WO2014147122A1

WO2014147122A1 - Imaging system and method

Info

Publication number: WO2014147122A1
Application number: PCT/EP2014/055504
Authority: WO
Inventors: Xiaoyu Qiao; Matthias Jobst
Original assignee: Ge Sensing & Inspection Technologies Gmbh
Priority date: 2013-03-22
Filing date: 2014-03-19
Publication date: 2014-09-25
Also published as: DE102013004924B4; DE102013004924A1

Abstract

An imaging ultrasonic testing system (10) comprises first and second linear arrays (12, 14), oriented in different directions in space, and an ultrasonic test unit (20) connected to the first and second linear arrays (12, 14). The ultrasonic test unit (20) is adapted to: insonify a plurality of ultrasonic signals into a test object with the first linear array (12); record the resulting echo signals from the test object with the second linear array (14), and generate a first three-dimensional image of at least a part of the test object volume. The ultrasonic test unit (20) is further adapted to: insonify a plurality of ultrasonic signals into a test object with the second linear array (14); record the resulting echo signals from the test object with the first linear array (12), and generate a second three-dimensional image of at least a part of the test object volume. The ultrasonic test unit (20) is further adapted to: compute a correlation of the first and second images to generate a three-dimensional image of at least a part of the test object volume therefrom.

Description

IMAGING SYSTEM AND METHOD

This disclosure relates generally to non-destructive testing and in particular to ultrasound imaging.

Non-destructive testing devices can be used to inspect test objects to identify and analyze flaws and defects in the objects. An operator is able to move a probe at or near the surface of the test object in order to perform testing of both the object surface and its underlying structure. Non-destructive testing can be particularly useful in some industries such as aerospace, power generation, oil and gas recovery and refining where object testing must take place without removal of the object from surrounding structures and where hidden defects can be located that would otherwise not be identifiable through visual inspection.

One example of non-destructive testing is ultrasonic testing. When conducting ultrasonic testing, an ultrasonic pulse can be emitted from a probe and passed through a test object at the characteristic sound velocity of that particular material. The sound velocity of a given material depends mainly on the modulus of elasticity, temperature and density of the material. Application of an ultrasonic pulse to a test object causes an interaction between the ultrasonic pulse and the test object structure, with sound waves being reflected back to the probe. This corresponding evaluation of the signals received by the probe, namely the amplitude and time of flight of those signals can allow conclusions to be drawn as to the internal quality of the test object, such as cracks or corrosion without destroying it.

Generally, an ultrasonic testing system includes a probe for sending and receiving signals to and from a test object, a probe cable connecting the probe to an ultrasonic test unit and a screen or monitor for viewing test results. The ultrasonic test unit can include power supply components, signal generation, amplification and processing electronics and device controls used to operate the non-destructive testing device.

Some ultrasonic test units can be connected to computers that control system operations as well as test results processing and display. Electric pulses can be generated by a transmitter and can be fed to the probe where they can be transformed into ultrasonic pulses by ultrasonic transducers. Conventional ultrasound imaging systems have an array of ultrasonic transducer elements to scan a targeted object by transmitting a focused ultrasound beam towards the object. The reflected acoustic wave is received, beamformed and processed for display. The beamforming profile is determined by the array structure. With less elements in an array structure, it suffers from the inherent drawback of having a less focused, wide main lobe corresponding to the main transducer beam and higher level side lobes corresponding to secondary, smaller acoustic beams located around the main lobe at predictable angles. This produces a lower quality resultant image with low contrast and resolution between the true reflections and suffering from significant interference. Whilst the level of the side lobes can be reduced by using different shading windows, this widens the main lobe, further decreasing image resolution. Other methods have been considered for reducing the effect of the side lobes, but these generally involve a considerable level of calculation, resulting in increased costs and reduced speed. Whilst the level of the side lobes can be reduced by using a larger array structure, this increases the cost, size and complexity of any such system.

It would be desirable to have an imaging system and corresponding method which reduces the effect of the side lobes and produces a better quality output image without being excessively large, complex or expensive. This disclosure proposes a method according to appended claim 1 for operating an imaging ultrasonic testing system comprising at least a first and a second linear array with a plurality of individually controllable ultrasonic transducers, the first and second linear arrays being oriented in different directions in space. In this case, the method comprises at least the following method steps: a. insonifying a plurality of ultrasonic signals into a test object with the first linear array, b. recording the resulting echo signals from the test object with the second linear array, c. generating a first three-dimensional image of at least a part of the test object volume, d. insonifying a plurality of ultrasonic signals into a test object with the second linear array, e. recording the resulting echo signals from the test object with the first linear array, f. generating a second three-dimensional image of at least a part of the test object volume, g. computing a correlation of the first and second images and generating a three- dimensional image of at least a part of the test object volume from the correlation.

In this case, it is not necessary for the method steps to be carried out in the specified order in time. Alternatively, for example, the steps a. and b. can be carried out first, then the steps d. and e., then the steps c. and f. (optionally also in parallel), and finally the method step g. It was found within the context of elaborate investigations that the indicated method permits the generation of images of the test object of an exceptionally high quality, with the technical expenditure being significantly reduced as compared with conventional methods. On the one hand, this relates to the number of required ultrasonic transducers, which is significantly reduced owing to the use of (at least) two linear arrays instead of one actual 2D array. On the other hand, the reduced number of transducers also requires a significantly decreased computing power for generating the images of the test object, so that real time imaging is possible even with portable devices with the computing power that is available already at the filing date. The computation proposed according to the disclosure of a correlation of the first and second images generated over the course of the method moreover leads to a significant reduction of artifacts in the generated image of the test object, because the influences, which are clearly visible without this correction, of the geometry of the linear arrays of the discrete transducers used, i.e. of the irregularly formed sound field with relevant side lobes, are significantly reduced by means of the correlation. Image generation is in this case particularly simple if the first and second linear arrays used are oriented orthogonally relative to one another.

The generation of a three-dimensional image of the test object volume requires a good spatial resolution in all three directions in space. A good depth resolution can be obtained if the method of "synthetic transmit focusing", which is described for example in the documents Synthetic Aperture Techniques with a Virtual Source Element, Cathrine H. Frazier et al., IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control, Vol. 45 No. 1 , January 1998, 196 - 207, Experimental Study of Transmit Synthetic Focusing Combined with Receive Dynamic Focusing in B-mode Ultrasound Imaging Systems, Moo-Ho Bae et al, 1999 IEEE Ultrasonics Symposium, 1261 - 1264, and the online publication Practical Applications of Synthetic Aperture Imaging, Swetolslav Ivanov Nikolov et al, Technical University of Denmark, available at http://orbit.dtu.dk/en, is used during the insonification of the ultrasonic pulses and the subsequent signal processing. These documents are hereby incorporated by reference into the present disclosure.

An embodiment of the method according to the dislosure, the computation of the correlation comprises a pixel-by-pixel multiplication of the generated three- dimensional first and second images. In another advantageous development, a three dimensional image of the test object can be generated from the correlation computed in this way, by carrying out a normalization step subsequent to the computation of the correlation. The combination of a pixel-by-pixel multiplication and subsequent normalization step proposed herein does not require a large amount of computing power and is therefore possible in real time even with the hardware available at the filing date. At the same time, the interfering influences of the irregularly formed sound field are very effectively suppressed in this process.

An effective transsonification of the test object can be obtained if the sound field generated by the first or second array is pivoted while the method is being carried out. In particular, it is possible to carry out, for example, the method steps a. and b. as well as d. and e. several times while the insonification angle theta of the sound field generated by the first or second array is being changed. In particular, a pivoting of the sound field generated by the respective array by an overall angle of 60° and above is possible in this case. In this way, virtually the entire test object volume can be transsonified without the position of the ultrasound-generating array on the surface of the test object having to be changed for this purpose. The position of the respective array recording the echo signals returning from the test object can also be kept constant in this case.

In order to effectively acquire the test object volume, the method of "dynamic receive focusing" can also be applied, which is described, for example, in the documents Digital Beamforming in Ultrasound, Steinberg, B.D., IEEE Trans. UFFC, Vol. 39, pages 716 - 721 (Nov. 1992) and A pipelined sampling delay focusing in ultrasonic imaging systems, J.H.Kim et. AL, Ultrasonic Imaging, Vol. 9, S. 75 - 91 (1987). These documents are also hereby incorporated by reference into the present disclosure. The "dynamic receive focusing" method permits forming the ultrasound recorded by the receiving array by means of electronic signal processing. In particular, the electronic "adjustment" of a focusing depth of the reflected, electronically focused ultrasound, as well as of a receiving angle, is possible.

The quality of the three-dimensional image of at least a part of the test object volume generated within the context of the method according to the disclosure can be increased further if the steps a. and b. or d. and e., respectively, are carried out several times, preferably more than 10 times, particularly preferably more than 50 times, at a given position of the first and second arrays for a given insonification angle Theta, and if an averaging process is executed with regard to the first and second images generated in the steps c. and f. The computations to be performed within the context of the method according to the dislosusre are considerably simplified if the relative position of the test object as well as of the first and second linear arrays is not changed during the steps a., b., d. and e. This condition may be ensured by using a test probe which comprises both the first as well as the second linear array, with their position relative to one another being fixed. A configuration of the linear arrays is particularly advantageous in which the longitudinal axes of the arrays respectively extend along a straight line. Moreover, a configuration is advantageous in which the arrays intersect in a point at an angle of intersection Alpha in a point of intersection P. In particular in this configuration, it has proved particularly advantageous if the angle of intersection alpha is 90°. Furthermore, it has proved advantageous if the extent of the first array starting from the point of intersection P is equally large in both its extension directions. This also applies to the second array. The overall result is a configuration in which the first and second arrays are disposed in the shape of a plus sign.

A particularly effective transsonification of the test object volume can be obtained if the first and second linear arrays are disposed in a common plane which is non-planar. In particular, it is advantageous if the common plane is rotationally symmetric, particularly with respect to the normal on the test object surface at the location of the first and of the second array, particularly at their point of intersection. In this case, it is advantageous if the common plane is configured concavely. By way of example for this, mention is made of a common plane configured as an upwardly open surface of a spherical segment.

Therefore, the generation of a three-dimensional image of at least a part of the volume of a test object starting from a fixed insonification location by means of two linear arrays oriented in different directions in space, such as orthogonally relative to one another, particularly advantageously comprises the following method steps:

• insonifying ultrasonic signals into the test object with a first array while varying the insonification angle and applying "synthetic transmit focusing",

• recording of the resulting ultrasonic signals from the test object with the second array while applying "dynamic receive focusing". A first three-dimensional image is generated subsequently. These method steps are carried out again for generating a second three-dimensional image, with the role of the first and of the second array being reversed. Finally, an error-corrected three- dimensional image of the examined volume of the sample is generated by means of a correlation of the first and second images. Finally, it has proved advantageous if the method step c. or f, respectively, of generating a first or/and second three-dimensional image, which is carried out in the context of the method according to the dislosure, in particular comprises the following steps: cl ./fl . computing a coherence factor correlated with the proportion of coherent sound power contained in an echo signal, and c2./f2. weighting the echo signal with the computed coherence factor.

Preferably, the coherence factor can assume values between 0 and 1 (inclusively), with a higher coherence factor being correlated with a larger proportion of coherent sound power contained in an echo signal. In this regard, it is not the entire first or second linear array that is being used for insonifying the ultrasonic signal into the test object. Rather, only one sub-array is selected in each case for transmitting. In contrast, the entire other array is always used for receiving. As a matter of principle, this leads to a degradation of the achievable resolution. In order to obtain a high resolution regardless, the ultrasonic signals received from the test object are weighted and coherently added while applying the above-mentioned coherence factor.

Such a process is described in detail in the subsequently published European patent application EP 11 183 566.6 dated September 30, 2011, by the applicant of the present application. The content of the disclosure of this document is hereby added by this reference to the subject matter of the present application.

Furthermore, this disclosure proposes an imaging ultrasonic testing system having the features of claim 11. Such an ultrasonic testing system comprises at least a first and a second linear array of ultrasonic transducers, the first and second linear arrays being oriented in different directions in space. Advantageously, the first and the second linear arrays are in this case oriented orthogonally with respect to one another.

Furthermore, it comprises an ultrasonic test unit operatively connected to the first and second linear arrays. The ultrasonic test unit is configured to insonify with the first linear array a plurality of ultrasonic signals into a test object and to record with the second linear array the resulting echo signals from the test object. Moreover, it is configured to generate a first three-dimensional image of at least a part of the test object volume.

Furthermore, the ultrasonic test unit is configured to insonify with the second linear array a plurality of ultrasonic signals into a test object and to record with the first linear array the resulting echo signals from the test object. Furthermore, the ultrasonic test unit is configured to generate a second three-dimensional image of at least a part of the test object volume.

Finally, the ultrasonic test unit is configured to compute a correlation of the first and of the second image and to generate a three-dimensional image of at least a part of the test object volume from the correlation. This three-dimensional image can then be displayed on a suitable display unit, which can be part of the ultrasonic testing system according to the dislosure. Furthermore, a data file can be generated which represents the generated three-dimensional image of at least a part of the test object volume. In an advantageous embodiment of the ultrasonic testing system according to disclosure, the ultrasonic test unit is furthermore configured to carry out a pixel-by- pixel multiplication of the generated three-dimensional first and second images during the computation of the correlation. Furthermore, the ultrasonic test unit of the ultrasonic testing system may be configured to carry out a normalization step during the generation of a three-dimensional image from the correlation generated by means of, for example, pixel-by-pixel multiplication.

In another padvantageous embodiment of the ultrasonic testing system according to the dislosure, the ultrasonic test unit is further configured to control the insonification angle Theta of the sound field generated by the first or second array, and to carry out the method steps a. and b. as well as d. and e. several times while the insonification angle Theta of the sound field generated by the first or second linear array is changed. In this manner an effective transsonification of the test object volume can be achieved already from a single insonification location. In another advantageous embodiment of the ultrasonic testing system according to the dislosure, the ultrasonic test unit is further configured to vary the focusing depth of the sound field generated by the first or second array in accordance with the method of "synthetic transmit focusing" described in the introduction. In another advantageous embodiment of the ultrasonic testing system according to the dislosure, the ultrasonic test unit is further configured to vary the focusing depth and/or the angle of the sound field recorded by the first or second array in accordance with the method of "dynamic receive focusing" described in the introduction.

Advantageously, both the first and the second linear arrays are integrated into a common test probe. The relative position of the first and second arrays relative to one another is fixed therein. In this case, a configuration of the linear arrays is particularly advantageous in which the longitudinal axes of the arrays respectively extend along a straight line. Moreover, a configuration is advantageous in which the arrays intersect in a point at an angle of intersection Alpha in a point of intersection P. In particular in this configuration, it has proved particularly advantageous if the angle of intersection alpha is 90°. Furthermore, it has proved advantageous if the extent of the first array starting from the point of intersection P is equally large in both its extension directions. This also applies to the second array. The overall result is a configuration in which the first and second arrays are disposed in the shape of a plus sign. A particularly effective transsonification of the test object volume can be obtained if the first and second linear arrays are disposed, in the common test probe, in a common plane which is non-planar. In particular, it is advantageous if the common plane is rotationally symmetric, particularly with respect to the normal on the test object surface at the location of the first and of the second array, particularly at their point of intersection. In this case, it is advantageous if the common plane is configured concavely. By way of example for this, mention is made of a common plane configured as an upwardly open surface of a spherical segment.

In principle, it is possible that in the ultrasonic testing system according to the disclosure the first and second linear arrays are configured as sections of an extended two-dimensional array comprising a plurality of individually controllable ultrasonic transducers. The two-dimensional array can be, for example, a rectangular, in particular square, array. Furthermore, the individually controllable ultrasonic transducers of the array can be rectangular, in particular square. It is pointed out that neither the first and second arrays, nor a possible two-dimensional array, nor their respective individually controllable ultrasonic transducers have to be planar. Rather, in an advantageous embodiment, both the linear arrays and a two-dimensional array, and optionally also the individually controllable ultrasonic transducers, are curved and form surfaces of a spherical segment, for example.

In an advantageous development, the ultrasonic test unit of the ultrasonic testing system is configured to respectively control, within the context of generating the first or/and the second three-dimensional image, not the entire first or second array, but only a subset of the transducers contained in a linear array (a so-called "sub-array"). In contrast, the entire other array is always used for receiving. As a matter of principle, this leads to a degradation of the achievable resolution, as was already mentioned above. In order to obtain a high resolution regardless, the ultrasonic test unit is configured to weight the ultrasonic signals received from the test object and to coherently add them up while applying a so-called coherence factor. The transmission of several ultrasonic signals with different sub-apertures of the array provides an improved sensitivity without a large array being necessary, and the use of the coherence factor increases the suppression of the side lobes, which improves the clarity, the contrast and the resolution of the resulting image.

For this purpose, the ultrasonic test unit is furthermore configured to compute a coherence factor correlated with the proportion of coherent sound power contained in an echo signal. The ultrasonic test unit furthermore weights the echo signal with the computed coherence factor. The coherence factor may assume values between 0 and 1 (inclusively), with a higher coherence factor being correlated with a larger proportion of coherent sound power contained in an echo signal. A larger value indicates a higher proportion of coherent intensity contained in the entire received signal, and thus a higher confidence of a good focusing quality. Implenentations of the present disclosure will now be described by means of examples with reference to the accompanying drawings, in which:

Figure 1 shows an example of a typical ultrasonic testing system;

Figure 2 shows an example of an ultrasonic testing system illustrating the

present dislosure;

Figure 3 schematically shows the functional configuration of the ultrasonic test unit 20 according to Figure 2 in the transmitting mode;

Figure 4 schematically shows the functional configuration of the ultrasonic test unit 20 according to Figure 2 in the receiving mode; Figures 5a-5c show examples for different configurations of orthogonal linear arrays for carrying out the present dislosure;

Figure 6 shows an example for non-planar configurations of two orthogonal linear arrays for carrying out the present dislosure, and

Figure 7 shows a schematic representation of an ultrasonic system according to the dislosure.

Figure 1 represents an example of an ultrasonic testing system 1. The system comprises a test probe 10 for transmitting and receiving signals to and from a test object 100. In this example, the test probe 10 is disposed for transmitting and receiving a reflected ultrasonic signal from the test object 100. In other examples, however, the test probe could instead be arranged so as to receive ultrasonic signals transmitted through a test object 100. The test object 100 could be any suitable object to be analyzed with regard to flaws and defects, such as panels of a vehicle, such as of an airplane or of a ship, sections of a pipeline or parts of an industrial plant, which can take place without having to remove the object from surrounding structures. In use, the test probe 10 is moved over the surface of the test object 100 in order to analyze the structure of the object. The test probe 10 comprises an array of ultrasonic transducers 11. A test probe cable 16 connects the test probe 10 to an ultrasonic test unit 20.

The ultrasonic test unit 20 comprises a control unit for signal generation, electronic amplification and processing units, in order to, for example, generate electronic pulses that are fed to the test probe 10, where they can be converted into ultrasonic pulses by the ultrasonic transducers 11. The ultrasonic test unit 20 can also receive the reflected signal generated by the test probe 10. The ultrasonic test unit 20 comprises an electrical output and can include a display unit 30 in the form of a screen or monitor, or - as shown - can be connected thereto in order to display results based on an output to a user in order to enable him to analyze the structure of the test object 100 and detect possible faults or defects in the test object. The display unit 30 can be provided by a computer (not shown), which can be connected to the ultrasonic test unit 20 and which can also provide some of the functions of the ultrasonic test unit 20.

Figure 2 schematically shows an ultrasonic testing system 1 according to an example of the present dislosure whose structure substantially corresponds to that shown in Fig. 1. In this case, the test probe 10 is depicted in more detail, which comprises two linear arrays 12, 14 which are disposed orthogonally relative to one another and each comprise a plurality N of individually controllable ultrasonic transducers 11. Every ultrasonic transducer 11 can act both as a transmitter for ultrasonic signals and as an ultrasonic receiver. In Figure 2, the arrays 12, 14, by way of example, each comprise seven ultrasonic transducers 11, i.e. N=7. In practice, the number N will be higher and be, for example, 32, 64 or 128, in order to obtain a sufficient image quality.

In this case, the ultrasonic test unit is configured to control the ultrasonic transducers 11 of each of the arrays 12, 14 individually and phase-accurately in the form of a "phased array", so that the fan-shaped ultrasonic field generated by an array 12, 14 can be both pivoted and focused to a different depth. Preferably, the ultrasonic test unit 20 is configured to control, in a transmission mode, each of the two arrays 12, 14 in a "synthetic transmit focusing" mode.

Moreover, the ultrasonic test unit 20 is configured to be operated in a receiving mode in which the ultrasonic signals recorded by the individual ultrasonic transducers 1 1 of one array 12, 14, respectively, are individually detected and processed in a time- resolved manner. The ultrasonic test unit 20 may be configured to evaluate, in a receiving mode, the recorded ultrasonic signals of each of the two arrays 12, 14 in a "dynamic receive focusing" mode. In this "dynamic receive focusing" mode, the received ultrasonic signals are processed in a time-resolved manner in such a way that the direction and/or focusing depth of a "virtual" echo sound field can be controlled in a targeted manner.

A combination of a fan- shaped transmission ultrasonic field with a "synthetic transmit focusing" transmitting mode and a variation of the insonification angle theta into the test object 100 and a processing of the received ultrasonic signals in the "dynamic receive focusing" receiving mode leads to an effective transsonification of a large sector of the test object with a significantly reduced number of ultrasonic pulses, even from a single insonification location.

Furthermore, the ultrasonic test unit 20 is configured to control either array 12 as a transmitting array and array 14 as a receiving array, or vice versa.

Figure 3 schematically shows the functional configuration of the ultrasonic test unit 20 according to Figure 2 in the transmitting mode. The control signal generated by a pulse generator 5 is supplied to every individual ultrasonic transducer 11 of a linear array, e.g. to the array 12, with an individual delay generated in a delay unit 27 being applied in each case for the purpose of beamforming. It is thus possible to generate beamformed, controlled, focused transmission beams 28 that penetrate into the test object 100 with a requisite scanning control angle and a requisite focusing depth.

Figure 4 schematically shows the functional configuration of the ultrasonic test unit 20 according to Figure 2 in the receiving mode. In this example, all of the ultrasonic transducers 11 of the other array, e.g. of the array 14, are used to receive the reflected signal. In this example, the electrical signal generated by each transducer element 11 in accordance with its recorded ultrasonic signal obtains a suitable beamforming delay from the delay unit 21 , and the delayed received signals are added up in the summing unit 22 and normalized in the normalization unit 23. The synthesis unit 25 generates a three-dimensional image of the volume of the test object 100 therefrom. In practice, the delay 21, the adding 22, the normalizing 23 and the synthesizing 25 are carried out in a control unit, such as, for example, a computer, a freely programmable or an application-specific microprocessor, or by a hard-wired electronic unit. An optional development, the ultrasonic test unit moreover comprises a coherence unit 24, which is configured to determine a coherence factor for the signal of each receiving ultrasonic transducer 11 , as this is explained in detail in EP 11 183 566.6, to which reference has already been made above. Reference is made to the explanations there. The generally time-dependent coherence factor CF(t) serves for transducer- specific weighting of the received signals corresponding to the proportion of coherent energy in each received signal. The signal of each receiving ultrasonic transducer 11 is weighted with the associated coherence factor CF(t), e.g. by multiplication in a multiplication unit 26. The signals thus weighted are then used for generating the first three-dimensional image of the transsonified test object volume. The introduction of a coherence unit 24 mentioned above makes it possible, moreover, to operate the ultrasonic test unit 10 according to the disclosure in such a way that, in a preferred development, sub-apertures consisting of a subset of the ultrasonic transducers 1 1 of one array 12, 14 are used for transmitting, with, however, all of the ultrasonic transducers 11 of the other array 14, 14 being used in each case for recording the resulting ultrasonic echo signals. The transmission sub-apertures can assume any desired focusing strategy, such as, for example, fixed focusing or dynamic focusing, by corresponding control of the delays in the unit 27. The corresponding delays are used in the signal processing of the receiving ultrasonic transducers 11 of the other array 14, 12. The received signals from every sub-aperture transmission are then weighted by a coherence factor CFl(t), CF2(t), CF3(t) ... according to the proportion of coherent energy in every received signal. The weighting 23 can include a multiplication with the coherence factor. In this example, the coherence factor is determined in a coherence factor unit 24 in order to correspond to the proportion of coherent energy in the total non-coherent energy of each received signal from each sub-aperture transmission. The received signals from each sub-aperture transmission weighted with the coherence factor are then added up by the synthesis unit 25 in order to provide a clearer three-dimensional image of the transsonified test object volume in such a way that possible flaws or faults can be easily identified. The use of sub-apertures of the first arrays 12 or 14 used for the generation of the ultrasonic field insonified into the test object 100 while, in contrast, all elements 11 of the second arrays 14 or 12 used for receiving are active during the reception of the echo signal, increases, in particular, one or more of the quantities sensitivity, penetration depth or signal-to-noise ratio, which characterize image quality. Figure 5a shows an alternative transducer arrangement of a test probe 10 consisting of a two-dimensional orthogonal array of N x N individually controllable ultrasonic transducers 11 , with the number N of ultrasonic transducers per line and column being seven in the Figure. In practice, arrays with a larger number N, e.g. 32, 64 or 128, will be used also in this case. In the example embodiment according to Figure 5a, only the central column and the central line of the array will be controlled in each case, so that the result is an orthogonal, cross-shaped configuration of the array 12, 14.

Fig. 5b shows a different control of the orthogonal two-dimensional array from Fig. 3 a, in which only the ultrasonic transducers 11 disposed on the diagonals of the arrays are being controlled as linear arrays 12, 14. Fig. 5c finally shows another control of the orthogonal two-dimensional array from Fig. 5a, in which two first linear arrays 12, 12' that extend in parallel and two second linear arrays 14, 14' that are oriented orthogonally thereto are controlled in each case. By way of example, these linear arrays 12, 12', 14, 14' are formed by the ultrasonic transducers 11 of the two-dimensional array situated at the edges. Figure 6 shows a two-dimensional array of ultrasonic transducer, which can be used advantageously within the context of the present dislosure and which, in contrast to the arrays from the Figures 2 and 5a - c, is non-planar. Rather, the ultrasonic transducers are disposed in orthogonal lines and columns on a surface of a spherical segment. This configuration enables the realization of a particularly large pivoting angle both for the insonified ultrasonic field and for the recorded ultrasonic echo.

The mode of operation of an example embodiment of an ultrasonic testing system according to the dislosure is explained with reference to Figure 7. In a first step, which is symbolized by the left half of Figure 7, ultrasonic pulses are insonified into the test object 100 at different insonification angles theta by means of all ultrasonic transducers 11 of the first linear array 12 in the form of an ultrasonic field that is focused in the y direction, but is divergent and thus fan-shaped in the x direction, as this was explained above with reference to Fig. 3. The resulting ultrasonic echoes are acquired in a time-resolved manner by means of all ultrasonic transducers of the orthogonally oriented second array 14. The received ultrasonic signals are subsequently processed in accordance with the above-described method of the "dynamic receive focusing" mode. The received ultrasonic signals are processed in a solid-angle-resolved and depth-resolved manner, whereby a three-dimensional first image of the transsonified volume of the test object 100 is generated, as this was explained in detail above with reference to Fig. 4. Due to the rather small number of receiving transducer elements 11 in the second array 14, which here acts as a receiver, the achievable image quality is lower than if all of the ultrasonic transducers 1 1 of the entire two-dimensional array are being used. However, the significant reduction of the number of ultrasonic signals of the individually receiving ultrasonic transducers that are to be processed individually - advantageously in parallel and in real time - considerably reduces the technical expenditure, so that, on the whole, the generation of a three-dimensional image of the transsonified volume of the test object 100 in real time becomes possible even with a portable ultrasonic testing system. With regard to transmitting and receiving of the ultrasonic signals, the role of the first and of the second array 12, 14 is subsequently swapped so that now, the above- described imaging method is carried out again, using the second linear array 14 as a transmitting array and the first linear array 1 as a receiving array. As a result, a second three-dimensional image of the same transsonified sector of the test object 100 as in the first image is provided. According to the dislosure, a correlation of the first and of the second three- dimensional image is now generated in a correlation unit 28 and used for generating a normalized three-dimensional image of the transsonified sector of the test object, which can be stored as a data file for further use or for archiving purposes and/or displayed on a display unit 30, e.g. a suitable display. For this purpose, a variety of methods is known from the prior art, particularly from image processing, from which the person skilled in the art can select a method that is particularly suitable for the specific application without having to resort to an inventive step. By way of example, reference is made here to a pixel-by-pixel multiplication of the brightness values with a subsequent normalization, which has proved to be an effective correlation method in the sense of the present dislosure.

The correlation proposed according to the dislosure of the low-resolution three- dimensional images generated within the context of the proposed method leads to a significant improvement of the image quality, without any substantially increased technical effort having to be made in obtaining the ultrasonic signals on which the image generation is based. The computation of the correlation can in this case be carried out by means of freely programmable microprocessors; however, the use of hardware custom-made for this purpose, e.g. in the form of ASICs, is generally advantageous due to advantages with regard to speed. Whilst the above description relates to particular implementations of the present disclosure, the skilled person will appreciate that variations and modifications may be made to the described implementations that fall within the scope of the subject mater defined in the appended claims.

Claims

CLAIMS:

1. A method for operating an imaging ultrasonic testing system comprising at least a first and a second linear array with a plurality of individually controllable ultrasonic transducers, the first and second linear arrays being oriented in different directions in space, comprising the following method steps: a. insonifying a plurality of ultrasonic signals into a test object with the first linear array, b. recording the resulting echo signals from the test object with the second linear array, c. generating a first three-dimensional image of at least a part of the test object volume, d. insonifying a plurality of ultrasonic signals into a test object with the second linear array, e. recording the resulting echo signals from the test object with the first linear array, f. generating a second three-dimensional image of at least a part of the test object volume, g. computing a correlation of the first and second images and generating a three-dimensional image of at least a part of the test object volume from the correlation.

2. The method according to claim 1, wherein the first and the second arrays are oriented orthogonally with respect to one another.

3. The method according to claim 1 or claim 2, wherein the method of "synthetic transmit focusing" is applied for insonifying the plurality of ultrasonic signals with the first and/or the second linear array.

4. The method according to claim 1, 2 or 3, wherein the insonification angle is varied during the insonification of the plurality of ultrasonic signals with the first and/or the second linear array.

5. The method according to any preceding claim, wherein the receiving angle is varied during the recording of the resulting echo signals from the test object with the second and/or the first linear array.

6. The method according to any preceding claim, wherein the method of "dynamic receive focusing" is applied during the recording of the resulting echo signals from the test object with the second and/or the first linear array.

7. The method according to any preceding claim, wherein the computation of the correlation comprises a pixel-by-pixel multiplication of the generated three- dimensional first and second images.

8. The method according to any preceding claim, wherein the generation of a three-dimensional image from the correlation comprises a normalization step.

9. The method according to any preceding claim, wherein the first and the second linear arrays are disposed in a common plane, which is non-planar.

10. The method according to any preceding claim, wherein the common plane is rotationally symmetric.

11. The method according to any preceding claim, wherein the common plane is a surface of a spherical segment.

12. The method according to any preceding claim, wherein the generation of the first or/and second three-dimensional images comprises the following further steps: a. computing a coherence factor correlated with the proportion of coherent sound power contained in an echo signal, and b. weighting the echo signal with the computed coherence factor.

13. The method according to claim 12, wherein the coherence factor can assume values between 0 and 1 (inclusively), with a higher coherence factor being correlated with a larger proportion of coherent sound power contained in an echo signal.

14. Imaging ultrasonic testing system (10), comprising: a. at least a first and a second linear array of ultrasonic transducers (12, 14), the first and second linear arrays (12, 14) being oriented in different directions in space, b. an ultrasonic test unit (20) connected to the first and second linear arrays (12, 14) and adapted to: i. insonify a plurality of ultrasonic signals into a test object with the first linear array (12), ii. record the resulting echo signals from the test object with the second linear array (14), iii. generate a first three-dimensional image of at least a part of the test object volume, iv. insonify a plurality of ultrasonic signals into a test object with the second linear array (14), v. record the resulting echo signals from the test object with the first linear array (12), vi. generate a second three-dimensional image of at least a part of the test object volume, vii. compute a correlation of the first and of the second image and to generate a three-dimensional image of at least a part of the test object volume from the correlation.

15. The ultrasonic testing system according to claim 14, wherein the ultrasonic test unit (20) is furthermore configured to carry out a pixel-by-pixel multiplication of the generated three-dimensional first and second images during the computation of the correlation.

16. The ultrasonic testing system according to claim 15, wherein the ultrasonic test unit (20) is furthermore configured to carry out a normalization step during the generation of a three-dimensional image from the correlation.

17. The ultrasonic testing system according to claim 14, 15 or 16, wherein the ultrasonic test unit (20) is configured to: a. control the insonification angle theta of the sound field generated by the first or second array, and b. carry out the method steps a. and b. as well as d. and e. several times while the insonification angle theta of the sound field generated by the first or second array is being changed.

18. The ultrasonic testing system according to any one of claims 14 to 17, wherein the first and the second linear arrays (12, 14) are disposed in a common plane, which is non-planar.

19. The ultrasonic testing system according to any one of claims 14 to 18, wherein the common plane is rotationally symmetric.

20. The ultrasonic testing system according to any one of claims 14 to 19, wherein the common plane is a surface of a spherical segment.

21. The ultrasonic testing system according to any one of claims 14 to 20, wherein the first and second linear arrays (12, 14) are configured as sections of an extended two-dimensional array comprising a plurality of individually controllable ultrasonic transducers.

22. The ultrasonic testing system according to any one of claims 14 to 21, wherein the ultrasonic test unit (20) is further configured to: a. compute a coherence factor correlated with the proportion of coherent sound power contained in an echo signal, and b. weight the echo signal with the computed coherence factor during the generation of the first or/and the second three-dimensional image.

23. The ultrasonic testing system according to claim 22, wherein the coherence factor computed by the ultrasonic test unit (20) can assume values between 0 and 1 (inclusively), with a higher coherence factor being correlated with a larger proportion of coherent sound power contained in an echo signal.