Method for processing signals from an ultrasound probe acquisition, corresponding computer program and ultrasound probe device

Abstract

A method including control of M emission transducers for L successive ultrasound plane wave emissions having L different emission angles, control of N reception transducers for simultaneously receiving N measurement time signals for each emission and reconstitution of an imaged zone by calculating, at each point, a value resulting from a processing of the measurement time signals received. The reconstitution of the imaged zone includes calculating LN flight times, LL, each flight time t.sub.l,n being the time taken for the l-th plane wave, the emission zone of which includes the point considered, where 1lL, to be received by the n-th reception transducer, where 1nN, passing through the point considered according to a predetermined propagation mode, and coherent summing LN instantaneous values taken, respectively, by the LN measurement time signals received corresponding to the L emissions, at the LN flight times.

Claims

1. A method for processing signals from an ultrasound probe acquisition, the method comprising: controlling, via processing circuitry, of an array of M emission transducers for L successive emissions of ultrasound plane waves having L different successive emission angles in L emission zones; controlling, via the processing circuitry, of an array of N reception transducers so as to simultaneously receive for a predetermined duration, for each emission, N measurement time signals, measuring echoes due to reflections of the emission considered; reconstituting, via the processing circuitry, of an imaged zone by calculating, at each point of a plurality of predetermined points of the imaged zone, a value resulting from a processing of at least some of LN measurement time signals received, wherein the reconstituting of the imaged zone includes the following, performed by the processing circuitry for each point of the imaged zone: determining of L emissions, LL, among the L successive emissions, of which the emission zones include the point considered, calculating of LN flight times, each flight time t.sub.l,n being time taken for an l-th plane wave, the emission zone of which includes the point considered, where 1lL, to be received by an n-th reception transducer, where 1nN, passing through the point considered according to a predetermined propagation mode, coherent summing of LN instantaneous values taken respectively, by the LN measurement time signals received corresponding to the L emissions determined, at the LN flight times calculated; and generating and displaying the imaged zone on a display.

2. The method for processing signals according to claim 1, wherein the emission transducers are controlled by a delay law defined for each of the successive L emissions of ultrasound plane waves, each delay law enabling generation of an ultrasound plane wave at a desired emission angle among the L different successive emission angles.

3. The method for processing signals according to claim 1, wherein M=N and the transducers are sequentially emitters and receivers.

4. The method for processing signals according to claim 1, wherein the L different successive emission angles are defined around a mean direction ((L+1)/2) not perpendicular to the emission transducer array.

5. The method for processing signals according to claim 1, wherein, for each emission angle, the predetermined propagation mode is chosen from one of the following modes: a direct-path propagation mode, with or without Longitudina.Math.Transversal mode conversion, according to which the plane wave emitted is received directly by each point of the imaged zone and directly returned to the reception transducers without any other reflection, a corner echo propagation mode, with or without Longitudina.Math.Transversal mode conversions, according to which the wave emitted undergoes a reflection on a predetermined surface of the imaged zone, either between the emission transducers and each point of the imaged zone or between each point of the imaged zone and the reception transducers, and an indirect-path propagation mode, with or without Longitudinal.Math.Transversal mode conversion, according to which the wave emitted undergoes at least two reflections against at least one predetermined surface of the imaged zone, at least once between the emission transducers and each point of the imaged zone and one other time between each point of the imaged zone and the reception transducers.

6. The method for processing signals according to claim 1, wherein the imaged zone is included in a union of the L emission zones, the reconstitution of the imaged zone including a following: resetting of each point of the imaged zone at a zero value, for any value of an index I ranging from 1 to L and for each point located in the intersection of the imaged zone and an I-th emission zone: calculating of the N flight times t.sub.l,n, coherent summing of the N instantaneous values taken, respectively, by the N measurement time signals received in response to the I-th emission, at the N calculated flight times, adding of the result of said coherent summing to the value of the point considered, and calculating of a modulus of a value finally obtained at each point of the imaged zone.

7. The method for processing signals according to claim 1, wherein, for each emission, an apodization of the M ultrasound signals emitted by the M emission transducers in order to form an ultrasound plane wave is performed by an apodization window such as a trapezoidal, Hamming or Blackman-Harris amplitude law.

8. The method for processing signals according to claim 1, wherein the imaged zone takes the form of a sectorial zone delimited by the ends of the emission zones of maximum and minimum angles.

9. A non-transitory computer-readable storage medium including computer executable instructions, wherein the instructions, when executed by a computer, cause the computer to perform: controlling of an array of M emission transducers for L successive emissions of ultrasound plane waves having L different successive emission angles in L emission zones; controlling of an array of N reception transducers so as to simultaneously receive for a predetermined duration, for each emission, N measurement time signals, measuring echoes due to reflections of the emission considered; reconstituting of an imaged zone by calculating, at each point of a plurality of predetermined points of the imaged zone, a value resulting from a processing of at least some of LN measurement time signals received, wherein the reconstituting of the imaged zone includes the following, performed by a processor for each point of the imaged zone: determining of L emissions, LL, among the L successive emissions, of which the emission zones include the point considered, calculating of LN flight times, each flight time t.sub.l,n being time taken for an l-th plane wave, the emission zone of which includes the point considered, where 1lL, to be received by an n-th reception transducer, where 1nN, passing through the point considered according to a predetermined propagation mode, and coherent summing of LN instantaneous values taken respectively, by the LN measurement time signals received corresponding to the L emissions determined, at the LN flight times calculated; and generating and displaying the imaged zone on a display.

10. An ultrasound probe device comprising: a probe including M ultrasound emission transducers and N ultrasound reception transducers; and processing circuitry configured to: control the M emission transducers for L successive emissions of ultrasound plane waves having L different successive emission angles in L emission zones; control the N reception transducers so as to simultaneously receive, for a predetermined period, for each emission, N measurement time signals, measuring echoes due to reflections of the emission considered; reconstitute an imaged zone by calculating, at each point of a plurality of predetermined points of the imaged zone, a value resulting from a processing of at least some of LN measurement time signals received, the processing circuitry being further configured to perform the following processing operations for each point of the imaged zone: determination of L emissions, LL, among the L successive emissions, of which the emission zones include the point considered, calculation of LN flight times, each flight time t.sub.l,n being time taken for an I-th plane wave, the emission zone of which includes the point considered, where 1l L, to be received by an n-th reception transducer, where 1nN, passing through the point considered according to a predetermined propagation mode, and coherent summing of LN instantaneous values taken, respectively, by the LN measurement time signals received corresponding to the L emissions determined, at the LN flight times calculated; and the processing circuitry is further configured to generate and display the imaged zone on a display.

Description

(1) The invention will be easier to understand in view of the following description provided solely as an example, and with reference to the appended drawings, wherein:

(2) FIG. 1 schematically shows the general structure of an ultrasound probe device according to an embodiment of the invention,

(3) FIG. 2 shows a principle of successive emissions of ultrasound plane waves implemented by the device of FIG. 1,

(4) FIGS. 3, 4 and 5 geometrically illustrate flight time calculations performed in the reconstitution of an imaged zone by the device of FIG. 1 when it implements the principle of FIG. 2,

(5) FIG. 6 shows the successive steps of a method for acquisition and processing of ultrasound signals implemented by the device of FIG. 1, according to an embodiment of the invention,

(6) FIG. 7 shows a first possible application of the method of FIG. 6,

(7) FIG. 8 shows a second possible application of the method of FIG. 6, and

(8) FIGS. 9 and 10 show different possible results of the application of FIG. 8.

(9) In reference to FIG. 1, a probe device 100 of an object 102 according to an embodiment of the invention has an ultrasound probe 104 having a housing 106, i.e. a non-deformable structural element that serves as a reference attached to the probe 104, wherein an array of N fixed or mobile transducers 108.sub.1, . . . , 108.sub.N are arranged, for example linearly or matricially.

(10) The object 102 is, for example, a mechanical part to be examined by non-destructive testing or, in a medical context, a human or animal body part to be examined non-invasively. In the embodiment of FIG. 1, the object 102 is immersed in a liquid, such as water 110, and the probe 104 is held at a distance from the object 102 so that the water 110 separates them. However, in another equivalent embodiment, the probe 104 may be in direct contact with the object 102.

(11) The transducers 108.sub.1, . . . , 108.sub.N are designed so as to individually emit ultrasound waves toward the object 102 in response to control signals identified under general reference C, according to main directions parallel to one another, indicated by dotted-line arrows in FIG. 1, and in the main plane of the figure.

(12) The transducers 108.sub.1, . . . , 108.sub.N are further designed to detect echoes of ultrasound waves reflected on and in the object 102 and to provide measurement signals identified by general reference S and corresponding to said echoes. Thus, in the non-limiting example of FIG. 1, the transducers 108.sub.1, . . . , 108.sub.N satisfy both the functions of emission and reception, but receivers different from the emitters may also be provided in different independent housings while remaining consistent with the principles of the invention.

(13) The probe device 100 further has an electronic circuit 112 for controlling the transducers 108.sub.1, . . . , 108.sub.N of the probe 104 and for processing the measurement signals S. This electronic circuit 112 is connected to the probe 104 so as to transmit the control signals C thereto and so as to receive the measurement signals S. The electronic circuit 112 is, for example, that of a computer. It has a central processing unit 114, such as a microprocessor designed to emit, to the probe 104, the control signals C and to receive, from the probe 104, the measurement signals S, and a memory 116 wherein in particular a computer program 118 is stored.

(14) The computer program 118 first has instructions 120 for generating the signals C for controlling the transducers 108.sub.1, . . . , 108.sub.N and receiving their echoes. These instructions are more specifically programmed so as to: activate the transducers 108.sub.1, . . . , 108.sub.N as emitters for L successive emissions of ultrasound plane waves having L different successive emission angles in L emission zones of the object 102, activate the transducers 108.sub.1, . . . , 108.sub.N as receivers so as, after each emission, to simultaneously receive, by said N receivers and for a predetermined duration, of the desired inspection depth, N measurement time signals, measuring in particular echoes due to reflections of each emission considered.

(15) The ultrasound plane waves are obtained upon emission by applying, to the transducers 108.sub.1, . . . , 108.sub.N delay laws recorded in the memory 116 in a delay law base 122. Each delay law defines delays to be applied to the transducers 108.sub.1, . . . , 108.sub.N in emission, so as to generate an ultrasound plane wave at a desired angle of emission among the L different successive emission angles. Therefore, there are as many delay laws as there are desired successive emissions.

(16) Upon reception, the set S of the LN measurement time signals received by the transducers 108.sub.1, . . . , 108.sub.N is returned by the probe 104 to the central processing unit 114.

(17) The computer program 118 further comprises instructions 124 for recording said signals, wherein K.sub.l,n(t) represents the measurement time signal received by the transducer 108.sub.n in response to the l-th ultrasound plane-wave emission.

(18) The computer program 118 further comprises instructions 126 for reconstituting an imaged zone by calculating, at each point of a plurality of predetermined points of said imaged zone, a value resulting from a processing of at least some of the LN measurement time signals received. More specifically, as the imaged zone is, for example, defined as being a digital image consisting of pixels, the instructions 126 are defined for, at each pixel of said image: determining L emissions, LL, among the L successive emissions, of which the emission zones include the pixel considered, calculating LN flight times, each flight time t.sub.l,n being the time taken for the l-th plane wave of which the emission zone includes the pixel considered, where 1lL, to be received by the n-th reception transducer, where 1nN, passing through the pixel considered according to a predetermined propagation mode, coherently summing the LN instantaneous values taken, respectively, by the LN measurement time signals received corresponding to the L emissions determined, at the LN flight times calculated, and calculating the modulus of the value obtained optionally weighted by the value L.

(19) Finally, the computer program 118 comprises instructions 128 for displaying the digital image obtained on a display device, not shown.

(20) As illustrated in FIG. 2, in a case where the number L of successive emissions is odd and where the angles of emission are followed with a constant step in a symmetrical angular sector with respect to the direction z orthogonal to the transducer array 108.sub.1, . . . , 108.sub.N, the first plane wave emission is associated with a delay law T.sub.1 concerning pulses emitted by the transducers 108.sub.1, . . . , 108.sub.N, enabling the emission of a plane wave having an angle of emission .sub.1 with respect to the direction z in a first emission zone ZE.sub.1 located partially outside the opening of the probe 104. The (L+1)/2-th plane wave emission is associated with a uniform delay law T.sub.(L+1)/2 for the emission of a plane wave having a zero angle of emission with respect to the direction z in a (L+1)/2-th emission zone ZE.sub.(L+1)/2 covering the opening of the probe 104. Finally, the last plane wave emission is associated with a delay law T.sub.L enabling the emission of a plane wave having an angle of emission .sub.L=.sub.1 with respect to the direction z in a last emission zone ZE.sub.L located partially outside of the opening of the probe 104. In general, the l-th plane wave emission is associated with a delay law T.sub.l enabling the emission of a plane wave having an angle of emission .sub.l=.sub.1+(l1).Math.(.sub.L.sub.1)/(L1) with respect to direction z.

(21) To improve the quality of the measurement signals used to reconstruct the imaged zone, it is also possible to apply an apodization of the ultrasound signals emitted by the transducers 108.sub.1, . . . , 108.sub.N so as to form an ultrasound plane wave of higher quality, without distortion due to edge effects. Such an apodization is performed at each emission spatially on all of the transducers by means of an apodization window such as a trapezoidal, Hamming or Blackman-Harris amplitude law. It has the effect of providing a better definition of the successive emission zones.

(22) In consideration of the acquisition technique used, the zone to be imaged must be contained in the union of the L successive emission zones. The result is that said zone may extend beyond the opening of the probe 104 as can be seen in FIG. 2. In particular, the imaged zone may take the form of a sectorial zone delimited by the ends of emission zones of maximum and minimum angles. An S-scan image may thus be obtained.

(23) In reference to FIG. 3 showing a 2D application of the invention, for the l-th ultrasound plane wave emission of emission angle .sub.l, the direct-path flight time t.sub.l,n(P) relative to a point P of coordinates (x, z) in the reference system (O, x, z) associated with the plane of the zone to be imaged where the axis (O, x) is the axis of the transducer array 108.sub.1, . . . , 108.sub.N, and relative to transducer 108.sub.n the coordinates of which are (x.sub.n, 0) in the same reference system (O, x, z), is broken down as follows:
t.sub.l,n(P)=t.sub.l.sup.tr(P)+t.sub.n.sup.re(P)
where t.sub.l.sup.tr(P) is the emission flight time between the plane (O, x) of emission of the plane wave and point P and t.sub.n.sup.re(P) is the reception flight time between the point P and the transducer 108.sub.n.

(24) By a geometric calculation, the emission flight time is expressed as follows:

(25) $t_l^{\mathrm{tr}}\left(P\right)=\frac{x.Math.\sin \left({}_l\right)+z.Math.\cos \left({}_l\right)}{c}$
where c is the speed of propagation of the plane wave in the medium considered (assuming that there is no change in medium, which is verified in cases of non-destructive testing on contact). It is noted that said emission flight time for point P is not dependent upon transducers 108.sub.1, . . . , 108.sub.N, but only upon the emission angle .sub.i.

(26) By a geometric calculation as well, the reception flight time is expressed as follows:

(27) $t_n^{\mathrm{re}}\left(P\right)=\frac{\sqrt{{\left(x_n-x\right)}^2+z^2}}{c}$
where it is noted that said reception flight time for point P is dependent only upon transducers 108.sub.1, . . . , 108.sub.N, but not upon the emission angle .sub.l.

(28) The total flight time is therefore expressed as follows:

(29) $t_{l,n}\left(P\right)=\frac{x.Math.\sin \left({}_l\right)+z.Math.\cos \left({}_l\right)}{c}+\frac{\sqrt{{\left(x_n-x\right)}^2+z^2}}{c}$

(30) To coherently sum LN instantaneous values taken, respectively, by LN measurement time signals received corresponding to L emissions determined contributing to the zone imaged at point P, at the LN flight times as calculated above, it is also appropriate, in practice, for the sake of time recalibration, to apply a constant .sub.l.sup.tr specific to each emission, the value of which is expressed as follows:

(31) ${}_l^{\mathrm{tr}}=\frac{\underset{n}{\max }\left(z_{l,n}^{}\right)}{2.Math.c}$
where z.sub.l,n represents the distance between the transducer 108.sub.n and the axis of the transducer array (O, x) virtually angularly shifted from the axis (O, x) by an angle equal to .sub.l. This distance may be calculated according to the following formula:

(32) $z_{l,n}^{}=x_n.Math.\sin \left({}_l\right)-\underset{n}{\min }\left[x_n.Math.\sin \left({}_l\right)\right]$

(33) The above calculation, and in particular the value min.sub.n[x.sub.n.Math.sin(.sub.l)], ensures that the delays applied in the delay laws always remain positive, even when the angle .sub.l is negative. Moreover, in the case of an inspected part with a planar surface, the constant .sub.l.sup.tr specific to each emission corresponds to the mean delay of the delay law applied to the l-th emission, or, equivalently, to half of the maximum delay.

(34) The result is that the modulus of the coherent summing defined above, involving the LN measurement time signals contributing to the zone imaged at point P, may be expressed as follows:
A(P)=|.sub.l=1.sup.L.sub.n=1.sup.NK.sub.l,n[t.sub.l,n(P)+.sub.l.sup.tr]|

(35) In practice, to obtain the image envelope, it is instead the analytical signals that are summed, in particular by means of Hilbert H.sub.l,n(t) of the signals K.sub.l,n(t). The above calculation then becomes, more specifically:
A.sub.env(P)=|.sub.l=1.sup.L.sub.n=1.sup.N(K.sub.l,n[t.sub.l,n(P)+.sub.l.sup.tr]+H.sub.l,n[t.sub.l,n(P)+.sub.l.sup.tr])|

(36) It is noted that, in accordance with said calculation close to an all-point focusing, no delay law is applied on reception.

(37) The advantage of reconstituting the zone to be imaged by such a technique based on flight time calculations at each point is that it is possible to take into account different configurations and modes of propagation of the ultrasound waves. Thus, for example, in the previous calculations, it was considered, for the sake of simplicity, that the mode of propagation of the ultrasound waves was direct, without changing the propagation medium (probe in contact with the object to be inspected) and without polarization conversion of the waves emitted, i.e. each plane wave emitted is received directly by each point of the imaged zone and returned directly to the transducers 108.sub.1, . . . , 108.sub.N without other reflection.

(38) However, other hypotheses may be made and it would then be sufficient to adapt the flight time calculation: the examination of the object considered may be performed with immersion without contact, with a more or less complex object surface, the mode of propagation of the ultrasound waves may be a corner echo mode, in particular in the vicinity of a crack-type defect and according to a certain plane wave incidence: in this case, the plane wave emitted is subject to a reflection against a predetermined surface of the imaged zone, for example the bottom of the object, either between the transducers 108.sub.1, . . . , 108.sub.N and each point of the imaged zone, or between each point of the imaged zone and the transducers 108.sub.1, . . . , 108.sub.N, the mode of propagation of the ultrasound waves may be an indirect-path mode, in particular also in the vicinity of a crack-type defect and according to a certain plane wave incidence: in this case, the plane wave emitted is subject to at least two reflections against at least one predetermined surface of the imaged zone, for example the bottom of the object, at least once between the transducers 108.sub.1, . . . , 108.sub.N and each point of the imaged zone and another time between each point of the imaged zone and the transducers 108.sub.1, . . . , 108.sub.N, regardless of the propagation mode, the longitudinal or transverse polarization of the ultrasound waves may vary upon a reflection: a transverse wave may become longitudinal and vice versa, said conversion having an impact on the propagation speed.

(39) The above hypotheses may also be combined.

(40) As an example, FIG. 4 shows an examination with immersion without contact according to which the propagation is performed, in direct mode, in a first medium at speed c.sub.1, for example water, then in a second medium at speed c.sub.2, for example the material (steel, . . . ) of the object inspected. The two media are, in this particular case, delimited by the planar surface of the object located at a distance H from the axis (O, x) of the transducer array 108.sub.1, . . . , 108.sub.N.

(41) Using the notations of FIG. 3 and adding those of FIG. 4, the calculation of points of impact x.sub.l.sup.tr(P) and x.sub.n.sup.re(P) on the planar surface of the object should be taken into account in order to calculate the total flight time t.sub.l,n(P).

(42) By emitting a plane wave in the first medium, above the planar surface of the object, with an angle of incidence .sub.l, it is possible to deduce the angle of incidence .sub.l of the plane wave in the second medium under the surface of the object by the Snell-Descartes law, written as follows:

(43) ${}_l={\sin }^{-1}\left(\frac{c_2}{c_1}\sin \left({}_l\right)\right)$

(44) Then, knowing the coordinates x and z of the focusing point P in the object inspected, the abscissa x.sub.l.sup.tr(P) of the point of impact on the surface associated with the emission path is deduced:
x.sub.l.sup.tr(P)=x(zH).Math.tan(.sub.l)

(45) Then, the emission flight time to point P is written:

(46) $t_l^{\mathrm{tr}}\left(P\right)=\frac{\sqrt{H^2+{\left[x_l^{\mathrm{tr}}\left(P\right)\right]}^2}}{c_1}+\frac{\left(x-x_l^{\mathrm{tr}}\left(P\right)\right)\sin {}_l+\left(z-H\right)\cos {}_l}{c_2}$

(47) As above, it is noted that said emission flight time for point P is not dependent upon transducers 108.sub.1, . . . , 108.sub.N, but only the angle of emission .sub.l.

(48) In reception, similarly:

(49) $t_n^{\mathrm{re}}\left(P\right)=\frac{\sqrt{H^2+{\left[x_n^{\mathrm{re}}\left(P\right)\right]}^2}}{c_1}+\frac{\sqrt{{\left(x_n^{\mathrm{re}}\left(P\right)-x\right)}^2+{\left(z-H\right)}^2}}{c_2}$
where the abscissa x.sub.n.sup.re(P) of the point of impact in reception is determined on the basis of Fermat's principle according to which the return path between point P and the receiver 108.sub.n must correspond to the shortest path. In the case of a planar part, the principle involves the search for zeroes of a function. The methods generally used to solve such a zero search are diverse: The Newton-Raphson method, the Ferrari method, the Laguerre method, the gradient descent method, and so on. The Newton-Raphson and gradient descent methods are the more beneficial because they remain valid for complex surface geometries. There are in particular numerous prior art documents on the search for shorter paths and on the calculation of points of impact. These methods therefore will not be mentioned.

(50) As above, it is demonstrated that the reception flight time for point P is dependent only on transducers 108.sub.1, . . . , 108.sub.N, but not on the angle of incidence .sub.l.

(51) As an additional example, FIG. 5 shows an examination with immersion without contact according to which the propagation is performed, in corner echo mode, in a first medium at speed c.sub.1, for example water, then in a second medium, for example the material (steel, etc.) of the object inspected. In the second medium, the propagation is performed at speed c.sub.2 for the longitudinal waves and at speed c.sub.3 for transverse waves. The two media are, in this particular case, delimited by the planar surface of the object located at a distance H.sub.i from the axis (O, x) of the transducer array 108.sub.1, . . . , 108.sub.N. The object also has a planar bottom, at a distance H.sub.r from the axis (O, x), against which a planar incident wave is reflected according to the corner echo propagation principle. Only the incident path at point P is shown, the return path being similar to that of the previous example.

(52) The incident path of the l-th planar wave emission is thus broken down into three parts: a first part T.sub.a between its theoretical emission point E of coordinates (x.sub.1, z.sub.1) on the axis (O, x) and a point of impact Ip of coordinates (x.sub.i, z.sub.i=H.sub.i) at the interface of the two media, oriented according to an angle l with respect to the direction (O, z), a second part T.sub.b between the point of impact Ip and a reflection point R of coordinates (x.sub.r, z.sub.r=H.sub.r) at the bottom of the object, oriented according to angle .sub.l with respect to the direction (O, z), and a third part T.sub.c between the point of reflection R and the point P of coordinates denoted (x.sub.2, z.sub.2), oriented according to an angle .sub.l with respect to the direction (O, z).

(53) It is also assumed that the wave is longitudinal on paths T.sub.a and T.sub.b, then transversal on the path T.sub.c, a polarization conversion occurring upon the reflection against the bottom of the object.

(54) In accordance with the Snell-Descartes law, the refraction principle must be verified at point Ip and the reflection principle must be verified at point R. This produces the following system of equations:

(55) $\{\begin{array}{c}\frac{\sin \left({}_l\right)}{c_1}=\frac{\sin \left({}_l\right)}{c_2}\\ \frac{\sin \left({}_l\right)}{c_2}=\frac{\sin \left({}_l\right)}{c_3}\end{array}$

(56) To pose the problem in Cartesian coordinates, the sines are expressed as a function of the coordinates of points E, Ip, R and P:

(57) 0$\sin \left({}_l\right)=\frac{\left(x_i-x_1\right)}{\sqrt{{\left(x_i-x_1\right)}^2+{\left(H_i-z_1\right)}^2}}$$\sin \left({}_l\right)=\frac{\left(x_r-x_i\right)}{\sqrt{{\left(x_r-x_i\right)}^2+{\left(H_r-H_i\right)}^2}}$$\sin \left({}_1\right)=\frac{\left(x_2-x_r\right)}{\sqrt{{\left(x_2-x_r\right)}^2+{\left(z_2-H_r\right)}^2}}$

(58) The system of equations above may then be expressed as follows:

(59) $\{\begin{array}{c}{f}_1\text{:}{c}_2\left(x_i-x_1\right)\sqrt{{\left(x_r-x_i\right)}^2+{\left(H_r-H_i\right)}^2}-\\ c_1\left(x_r-x_i\right)\sqrt{{\left(x_i-x_1\right)}^2+{\left(H_i-z_1\right)}^2}=0\\ f_2\text{:}{c}_3\left(x_r-x_i\right)\sqrt{{\left(x_2-x_r\right)}^2+{\left(z_2-H_r\right)}^2}-\\ c_2\left(x_2-x_r\right)\sqrt{{\left(x_r-x_i\right)}^2+{\left(H_r-H_i\right)}^2}=0\end{array}$

(60) This system of two nonlinear equations with two unknowns, x.sub.i and x.sub.r, is classically solved by means of the Newton-Raphson method. It makes it possible to determine points Ip and R, then to deduce the emission flight time t.sub.l.sup.tr(P).

(61) In reference to FIG. 6, an example of an ultrasound signal acquisition and processing method 600 implemented by the device 100 of FIG. 1 will now be described according to a preferred embodiment of the invention.

(62) In a step 602, the processing unit 114 carrying out the instructions 120 orders the sequences of emissions and receptions of transducers 108.sub.1, . . . , 108.sub.N for the acquisition of measurement signals K.sub.l,n(t).

(63) There are L of these sequences, L being an integer number capable of being much lower than the number N of transducers 108.sub.1, . . . , 108.sub.N. After each firing, the signals are received on the set of N transducers, digitized and transmitted to the electronic circuit 112.

(64) In a step 604, the processing unit 114 carrying out the instructions 124 records the measurement signals K.sub.l,n(t), said signals being digitized so as to facilitate their subsequent processing. Steps 602 and 604 may be carried out simultaneously, i.e. it is unnecessary to wait for all of the firings to be performed in order to begin to record the measurement signals and reconstitute an image.

(65) In a next step 606, the processing unit 114 carrying out the instructions 126 resets each pixel of the zone to be imaged, chosen in the union of the L emission zones, at a zero value. Moreover, an index I intended to vary from 1 to L is reset to 1. This step may be carried out independently of steps 602 and 604, before, during or after.

(66) In the next step 608, for each pixel of the zone to be imaged located in the l-th emission zone, the N flight times t.sub.l,n 1nN, are calculated according to a propagation mode chosen specifically for the angle of emission .sub.l, according, for example, to one of the calculations presented above. It therefore appears that multiple propagation modes with or without polarization conversions may respectively be chosen for the L successive emissions. The invention therefore makes it possible to fuse multiple ultrasound reconstruction modes in a single zone to be imaged.

(67) In a step 610, for each pixel of the zone to be imaged located in the l-th emission zone, the N instantaneous values taken, respectively, by the N measurement time signals received in response to the l-th emission, at the N flight times calculated above, are summed in accordance with the following coherent summing operation:
A.sub.l(P)=.sub.n=1.sup.N(K.sub.l,n[t.sub.l,n(P)+.sub.l.sup.tr]+H.sub.l,n[t.sub.l,n(P)+.sub.l.sup.tr])
where A.sub.l(P) is the amplitude of the pixel P for the l-th emission.

(68) In a step 612, for each pixel of the zone to be imaged located in the l-th emission zone, the result of the coherent summing is added to the current value of the pixel considered and the index I is incremented by one unit.

(69) Then, in a test step 614, if l is strictly lower than L, the method returns to step 608. Otherwise, it goes to a final step 616.

(70) In the final step 616, the modulus of the value finally obtained at each point of the imaged zone is calculated, so that the value A.sub.env(P) defined above is obtained at each pixel P:
A.sub.env(P)=|.sub.l=1.sup.LA.sub.l(P)|

(71) A weighting of the pixel values by the number of firings having contributed to the value of each of them may optionally be performed, with the understanding that the pixels close to the mean angle of the emissions receive more ultrasound waves than those farther away.

(72) Each loop of steps 608, 610 and 612 of the iterations on the index I may be carried out in parallel with steps 602 and 604 since the processing performed in each of said loops is dependent only upon the results of a single ultrasound firing. It is in particular unnecessary to wait for all of the firings to be performed in order to begin the calculations of the iterative process 608-610-612-614. Steps 608, 610, 612, 614 and 616 are, moreover, like step 606, carried out by the processing unit 114 by means of instructions 126.

(73) In the last step 616 also, the processing unit 114 carrying out instructions 128 displays the resulting image.

(74) Owing to the implementation of this preferred embodiment, the imaged zone may be progressively reconstructed, angle by angle, updating for each firing angle only the values of pixel located in the firing zone, the contours of said zone being better defined as an effective apodization has been performed on the emission of the ultrasound plane wave. It is therefore unnecessary to perform calculations for all of the pixels of the final image on each firing. This principle is particularly advantageous when the final image extends largely beyond the dimensions of the sensor.

(75) As shown in FIG. 7, one of the advantages of the invention is also that the L different successive emission angles .sub.1 to .sub.L may be defined around a mean direction .sub.(L+1)/2 not perpendicular to the transducer array 108.sub.1, . . . , 108.sub.N. In particular, when it involves detecting defects such as a crack F at the bottom of an object to be inspected in a non-destructive testing, said crack F also being perpendicular to the transducer array, it is preferable to laterally offset the zone to be inspected with respect to the probe 104 and to emit around a mean of 45 for example. Thus, the crack becomes more visible. It is also advantageously detected over its entire length by choosing a corner echo propagation mode in its vicinity with a possible polarization conversion, which is also possible in the implementation of the invention as seen above. The zone to be inspected may even be offset so as to completely leave the opening of the probe 104, which is not possible in conventional beamforming methods.

(76) Another example of a concrete application of the method described above is shown in FIG. 8. This application was the subject of real tests, the various results of which will be discussed in reference to FIGS. 9 and 10.

(77) The inspected part 800 is made of stainless steel and has three artificial notch-type defects. Two notches, h.sub.1 and h.sub.3, are oriented perpendicularly to the transducer array 108.sub.1, . . . , 108.sub.n and located at the bottom of the inspected part 800 outside of the opening of the sensor formed by the set of transducers 108.sub.1, . . . , 108.sub.n. A third notch, h.sub.2, has an orientation parallel to the transducer array 108.sub.1, . . . , 108.sub.n and is located in the vicinity of the normal to the sensor that they form. The depth of the part is H=70 mm, the number N of transducers is equal to 64 (inter-transducer step of 0.6 mm, central frequency of the ultrasound waves emitted of 5 MHz), the inspected part is in contact with the transducers centered on the notch h.sub.2. The three notches have a length of 10 mm.

(78) Thirty-one successive ultrasound plane wave emissions with emission angles ranging from 60 to +60 (with respect to the normal of the sensor) per 4 step are carried out. Among these 31 successive emissions, the following are distinguished: a first group of emissions oriented around 45, more specifically between 60 and 30, this first group covering an angular sector wherein the notch h.sub.1 is located; a second group of emissions oriented around the normal (0), more generally between 30 and +30, this second group covering an angular sector wherein the notch h.sub.2 is located; and a third group of emissions oriented around +45, more specifically between +30 and +60, this third group covering an angular sector wherein the notch h.sub.3 is located. The imaged zone is included in the union of the emission zones and is identified by reference Z.

(79) According to a first possible reconstitution of the imaged zone Z, the same direct-path and longitudinal wave propagation mode is applied to the processing of measurement signals of all ultrasound firings. This mode is symbolized by the notation L.sup.dL, where L indicates a rectilinear sub-path in longitudinal polarization and .sup.d indicates a plane wave interaction with a defect. More generally, the direct-path propagation mode is symbolized by the notation X.sup.dX, where X may take the value L or T (for a rectilinear sub-path in transversal polarization) on each sub-path. The reconstitution with the single mode L.sup.dL provides the result of FIG. 9(a). The notch h.sub.2 is imaged in its entirety, while the two other notches h.sub.1 and h.sub.3 are characterized each by a single diffraction echo, located at the bottom of each defect. The diffraction echo of the top of the notches is in fact barely visible in the two cases.

(80) According to a second possible reconstitution of the imaged zone Z, the same corner echo and longitudinal wave propagation mode is applied to the processing of measurement signals of all ultrasound firings. This mode is symbolized by the notation L.sup.dL.sup.rL, where .sup.r indicates a plane wave interaction with the bottom of the inspected part. More generally, the corner echo propagation mode is symbolized by the notation X.sup.dX.sup.rX, where X may take the value L or T on each sub-path. The reconstitution with the single mode L.sup.dL.sup.rL provides the result of FIG. 9(b). The notches h.sub.1 and h.sub.3 are this time clearly visible over their entire length, but the notch h.sub.2 is completely masked by a reconstruction artifact. This artifact is explained by the presence of the bottom echo, which is a direct echo. More specifically, the artifact located in the opening of the sensor corresponds to the repositioning of the direct echo coming from the bottom of the part inspected at flight times corresponding to corner echo paths. There is a contradiction between the nature of the echo and the reconstruction mode.

(81) According to a third possible reconstitution of the imaged zone Z, taking advantage of the two previous imperfect reconstitutions: a first mode of propagation L.sup.dL.sup.rL is applied to the processing of the measurement signals from the first group of emissions between 60 and 30, this propagation mode being adapted to the configuration of notch h.sub.1, a second mode of propagation L.sup.dL is applied to the processing of the measurement signals from the second group of emissions between 30 and +30, this propagation mode being adapted to the configuration of notch h.sub.2, and a third mode of propagation L.sup.dL.sup.rL is applied to the processing of the measurement signals from the third group of emissions between +30 and +60, this propagation mode being adapted to the configuration of notch h.sub.3.

(82) This third reconstitution provides the progressive results of FIG. 10. FIG. 10(a) shows the intermediate result of the ultrasound firing n.sup.o1 (60), while the propagation mode chosen is L.sup.dL.sup.rL. FIG. 10(b) shows the intermediate result of the ultrasound firing n.sup.o2 (56), while the propagation mode chosen is still L.sup.dL.sup.rL. FIG. 10(c) shows the intermediate result of the ultrasound firing n.sup.o9 (28), while the propagation mode chosen is L.sup.dL. FIG. 10(d) shows the intermediate result of the ultrasound firing n.sup.o10 (24), while the propagation mode chosen is still L.sup.dL. FIG. 10(e) shows the intermediate result of the ultrasound firing n.sup.o11 (20), while the propagation mode chosen is still L.sup.dL. FIG. 10(f) shows the intermediate result of the ultrasound firing n.sup.o24 (+32), while the propagation mode chosen is again L.sup.dL.sup.rL. FIG. 10(i) shows the intermediate result of the ultrasound firing n.sup.o26 (+40), while the propagation mode chose in still L.sup.dL.sup.rL. Finally, FIG. 10(j) shows the final result of the ultrasound firing n.sup.o31 (+60), while the propagation mode chosen is still L.sup.dL.sup.rL.

(83) The image 10(j) is to be compared with images 9(a) and 9(b). The visibility of the three notches is clearly greater, owing to the possibility offered by a method according to the invention of adapting the propagation mode chosen during reconstitution as a function of the angle of emission of successive ultrasound firings.

(84) It clearly appears that a method and a device as described above make it possible to perform a smaller number of firings than that necessary in the all-point focusing methods for an equivalent image quality finally obtained or in order to achieve better performance in terms of image quality for an equivalent number of firings. The main reason for this improvement, i.e. higher speed or better image quality, is that, on each firing, all of the emission transducers are used.

(85) In addition, the method proposed remains compatible with complex geometries or materials and makes it possible to fuse a plurality of propagation modes in the same image, according to the firing angle. Images extending beyond the opening of the sensor may also be obtained.

(86) In the case of a progressive reconstitution of the image, angle by angle, as is made possible by the invention, the outcome is finally improved.

(87) Finally, experimental tests show that the detection amplitude is clearly higher with a method according to the invention than with a classic all-point focusing method. In comparative studies, a factor of 10 between the two methods was demonstrated. The reason for this difference is again the use of all of the emission transducers in each firing.

(88) It should also be noted that the invention is not limited to the embodiment described above. It will indeed appear to a person skilled in the art that various modifications may be made to the embodiment described above, in light of the teaching disclosed above.

(89) In particular, the computer program instructions may be replaced by electronic circuits dedicated to functions performed during the execution of said instructions.

(90) In general, in the claims below, the terms used must not be interpreted as limiting the claims to the embodiment described in the present description, but must be interpreted so as to include all of the equivalents that the claims are intended to cover owing to their wording, and which are available to a person skilled in the art applying general knowledge to the implementation of the teaching disclosed above.