process of the measurement system according to the present embodiment will be described. FIG. 2 is a flowchart showing an example of the signal processing method according to the present embodiment. As shown in FIG. 2, the signal processing method according to the present embodiment includes a reading step S10, an acquisition step S20, and an imaging step S30.

[0020] In the reading step S10, the arithmetic unit 20 reads a predetermined propagation model indicating the propagation path of a plane wave when the plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface 41 of the inspection target 40 from a plurality of transmission / reception elements 11 propagates inside the inspection target 40, reflects at a focal point, and reaches the transmission / reception element 11.

[0021] In the measurement process of the measurement system according to the present embodiment, when the beam i of the transmitted wave reaches the focal position inside the imaging range

Equation

Equation

Number

Number

[0022] Figure 3 is a diagram schematically showing an example of the propagation path (forward path) of the transmitted wave in the propagation model. In Figure 3, the path from the transmission part of the sensor 10 to the focal point p is shown. In the forward path, it is regarded as the propagation of a plane wave. d i (1) 、d i (2) represents each distance of the propagation path of the forward path shown in Figure 3.

[0023] Figure 4 is a diagram schematically showing an example of the propagation path (return path) of the transmitted wave in the propagation model. In Figure 4, the path from the focal point p to the reception part of the sensor 10 is shown. In the return path, it is regarded as the propagation of a spherical wave. d i (3) 、d i (4) represents each distance of the propagation path of the return path shown in Figure 4.

[0024] Next, in the reception data acquisition step S20, the calculation unit 20 acquires reception data received by the plurality of transmission / reception elements 11 when the plurality of transmission / reception elements 11 transmit a plane wave in an inclined direction with respect to the inspection object 40.

[0025] Next, in the compressive sensing processing step S30, the arithmetic unit 20 generates imaging data of the inspection target 40 by compressive sensing based on the read propagation model and the acquired reception data. The arithmetic unit 20 divides the imaging range of the inspection target 40 into a plurality of partial images, and in compressive sensing, the reception data is processed in parallel for each partial image using the unit time data which is the reception data for each reception time of the plane wave.

[0026] FIG. 5 and FIG. 6 are diagrams schematically showing an example of the imaging range by the measurement system. The mathematical model representing the correspondence relationship between the signal vector and the reflection intensity vector necessary for applying compressive sensing to imaging is

Number

Number

Number

Number

Number

Number

Number

[0027] In [Equation 16], j and ω are imaginary units and represent the angular frequencies of the transmission signals respectively. Also, A i (k) is denoted as the observation matrix. It is the coefficient in [Equation 16] [Mathematics] is an expression representing whether the reflected wave of beam i from position p i arrives at the transceiver element 11 at time k. If it arrives, it takes the value "1", and if it does not arrive, it takes the value "0".

[0028] Figure 7 is a flowchart showing the processing flow in the arithmetic unit. Figure 7 shows the processing (offline processing) that does not use the received signal at the transceiver element 11.

[0029] In the offline processing, the pixel information (number of pixels, pixel size, coordinates of each observation pair focus in the imaging range), sampling period, and observation matrix A i (k) of the imaging range of the inspection target 40 are input. In the offline processing, as shown in Figure 6, the arithmetic unit 20 divides the imaging range 60 into sub-images 61 (step S110). The arithmetic unit 20 performs beam passing determination for each divided sub-image (step S120). Also, the arithmetic unit 20 selects the time zones contributing to imaging in each sub-image (step S130). As a result, it is not necessary to use the received data of all beams and all times for parallel processing, so the memory occupancy in the arithmetic unit 20 can be suppressed.

[0030] The calculation unit 20 sets bins in the partial image and selects the times corresponding to the set bins (step S140). Then, the calculation unit 20 divides the observation matrix for each bin in the partial image

Number

Number

Number

Number

Number

Number

Number

Number

Number

[0031] FIG. 8 is a flowchart showing the processing flow in the arithmetic unit. FIG. 8 shows the processing (online processing) using the received signal. In the online processing, after measuring the received signal vector by the transceiver element 11, the arithmetic unit 20 performs processing for real-time imaging. In the online processing, ξ(k) indicated by [Equation 2] is input. As shown in FIG. 8, the arithmetic unit 20 performs a Hilbert transform on ξ(k) to calculate η(k) (step S210). η(k) is a vector that summarizes the received signal vectors of all beams in the same way as ξ(k) indicated by [Equation 2].

Equation

[0032] The arithmetic unit 20 performs a process (preprocessing) of extracting snapshots that contribute to the imaging of each bin using the times corresponding to each bin (indicated by [Equation 22]) calculated in the offline processing (step S220). The snapshots extracted in the preprocessing are adjusted according to [Equation 18].

Equation

[0033] The arithmetic unit 20 transfers [Equation 18] and [Equation 28] from the CPU to the GPU based on this (step S230).

Equation

Equation

[0034] At this time, [Number] Regarding this, by performing the ADMM operation in parallel processing, imaging by parallel processing on the GPU is performed (step S240). Since the ADMM process mainly involves arithmetic operations, appropriate parallel processing can be performed.

[0035] The operation unit 20 returns the reflection intensity vector of each bin calculated by the parallel processing in step S240 [Number] from the GPU to the CPU (step S250), and the overlapping part is added in post-processing (post-processing: step S260) to obtain the pixel value s.

[0036] As described above, in the present disclosure, the signal processing apparatus according to the first aspect reads a propagation model showing the propagation path and propagation time of a plane wave when a plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface of the inspection target 40 from a plurality of transmission / reception wave elements 11 propagates inside the inspection target 40, is reflected at the focal point, and reaches the transmission / reception wave element 11, an acquisition process of acquiring reception data of the plane wave received by the plurality of transmission / reception wave elements 11 when the plurality of transmission / reception wave elements 11 transmit a plane wave in the inclined direction with respect to the inspection target 40, and an imaging process of generating imaging data of the inspection target 40 by compressive sensing based on the read propagation model and the acquired reception data. The operation unit 20 includes an operation unit 20 that divides the imaging range 60 of the inspection target 40 into a plurality of partial images 61 in the imaging process, and in compressive sensing, processes the reception data in parallel for each partial image 61 using bins 70 that are reception data for each reception time of the plane wave.

[0037] Therefore, since the reception data is processed in parallel for each partial image 61 using the bins 70 in compressive sensing, the imaging process of the inspection target can be performed at high speed by parallel processing.

[0038] In the signal processing apparatus according to the second aspect, in the signal processing apparatus according to the first aspect, the calculation unit 20 determines whether or not a plane wave has passed through each sub-image 61, and does not use the processing result of the sub-image 61 through which the plane wave has not passed for generating imaging data. Therefore, since it is not necessary to perform parallel processing on the reception signal data for all beams and all times, the processing amount of the calculation unit 20 can be suppressed. Further, by performing this processing offline, the time required for online processing can be shortened.

[0039] In the signal processing apparatus according to the third aspect, in the signal processing apparatus according to the first and second aspects, the calculation unit 20 sets the sub-images 61 so that the number of pixels included in each sub-image 61 is the same. Therefore, parallel processing for each sub-image 61 can be appropriately performed.

[0040] In the signal processing apparatus according to the fourth aspect, in the signal processing apparatus according to the first to third aspects, the calculation unit 20 sets the sub-images 61 so that the number of pixels included in each sub-image 61 is equal to or less than a threshold value. Therefore, parallel processing for each sub-image 61 can be appropriately performed.

[0041] The signal processing method according to the fifth aspect includes a reading step of reading a propagation model showing the propagation path and propagation time of a plane wave when the plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface of the inspection object 40 from a plurality of transmission / reception wave elements 11 propagates inside the inspection object 40 and is reflected at the focal point and reaches the transmission / reception wave elements 11, an acquisition step of acquiring reception data of the plane wave received by the plurality of transmission / reception wave elements 11 when the plurality of transmission / reception wave elements 11 transmit a plane wave in the inclined direction with respect to the inspection object 40, and an imaging step of generating imaging data of the inspection object 40 by compressive sensing based on the read propagation model and the acquired reception data. In the imaging step, the imaging range 60 of the inspection object 40 is partitioned into a plurality of sub-images 61, and in compressive sensing, the reception data is processed in parallel for each sub-image 61 using the unit time data which is the reception data at each reception time of the plane wave.

[0042] Therefore, in compressive sensing, by using bin 70 to process received data in parallel for each partial image 61, imaging can be performed at high speed through parallel processing. As a result, it becomes possible to accurately image the inspection target 40 over a wide range with high resolution.

Description of Reference Numerals

[0043] 10 Sensor 11 Transmitting / Receiving Element 20 Arithmetic Unit 30 Memory Unit 40 Inspection Target 41 Interface 50 Pulser Receiver 60 Imaging Range 61 Partial Image 70 Unit Imaging Range, Bin 100 Signal Processing Device SYS Measurement System

Claims

1. A reading process of reading a propagation model indicating a propagation path and a propagation time of a plane wave when the plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface of an inspection object from a plurality of transmitting and receiving elements propagates inside the inspection object, reflects at a focal point, and reaches the transmitting and receiving elements; An acquisition process of acquiring reception data of the plane wave received by the plurality of transmitting and receiving elements when the plurality of transmitting and receiving elements transmit the plane wave in the inclined direction with respect to the inspection object; A processing unit that performs an imaging process of generating imaging data of the inspection object by compressive sensing based on the read propagation model and the acquired reception data; In the imaging process, the processing unit divides an imaging range of the inspection object into a plurality of partial images, and in the compressive sensing, the reception data is processed in parallel for each partial image using unit time data that is the reception data for each reception time of the plane wave. A signal processing device.

2. The processing unit determines whether or not the plane wave has passed through each partial image, and does not use the processing result of the partial image through which the plane wave has not passed for generating the imaging data. The signal processing device according to claim 1.

3. The processing unit sets the partial images so that the number of pixels included in each partial image is the same. The signal processing device according to claim 1.

4. The processing unit sets the partial images so that the number of pixels included in each partial image is equal to or less than a threshold value. The signal processing device according to claim 3.

5. A reading step of reading a propagation model indicating a propagation path and a propagation time of a plane wave when the plane wave transmitted in an inclined direction inclined with respect to the normal direction of the interface of an inspection object from a plurality of transmitting and receiving elements propagates inside the inspection object, reflects at a focal point, and reaches the transmitting and receiving elements; An acquisition step of acquiring reception data of the plane wave received by the plurality of transmitting and receiving elements when the plurality of transmitting and receiving elements transmit the plane wave in the inclined direction with respect to the inspection object; An imaging step of generating imaging data of the inspection object by compressive sensing based on the read propagation model and the acquired reception data is included. In the imaging step, the imaging range of the inspection target is partitioned into a plurality of partial images, and in the compressive sensing, the received data is processed in parallel for each of the partial images using the unit time data which is the received data at each reception time of the plane wave. Signal processing method.

Images