CN106651740B - A kind of full focus data fast imaging method of ultrasound based on FPGA and system - Google Patents
A kind of full focus data fast imaging method of ultrasound based on FPGA and system Download PDFInfo
- Publication number
- CN106651740B CN106651740B CN201611077700.6A CN201611077700A CN106651740B CN 106651740 B CN106651740 B CN 106651740B CN 201611077700 A CN201611077700 A CN 201611077700A CN 106651740 B CN106651740 B CN 106651740B
- Authority
- CN
- China
- Prior art keywords
- pixel
- port
- array element
- signal
- pixel value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/76—Architectures of general purpose stored program computers
- G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
- G06F15/7807—System on chip, i.e. computer system on a single chip; System in package, i.e. computer system on one or more chips in a single package
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/76—Architectures of general purpose stored program computers
- G06F15/78—Architectures of general purpose stored program computers comprising a single central processing unit
- G06F15/7839—Architectures of general purpose stored program computers comprising a single central processing unit with memory
- G06F15/7842—Architectures of general purpose stored program computers comprising a single central processing unit with memory on one IC chip (single chip microcontrollers)
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The full focus data fast imaging method of the ultrasound that the present invention relates to a kind of based on FPGA and system, the first ultrasonic signal that the full focus data fast imaging method of ultrasound based on FPGA, which includes: acquisition, to be emitted by each array element of at least one the first array element is to pixel, and at least one second ultrasonic signal is reflected at least one the second array element by pixel, and the data after acquisition are stored；According to the first ultrasonic signal and at least one second ultrasonic signal, at least one corresponding propagation delay time of pixel is determined；According to the corresponding data of pixel of at least one propagation delay time of pixel retrieval storage；According to the corresponding pixel value of the corresponding Data Synthesis pixel of pixel；The corresponding pixel value of pixel is converted, determines the analytic signal of pixel value, and be imaged after the envelope for obtaining analytic signal.Based on imaging method of the invention, imaging parallelization degree is improved, imaging efficiency is accelerated, while significantly improving the influence of array number and pixel number increase to efficiency.
Description
Technical field
The present invention relates to ultrasound detection field, in particular to the full focus data fast imaging side of a kind of ultrasound based on FPGA
Method and system.
Background technique
Ultrasonic full focus data (TFM) imaging is a kind of advanced phased array detection method that (FMC) is captured based on complete matrix,
Compared to conventional phased array imaging method, there is advantages, the tools such as high-resolution, detection range be big to have broad application prospects.
TFM is a kind of imaging method based on FMC, as shown in Figure 1, TFM is to detection zone gridding, according to array element and picture
The space length of element calculates transmission delay, retrieves corresponding FMC data, completes the superposition synthesis of pixel.
The value I of any pixel point P (x, z) is provided by formula 1.
hijFor excitation array element be i when, the reception data of j array element,
And it have passed through Hilbert transformation.xiAnd xjIt is the horizontal axis coordinate of array element i and j respectively, c is the velocity of sound, and N is element number of array.
For a long time, full focus data can only non real-time realization in the lab, a large amount of data processing and be calculated as
Its bottleneck for moving towards application.Concurrent operation platform is become with its efficient concurrent executive capability and is improved total focus imaging efficiency
Important method.
Concurrent operation platform mainly has GPU, multi-core CPU and FPGA etc..TFM algorithm acceleration based on GPU+CPU is in recent years
Carry out main implementation, but does not account for the influence that hardware acquisition comes with output transmission, while being limited to software processor
Parallel ability, TFM computational efficiency is still insufficient, and when array element or pixel increase, imaging efficiency can sharply decline.
FPGA provides good parallel tables for the efficient calculating of TFM.Have within 2013 and researchs and proposes based on multiple FPGA
TFM imaging, reaches 73Hz imaging rate under 16 array elements, 60*60 amount of pixels, but when pixel is increased to 128*128, efficiency is substantially
Drop to 20Hz or less.Reason is that current FPGA design method parallelization degree is insufficient.Delay is such as focused using software
Calculate and pass down the mode of storage, cause FPGA need to spend multiple periods could read latch value, calculate a pixel, limit
Computational efficiency is made；Pixel can only be synthesized serially simultaneously, and when pixel resolution is higher, imaging rate is decreased obviously, it is difficult to
Meet using mechanical scanning as the quick testing requirements of the TFM of representative.
Prior art TFM imaging, due to the parallelization of calculation method it is insufficient caused by imaging rate it is low, by array element and
Pixel quantity influences the problems such as obvious.
Summary of the invention
It is fast the purpose of the present invention is in view of the drawbacks of the prior art, proposing a kind of full focus data of the ultrasound based on FPGA
Fast imaging method and system effectively increases TFM imaging efficiency, and significantly improve array number and pixel number compared with prior art
Increase the influence to imaging efficiency, provides fast imaging scheme for the high speed TFM detection using mechanical scanning as representative.
To achieve the above object, on the one hand, the full focus data of the ultrasound that the present invention provides a kind of based on FPGA is rapid-result fastly
Image space method, should be specifically included based on the ultrasonic full focus data fast imaging of FPGA: acquisition is by the every of at least one the first array element
First ultrasonic signal of one array element transmitting reflects at least one second ultrasonic signal to extremely to pixel, and by pixel
Few second array element, and the data after acquisition are stored；Believed according to the first ultrasonic signal and at least one second ultrasonic wave
Number, determine at least one corresponding propagation delay time of pixel；According to the picture of at least one propagation delay time of pixel retrieval storage
The corresponding data of vegetarian refreshments；According to the corresponding pixel value of the corresponding Data Synthesis pixel of pixel；Pixel corresponding to pixel
Value is converted, and determines the analytic signal of pixel value, and be imaged after the envelope for obtaining analytic signal.
Preferably, the step of and storing after acquisition specifically includes: by corresponding first ultrasonic signal of pixel and extremely
Few second ultrasonic signal is stored in dual port memories.
Preferably, method further include: when acquisition and retrieval concurrently execute, by configured with first port and second port
The port that the first port of dual port memories is inputted as data, the dual port memories configured with first port and second port
Search and output of the second port as data；After acquisition, when retrieval executes, by being configured with first port and second port
The first port of dual port memories and the second port of the dual port memories configured with first port and second port make respectively
For search and output.
Preferably, according to the first ultrasonic signal and at least one second ultrasonic signal, determine that pixel is corresponding extremely
The step of few propagation delay time, specifically includes: calculating at least one corresponding first array element of pixel by Real-time Delay algorithm
Each array element and at least one second array element between sound path；According to sound path and the velocity of sound, at least one propagation delay time is determined.
Preferably, calculated by Real-time Delay algorithm each array element of corresponding at least one the first array element of pixel with
The step of sound path between at least one second array element are as follows:
Wherein, the abscissa x of i-th of array element of at least one the first array elementi, j-th gust of at least one the second array element
The abscissa x of memberj, pixel coordinate is (x, z).
Preferably, method further include: by field programmable gate array chip to the first ultrasonic signal and at least
One the second ultrasonic signal concurrent obtains at least one corresponding propagation delay time of pixel.
Preferably, the step of converting to pixel, obtaining the analytic signal of pixel specifically includes: being become according to Hilbert
It changes and pixel value is converted, pixel value is become into frequency-region signal from time-domain signal；According to the pixel value and frequency-region signal of time-domain signal
Pixel value synthesize analytic signal.
Preferably, the step of and being imaged after the envelope for obtaining analytic signal specifically includes: on the envelope by analytic signal
Host computer is reached to be imaged.
Preferably, Hilbert transform can be with are as follows: FIR filter.
On the other hand, the embodiment of the present invention provides a kind of full focus data fast imaging system of the ultrasound based on FPGA, should
The full focus data fast imaging system of ultrasound based on FPGA includes: acquisition module, memory module, computing module, retrieval module
And conversion module；Acquisition module is used for, and acquisition is believed by the first ultrasonic wave that each array element of at least one the first array element emits
Number at least one second ultrasonic signal is reflected at least one the second array element to pixel, and by pixel；Memory module is used
In by the data storage after acquisition；Computing module is used for, and is believed according to the first ultrasonic signal and at least one second ultrasonic wave
Number, determine at least one corresponding propagation delay time of pixel；Retrieval module is used for, according at least one propagation delay time of pixel
Retrieve the corresponding data of pixel of storage；Synthesis module is used for, corresponding according to the corresponding Data Synthesis pixel of pixel
Pixel value；Conversion module is used for, and is converted to the corresponding pixel value of pixel, is determined the analytic signal of pixel value.
The present invention is calculated using in FPGA piece, is participated in without CPU, and the parallelization degree of TFM imaging is effectively increased, and is accelerated
Imaging efficiency, while significantly improving array number and pixel number increases influence to efficiency.
Detailed description of the invention
Fig. 1 is ultrasound in the prior art totally according to focal imaging figure；
Fig. 2 is a kind of method knot for the full focus data fast imaging of ultrasound based on FPGA that the embodiment of the present invention provides
Structure flow chart；
Fig. 3 is polar coordinates transition diagram；
Fig. 4 is frequency, the amplitude, phase diagram of FIR filter；
Fig. 5 is that evidence is focused into the quickly structure as system to a kind of ultrasound based on FPGA provided in an embodiment of the present invention totally
Schematic diagram；
Fig. 6 is a specific image based on the full focus data fast imaging method of ultrasound in Fig. 1 based on FPGA.
Specific embodiment
Below by drawings and examples, technical scheme of the present invention will be described in further detail.
The present invention devises full parellel numerical procedure in FPGA piece, including is based on CORDIC (Coordinate
Rotational Digital Computer) Real-time Delay calculate, parallel pixel synthesis and pixel synthesis after
Hilbert transformation.Realize that the real-time parallel for focusing delay calculates respectively, the concurrently synthesis and pixel analytic signal of more pixels
Acquisition.
Fig. 2 is a kind of stream for the full focus data fast imaging method of ultrasound based on FPGA that the embodiment of the present invention provides
Cheng Tu.As shown in Fig. 2, the step of ultrasonic full focus data fast imaging, specifically includes:
Step S100: the first ultrasonic signal that acquisition is emitted by each array element of at least one the first array element is to pixel
Point, and at least one second ultrasonic signal is reflected at least one the second array element by pixel, and store after acquisition；
For with N number of array element excitation-reception combination, each FMC acquisition is only to one in ultrasonic phase array probe
First the first ultrasonic signal of array element excitation-emission, N number of second array element are primary to swash for N channel phased array as receiving unit
Hair can be obtained by N number of second ultrasonic signal, and the second ultrasonic wave is all via N number of received echo-signal of second array element.The
One array element is that the transmitting array element being excited is used to emit the first ultrasonic signal, and the second array element is then to receive the second ultrasonic signal
Reception array element.
For such as 16 array element systems, process are as follows: array element 1 emits, and array element 1-16 acquires transmitted wave and acquisition ultrasound instead
It is emitted back towards wave；After having acquired, array element 2 emits, and array element 1-16 acquires transmitted wave and ultrasonic reflection echo；And so on, Zhi Daozhen
Member 16 emits, and array element 1-16 acquires ultrasonic reflection echo.
Since the number of the clarity different pixels point of imaging is also different.It is after the completion of FMC acquisition, pixel is corresponding
In first ultrasonic signal and N number of second ultrasonic signal storage dual port memories.
Dual port memories (True Dual-port RAM) are resources in the piece of FPGA, can meet a variety of reading and writing datas and want
It asks.Compared to cache way outside piece, degree is read and write using the parallelization that channel can be improved in resource storage in piece.Money is stored in piece
In source, dual port memories are compared to single port RAM and FIFO, it can be achieved that double reading data bandwidth.
Step S110: according to the first ultrasonic signal and at least one second ultrasonic signal, determine that pixel is corresponding
At least one propagation delay time；
For with N number of array element excitation-reception combination, N number of receiving unit receives N number of second ultrasonic signal respectively,
Each receives array element and only receives second ultrasonic signal.
Time delay according to ultrasonic phase array pop one's head in first ultrasonic signal of array element excitation-emission to pixel, and by
The second ultrasonic signal of pixel point reflection to each of N number of reception array element sound path of array element and the ratio of the velocity of sound obtains.
Specifically, according to the first ultrasonic signal and at least one second ultrasonic signal, determine that pixel is corresponding extremely
The step of few propagation delay time, specifically includes:
The array element excitation-emission the in ultrasonic phase array probe is calculated according to Real-time Delay algorithm (cordic algorithm)
One ultrasonic signal is to pixel, and by the second ultrasonic signal of pixel point reflection to each of N number of reception array element array element
Sound path；By the ratio of sound path and the velocity of sound, propagation delay time is determined.
Cordic algorithm is by a series of fixations, angle related with operation radix, the continuous beat iteration of the angle
To approach required rotation angle, it is suitble to the calculating of coordinate conversion and trigonometric function etc..
Cordic algorithm be applied to coordinate conversion can by any input vector (X, Y) along unit circle rotate, until this to
The Y-coordinate of amount is 0, rotation angle, θ and vector field homoemorphism X ' is obtained, to realize the conversion of polar coordinates to rectangular co-ordinate.
The transmission delay of TFM is obtained by sound path/velocity of sound.Sound path between pixel and array element can be converted in the coordinate of CORDIC
In acquire, do not need direct extracting operation, greatly facilitate sound path calculating.
Such as, the input coordinate X of array element i and pixel P (x, z), CORDIC are the abscissa of array element i and pixel P (x, z)
Difference, ordinate Y be array element i and pixel P (x, z) Z axis coordinate difference, the postrotational amplitude result of coordinate is exactly pixel P
Unidirectional sound path between (x, z) and array element i.The mathematical expression of the process are as follows:
Two-way sound path between combination and pixel P (x, z) is received by array element i excitation-array element j are as follows:
One propagation delay time are as follows:
Wherein, propagation delay time of the T between array element i and array element j, c are the velocity of sound.
That is array element i excitation-N number of array element receives the ratio of combination with the velocity of sound, i.e. N number of needed for each pixel in TFM
Propagation delay time.
Optionally, the embodiment of the present invention uses N number of CORDIC computing module, it can be achieved that N number of propagation delay time based on FPGA
Concurrent meets the quick calculating requirement of the corresponding pixel value of pixel.Compared to conventional software calculation, the present invention
Embodiment by FPGA internal calculation, does not need any storage and reading completely, can really realize that real-time parallel calculates, effectively improve
TFM pixel computational efficiency.
Step S120: according to the corresponding data of pixel of at least one propagation delay time of pixel retrieval storage；
Step S130: according to the corresponding pixel value of the corresponding Data Synthesis pixel of pixel.
For being combined by N number of array element excitation-reception, each pixel can correspond to N number of propagation delay time, by N number of transmission
Time delay retrieves the data being stored in dual port memories, is superimposed the pixel value of the data pixel retrieved.
Since TFM imaging can be related to many pixel values, generally first pixel value can be stored.It is needing to carry out to pixel
When value is imaged, directly from the position acquisition pixel value of storage.
Step S140: the corresponding pixel value of pixel is converted, determines the analytic signal of pixel value.
Pixel value is converted according to Hilbert transformation, the pixel value of time-domain signal is changed into the picture of frequency-region signal
Element value；The pixel value of the pixel value of time-domain signal and frequency-region signal is combined into analytic signal.
Hilbert transformation is carried out to array of pixel values, obtains the analytic signal of pixel value, while guaranteeing imaging effect,
Meet the pipeline design of FPGA, and reduces resource occupation.
In the field of communications, Hilbert transformation is the effective means that envelope is taken to signal.The post-processing based on FMC at
As in, usually first convert to obtain the analytic signal of ultrasonic echo with Hilbert, then obtain the envelope signal of analytic signal.It compares
In original signal direct imaging, the envelope of analytic signal keeps imaging more smooth.
In conventional TFM imaging, mainly by software realization, point-by-point operation is improved to software real-time for Hilbert transformation
It makes troubles.Hilbert transformation is linear transform process, meets principle of stacking, therefore in TFM imaging, signal is first superimposed and obtains
Hilbert converts to obtain analytic signal and be superimposed again with elder generation Hilbert transformation theoretically to have again after the corresponding pixel value of pixel
There is same effect.However the latter more meets the resource-effective and flowing water line computation of FPGA.
The execution sequence of step S130 and step S140, which can be, to be first carried out step S140 and executes step S130 again.The present invention
Embodiment executes step S140 after only enumerating execution step S130, first carries out step S140 and is not discussing here, imaging
Effect is identical.
Hilbert transformation is that physics is not achievable, therefore realizes that Hilbert becomes using FIR filter in FPGA
Change identical function.
The embodiment of the present invention uses 128 rank FIR filters, and frequency amplitude phase response is as shown in figure 3, wherein amplitude
Spectrum has all-pass characteristic, and phase has strictly linear, meets Hilbert transformation and requires.
Step S150: the envelope for obtaining analytic signal is uploaded to host computer imaging.
Obtain the envelope of analytic signal；Envelope signal is uploaded to host computer and shows image, to complete TFM imaging.
Since there are two configurable first ports and second port for dual port memories configuration, ultrasound is totally according to focal imaging
Method further include:
Step S160: when FMC, which is acquired, concurrently to be executed with retrieval, by the dual-port for being configured with first port and second port
The port that the first port of memory is inputted as data, the second end of the dual port memories configured with first port and second port
Search and output of the mouth as data；
After acquisition, when retrieval executes, by first configured with first port and the dual port memories of second port
Port is defeated as retrieving by the second port configured with first port and the dual port memories of second port as search and output
Out.
Specifically, for dual port memories configured with configurable first port and second port, first port is port A, the
Two-port netwerk is port B.
When FMC slow down and retrieve concurrently execute when, the port that port A is inputted as data, port B is defeated as data retrieval
Out；When FMC acquisition finishes, when only carrying out retrieval, port A and port B can be mutually as search and output, port A and port B
It is independent, the calculating of 2N propagation delay time of 2 pixels can be supported simultaneously, and then picture is retrieved by port A and port B respectively
The corresponding data of vegetarian refreshments synthesize the corresponding pixel value of 2 pixels.
When for a pixel, it is based on FPGA, using N number of CORDIC computing module, it can be achieved that the meter of N number of propagation delay time
It calculates；For 2 pixels, by N number of cordic algorithm module composition at a time delay module, each pixel is according to delay mould
Block obtains corresponding N number of transmission delay, and then can obtain the calculating of 2N transmission delay of 2 pixels, and then synthesizes 2 pixels
The corresponding pixel value of point.
Compared to normal cache mode in data buffer storage outside piece and piece, the embodiment of the present invention can realize the concurrent of 2 pixels
It calculates, pixel computational efficiency is doubled, imaging rate is accelerated.With the promotion of resource in FPGA, increase in array number N
The example of dual port memories can be improved in added-time, guarantees the FMC acquisition in each channel, reduces influence of the N increase to TFM efficiency.
It should be noted that the one-dimensional linear array probe of N number of array element carries out FMC acquisition, each array element is sequentially excited, often
All array elements while and respective individual reception data after primary excitation.Sequence excites each array element, when can use N number of delay
Sequence, delay timing is corresponding with array element, realizes the sequence excitation of array element；It can also be by the N number of timing of software realization, when reaching first
Between when, excite first array element to pixel, and so on.
The embodiment of the present invention proposes the full parellel TFM computing architecture based on FPGA, including is based on CORDIC
The Real-time Delay of (Coordinate Rotational Digital Computer) calculates, parallel pixel synthesis and pixel are closed
Hilbert transformation after.The Real-time and Concurrent that the present invention can be achieved to focus delay calculates, and is not necessarily to any storage, while can realize
The concurrently synthesis of multiple pixels.Compared to conventional method, the present invention has preferably played FPGA at parallel computation and signal
Advantage in reason effectively increases TFM imaging rate, can significantly improve the influence of array number and pixel number increase to efficiency.This
Invention can detect for the high speed TFM using mechanical scanning as representative and provide fast imaging scheme.
Fig. 3 is polar coordinates transition diagram.It converts as shown in figure 3, cordic algorithm is applied to polar coordinates by any input vector
(Input Vector)It is rotated along unit circle, until input vectorIt is zero, obtains rotation angle (Output
Phase) and the mould of input vector (Output Mag), to realize the conversion of polar coordinates to rectangular co-ordinate.
Sound path between pixel and each array element and at least one second array element of at least one the first array element can be with
It is acquired in conversion of the polar coordinates to rectangular co-ordinate of CORDIC.
Such as, the input coordinate X of array element i and pixel P (x, z), CORDIC are the abscissa of array element i and pixel P (x, z)
The difference of x, ordinate Y are the difference of the ordinate z of array element i and pixel P (x, z), and the postrotational amplitude result of coordinate is exactly pixel
Unidirectional sound path between point P (x, z) and array element i.The mathematical expression of the process are as follows:
Wherein, abscissa X is the abscissa x of an array element of at least one the first array elementiWith pixel abscissa x it
Difference, ordinate Y are the ordinate z of an array element of at least one the first array elementiAnd the difference of the axial coordinate z of pixel.
Have identical function can equivalent substitution with cordic algorithm.
Fig. 4 is frequency, the amplitude, phase diagram of FIR filter.As shown in figure 4, frequency (Frequency), unit MHz；
Amplitude (Magnitude), unit are dB (decibels)；Phase (Phase), unit are radians (radian).
FIR filter is the filter of 128 ranks, for the pixel value of time-domain signal to be changed into the pixel of frequency-region signal
Value.
As shown in figure 4, the amplitude spectrum of FIR filter has all-pass characteristic, the phase of FIR filter has stringent line
Property, meet Hilbert transformation and requires.
Have identical function can equivalent substitution with Hilber transformation.
Fig. 5 is a kind of structure of the full focus data fast imaging system of ultrasound based on FPGA provided in an embodiment of the present invention
Schematic diagram.As shown in figure 5, the full focus data fast imaging system of ultrasound includes:
Hardware components are that the full focus data fast imaging system of ultrasound based on FPGA, based on FPGA includes FMC acquisition mould
Block 10, memory module 20, computing module 30, retrieval module 40, synthesis module 50 and conversion module 60.
Acquisition module 10 is used to acquire by the first ultrasonic signal of each array element transmitting of at least one the first array element
At least one second ultrasonic signal is reflected at least one the second array element to pixel, and by pixel；Memory module 20 is used
In by acquisition after data be stored in dual port memories；Computing module 30 is used for, according to the first ultrasonic signal and at least one
Second ultrasonic signal determines at least one corresponding propagation delay time of pixel；Retrieval module 40 is used for, extremely according to pixel
The corresponding data of pixel of few propagation delay time retrieval storage；Synthesis module 50 is used for, according to the corresponding data of pixel
The corresponding pixel value of synthesized image vegetarian refreshments；Conversion module 60 is used for, and is converted to the corresponding pixel value of pixel, is determined pixel value
Analytic signal.
Specifically, in the data storage and dual port memories after memory module 20 is used to acquire.
Dual port memories are port A configured with configurable first port and second port, first port, and second port is
Port B.
When FMC slow down and retrieve concurrently execute when, the port that port A is inputted as data, port B is defeated as data retrieval
Out；When FMC acquisition finishes, when only carrying out retrieval, port A and port B can be mutually as search and output, port A and port B
It is independent, the calculating of 2N propagation delay time of 2 pixels can be supported simultaneously, and then picture is retrieved by port A and port B respectively
The corresponding data of vegetarian refreshments synthesize the corresponding pixel value of 2 pixels.
When for a pixel, it is based on FPGA, using N number of CORDIC computing module, it can be achieved that the meter of N number of propagation delay time
It calculates；For 2 pixels, by N number of CORDIC computing module composition at a time delay module, each pixel is according to delay mould
Block obtains corresponding N number of transmission delay, and then can obtain the calculating of 2N transmission delay of 2 pixels, and then synthesizes 2 pixels
The corresponding pixel value of point.
Computing module 30 calculates each battle array of at least one corresponding the first array element of pixel by Real-time Delay algorithm
Sound path between member and at least one second array element；According to sound path and the velocity of sound, at least one propagation delay time is determined.
The embodiment of the present invention by FPGA use N number of CORDIC computing module, it can be achieved that N number of propagation delay time concurrently based on
It calculates, meets the quick calculating requirement of the corresponding pixel value of pixel.
Synthesis module 50 is converted according to Hilbert transform pairs pixel value, and pixel value is become frequency domain letter from time-domain signal
Number；Analytic signal is synthesized according to the pixel value of the pixel value of time-domain signal and frequency-region signal.
Since TFM imaging can be related to many pixel values, generally first pixel value can be stored.Ultrasound based on FPGA is complete
Data fast imaging system further includes memory module 70, and memory module 70 can be common caching for storing pixel value
Or memory, it does not limit, does not identify in figure 5 herein.
Software section, the full focus data fast imaging system of ultrasound based on FPGA includes system control module 80 and image
Image-forming module 90, for sequentially exciting array element, image image-forming module 90 is used for based on acquisition analytic signal system control module 80
Envelope be uploaded to host computer imaging.
Fig. 6 is a specific image based on the full focus data fast imaging method of ultrasound in Fig. 1 based on FPGA.
As shown in fig. 6, the embodiment of the present invention uses 16/64 (pulse recurrence frequency/array number), by TFM real time imagery, rail is realized
The quick detection of base material crackle and defect.
1mm horizontal perforation is common artificial defect in bloom, and the embodiment of the present invention is in order to verify TFM computational efficiency and imaging
Effect has carried out TFM detection imaging to the 1mm horizontal perforation of 20mm depth, imaging rate up to 312.5Hz, meet TFM it is quick,
Real-time Imaging.
The present invention effectively improves the efficiency of ultrasonic full data imaging, and imaging rate is that (pulse repeats frequency to PRF/N
Rate/array number), reach the physics limit of TFM detection；Pixel resolution increase when, efficiency reduce the case where be significantly improved.
Under 100*100 and 200*200 pixel resolution, the embodiment of the present invention is reachable using 16 array element TFM imaging rates
312.5Hz, efficiency do not increase with the number of pixel and are reduced；When 32 array element, TFM imaging rate up to 156.25Hz,
When array number increases, meet the physical limit of PRF/N.It can be detected for the high speed TFM using mechanical scanning as representative and fast short-term training is provided
Image space case.
The present invention is calculated using in FPGA piece, is participated in without CPU, and the parallelization degree of TFM imaging is effectively increased, and is accelerated
Imaging efficiency, while significantly improving array number and pixel number increases influence to efficiency.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not intended to limit the present invention the foregoing is merely a specific embodiment of the invention
Protection scope, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (7)
1. a kind of full focus data fast imaging method of ultrasound based on FPGA characterized by comprising
The first ultrasonic signal that acquisition is emitted by each array element of at least one the first array element is to pixel, and by the picture
Vegetarian refreshments reflects at least one second ultrasonic signal at least one the second array element, and the data after acquisition are stored；Wherein, will
Corresponding first ultrasonic signal of pixel and at least one described second ultrasonic signal are stored in dual-port
It deposits；When acquisition and retrieval concurrently execute, by the first end configured with first port and the dual port memories of second port
Port of the mouth as data input, the second port of the dual port memories configured with first port and second port is as number
According to search and output；After acquisition, when retrieval executes, by the dual port memories configured with first port and second port
The second port of the first port and the dual port memories configured with first port and second port is respectively as retrieval
Output；
By field programmable gate array (FPGA) chip to first ultrasonic signal and it is described at least one second
Ultrasonic signal concurrent obtains at least one corresponding propagation delay time of the pixel；
According to the corresponding data of the pixel of at least one propagation delay time of pixel retrieval storage；
According to the corresponding pixel value of pixel described in the corresponding Data Synthesis of the pixel；
The corresponding pixel value of the pixel is converted, determines the analytic signal of the pixel value, and obtaining the solution
It is imaged after analysing the envelope of signal.
2. imaging method according to claim 1, which is characterized in that described to pass through field programmable gate array
(FPGA) chip obtains the picture to first ultrasonic signal and at least one described second ultrasonic signal concurrent
The step of vegetarian refreshments corresponding at least one propagation delay time, specifically includes:
Each array element and the institute of corresponding at least one first array element of the pixel are calculated by Real-time Delay algorithm
State the sound path between at least one second array element；
According to the sound path and the velocity of sound, at least one described propagation delay time is determined.
3. imaging method according to claim 2, which is characterized in that described to calculate the pixel by Real-time Delay algorithm
The step of sound path between each array element and at least one described second array element of corresponding at least one first array element of point
Suddenly are as follows:
Wherein, xiFor the abscissa of i-th of array element of at least one the first array element, xjIt is j-th gust of at least one the second array element
The abscissa of member, (x, z) are pixel coordinate.
4. imaging method according to claim 1, which is characterized in that described to be carried out to the corresponding pixel value of the pixel
The step of converting, determining the analytic signal of the pixel value specifically includes:
The transformation of the pixel value according to Hilbert transform pairs, becomes frequency-region signal from time-domain signal for the pixel value；
Analytic signal is synthesized according to the pixel value of the pixel value of the time-domain signal and the frequency-region signal.
5. imaging method according to claim 4, which is characterized in that described and after the envelope for obtaining the analytic signal
The step of imaging, specifically includes:
The envelope of the analytic signal is uploaded to host computer to be imaged.
6. imaging method according to claim 4, which is characterized in that the Hilbert transform can be with are as follows: FIR filtering
Device.
7. a kind of full focus data fast imaging system of ultrasound based on FPGA characterized by comprising acquisition module (10),
Memory module (20), computing module (30), retrieval module (40), synthesis module (50) and conversion module (60)；
The acquisition module (10) is used for, and acquisition is believed by the first ultrasonic wave that each array element of at least one the first array element emits
Number at least one second ultrasonic signal is reflected at least one the second array element to pixel, and by the pixel；
The memory module (20) is used for, and the data after acquisition are stored；Wherein, described the first surpass the pixel is corresponding
Acoustic signals and at least one described second ultrasonic signal are stored in dual port memories；When acquisition and retrieval concurrently execute,
The port inputted by the first port configured with first port and the dual port memories of second port as data, is configured with
Search and output of the second port of the dual port memories of first port and second port as data；After acquisition,
When retrieval executes, by the first port configured with first port and the dual port memories of second port and it is configured with first end
The second port of the dual port memories of mouth and second port is respectively as search and output；
The computing module (30) is used for, and is believed by field programmable gate array (FPGA) chip first ultrasonic wave
Number and at least one described second ultrasonic signal concurrent, obtain at least one corresponding propagation delay time of the pixel；
The retrieval module (40) is used for, according to the pixel of at least one propagation delay time of pixel retrieval storage
Corresponding data；
The synthesis module (50) is used for, according to the corresponding pixel value of pixel described in the corresponding Data Synthesis of the pixel；
The conversion module (60) is used for, and converts to the corresponding pixel value of the pixel, determines the solution of the pixel value
Signal is analysed, and is imaged after the envelope for obtaining the analytic signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611077700.6A CN106651740B (en) | 2016-11-29 | 2016-11-29 | A kind of full focus data fast imaging method of ultrasound based on FPGA and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611077700.6A CN106651740B (en) | 2016-11-29 | 2016-11-29 | A kind of full focus data fast imaging method of ultrasound based on FPGA and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106651740A CN106651740A (en) | 2017-05-10 |
| CN106651740B true CN106651740B (en) | 2019-09-27 |
Family
ID=58814108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611077700.6A Expired - Fee Related CN106651740B (en) | 2016-11-29 | 2016-11-29 | A kind of full focus data fast imaging method of ultrasound based on FPGA and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106651740B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109975814B (en) * | 2017-12-28 | 2020-09-22 | 深圳先进技术研究院 | Ultrasound imaging method, system and device |
| CN111812205B (en) * | 2020-06-18 | 2023-05-26 | 广州多浦乐电子科技股份有限公司 | Full-focusing and phased array double-scanning imaging method |
| CN113156431B (en) * | 2021-04-30 | 2022-04-26 | 珠海微度芯创科技有限责任公司 | FPGA-based back projection rapid imaging architecture design method |
| CN117405780A (en) * | 2023-12-15 | 2024-01-16 | 广州多浦乐电子科技股份有限公司 | Dual-pixel ultrasonic full focusing system based on FPGA and implementation method |
| CN117710184B (en) * | 2024-01-04 | 2024-08-23 | 浙江大学 | Ultrasonic full-focusing wave number domain imaging method, system and device based on FPGA and GPU |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783138A (en) * | 2009-01-17 | 2010-07-21 | 华为技术有限公司 | Image-scaling method and device thereof |
| CN103908304A (en) * | 2014-03-14 | 2014-07-09 | 中瑞科技（常州）有限公司 | Ultrasonic elastography system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050148872A1 (en) * | 2003-12-23 | 2005-07-07 | Richard William D. | Apparatus and method for synthetic focus ultrasonic imaging |
-
2016
- 2016-11-29 CN CN201611077700.6A patent/CN106651740B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783138A (en) * | 2009-01-17 | 2010-07-21 | 华为技术有限公司 | Image-scaling method and device thereof |
| CN103908304A (en) * | 2014-03-14 | 2014-07-09 | 中瑞科技（常州）有限公司 | Ultrasonic elastography system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106651740A (en) | 2017-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651740B (en) | A kind of full focus data fast imaging method of ultrasound based on FPGA and system | |
| CN106596736B (en) | A kind of real-time ultrasound phased array total focus imaging method | |
| CN106680825B (en) | A kind of acoustic array imaging system and method | |
| CN105997148B (en) | Pulse Doppler ultrahigh spectral resolution imaging processing method and processing system | |
| CN106501367B (en) | Implementation method is imaged in phased array supersonic based on elliptic arc scan transformation | |
| WO2019238850A1 (en) | Method and apparatus for ultrasound imaging with improved beamforming | |
| Njiki et al. | A multi-FPGA architecture-based real-time TFM ultrasound imaging | |
| CN108490077A (en) | A kind of real-time 3D total focus phased array supersonic imaging method, device and storage medium | |
| KR101911734B1 (en) | Software-based ultrasound imaging system | |
| Plotnick et al. | Fast nearfield to farfield conversion algorithm for circular synthetic aperture sonar | |
| CN108693253A (en) | A kind of rapid phase-control battle array ultrasound total focus imaging technique | |
| JP2014512243A (en) | Harmonic ultrasound image processing by synthetic aperture sequential beamforming | |
| Jia et al. | Rigid and elastic acoustic scattering signal separation for underwater target | |
| US11719613B2 (en) | Method and device for quantifying viscoelasticity of a medium | |
| US20210272339A1 (en) | Systems and Methods for Generating and Estimating Unknown and Unacquired Ultrasound Data | |
| JP2006508729A (en) | High frame rate 3D ultrasound imager | |
| JPH06254092A (en) | Ultrasonic signal processor | |
| Vogel et al. | Efficient parallel beamforming for 3D ultrasound imaging | |
| Morse et al. | Backscattering of transients by tilted truncated cylindrical shells: Time-frequency identification of ray contributions from measurements | |
| CN105919624B (en) | A kind of high impulse repeats the coding and device of scanning frequency signal | |
| CN106108942B (en) | Parallel ARFI imaging method based on OpenCL | |
| WO2021135629A1 (en) | Beam forming processing method and apparatus, computer device, and storage medium | |
| US20050148872A1 (en) | Apparatus and method for synthetic focus ultrasonic imaging | |
| Romero-Laorden et al. | Strategies for hardware reduction on the design of portable ultrasound imaging systems | |
| Dziewierz et al. | A design methodology for 2D sparse NDE arrays using an efficient implementation of refracted-ray TFM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190927 Termination date: 20211129 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |