Method for locating an impact on a surface and device therefor

Translation of description of equivalent WO 03107261 A2

p0001A method for locating an impact on a surface and device for œuyre formatting method.

p0002The present invention relates to methods for locating an impact on a surface and to the devices for the implementation of these methods.

p0003More particularly, the invention relates to a method in which locates an impact on a surface belonging to an object forming an acoustic interface, provided with at least one acoustic sensor (the object forming an acoustic interface may be made of a single piece or of several components, assembled or at least in mutual contact), process in which captures at least one signal from acoustic waves. generated in the object forming an acoustic interface. said impact is localized and the impact by processing said sensed signal.

p0004FR-A-2 811 107 describes an example of such a process which is particularly applicable to glass. In this known process, the position is calculated from the impact on the surface of the object by measuring the time of flight differences of the acoustic waves to the different sensors.

p0005This known method, however, requires: that the glass used are perfectly homogenous and perfect surface finish - as the fields of the glass are treated specially in particular to avoid reflections of sound waves, that is known in advance speed of the acoustic waves in the glass, which means to know precisely its composition, which is used at least four sensors.

p0006This means that this known method is particularly expensive to implement and can not be applied to any pre-existing objects, including object made of heterogeneous assemblies parts, irregularly shaped objects, etc.

p0007The present invention is intended to overcome these drawbacks.

p0008To this end, according to the invention, a method of the kind in question is characterized in that it comprises a recognition step during which the sensed signal is compared with at least one predetermined signal corresponding to the signal that is sensed when generating an impact on at least one active zone belonging to the surface of the object forming an acoustic interface (this comparison that can be made both in the time domain than in the frequency domain, may optionally be performed on only a portion the received signal or on data from the signal picked up after processing, in which case said predetermined signal may be reduced to the portion on which the comparison is made or the data on which the comparison is made), and the impact is associated with said active area if sufficient signal neighbor said predetermined signal. Thanks to these features, we get an impact positioning method which is robust, adaptable to all objects (including heterogeneous objects made by assembling several parts or by contacting several pieces), easy and inexpensive to enforce. In preferred embodiments of the invention, may optionally be furthermore employed one and / or to the other as follows: the surface of the object forming an acoustic interface comprises several active zones, and in the recognition step, comparing the signal picked up at several predetermined signals each corresponding to the sensed signal when generating an impact on one of said active areas; one acoustic sensor is used; - Using several acoustic sensors and during the recognition step, one captures a signal for each acoustic sensor and the signals sensed by the various acoustic sensors are compared with the predetermined signals independently of each other; - The signals sensed by the various acoustic sensors are compared with predetermined signals differently from each other; a plurality of acoustic sensors for measuring several different magnitudes are used; - Is used at most two acoustic sensors; the method includes an initial learning step during which each predetermined experimentally determining signal for generating at least one impact on each active zone; - Each predetermined signal is a theoretical signal (calculated or experimentally determined on an object identical or very similar the acoustic point of view to the one used);

p0009- During the recognition step, comparing the sensed signal to said at least one predetermined signal by intercorrelation; during the recognition step, comparing the sensed signal to said at least one predetermined signal by a process of recognition chosen from voice recognition, signal recognition, shape recognition and recognition by neural network; during the recognition step, one associates the received signal either to a single active region, or no active area; each active zone is associated with a predetermined information (for example, an alphanumeric character, a command, etc.) and when combined impact to an active area, we actually use the predetermined information element corresponding to that active zone by device electronics; the surface of the object forming an acoustic interface comprises a number n of active zones, n being at least equal to 2, and the recognition step comprises the following sub-steps of: conducting a cross-correlation of the sensed signal

p0010(Usually after normalization) with said predetermined signals Ri (t), i being a natural number between 1 and n which designates an active zone, thus obtaining the functions of correlation Ci (t), it is determined an active zone j potentially activated corresponding to the result of intercorrelation Cj (t) having a maximum amplitude greater than those of the other results Ci (t). it also determines the distribution D (i) of the maximum amplitude of the results of correlation: D (i) = Max ((C (t)), it also determines the distribution D '(i) the maximum amplitude results intercorrelation C'i (t) between Rj (t) and the various predetermined signals Ri (t) = D '(i) = Max ((C' i (t)), it is determined whether the impact was generated on the active zone j as a function of a level of correlation between the distribution D (i) and D '(i) - during the recognition step, the sensed signal is processed to extract data representative of certain characteristics of the sensed signal and comparing the data thus extracted to the reference data extracted from the signal that is sensed when an impact is generated on each active zone, during the recognition step, a code is determined from said data extracted from the sensed signal and this code is compared with a table which gives a correspondence between at least certain codes and each active zone; the object forming an acoustic interface comprises at least two active zones and during the recognition step, it is determined similarity representative values ​​of the resemblance between the sensed signal and the predetermined signals (such as a value from the function cross-correlation, for example its maximum), the impact is associated with several adjacent active zones corresponding to a maximum resemblance, called reference active zones, then the position of impact is determined on the surface based on assigned resemblance values the reference active zones; determining the position of impact on the surface so that the resemblance values ​​attributed to the reference active zones correspond as well as possible to theoretical resemblance values ​​computed for said reference active zones for an impact generated in said position on the surface ; the theoretical resemblance values ​​are functions of the position of the impact on the surface, determined in advance for each possible set of reference active zones; identifying the active region by comparison between the phase of the predetermined signals Ri (t) and the received signal; - During the learning phase, calculating the Fourier transform R (ω) = | Ri (ω) | .e^{φι j (ω)} of each acoustic signal R (t) generated by an impact on the active zone i, where i is an index lying between 1 and n, and retains this Fourier transform only the phase component e^{3 φι (ω)}. only in the frequency bands ω in which the amplitude | Ri (ω) | is greater than a predetermined threshold, then applies the same processing to each sensed acoustic signal S (t) during the normal operation of the device; - The predetermined threshold is equal to the maximum of AX / D and | B (ω) | , or :

p0011MAX is chosen from the maximal value of the modules

p0012| Ri (ω) |, the maximal value of the modules | Ri (ω) | each normalized in energy, and the maximum value of the envelope of the average of the modules | Ri (ω) | each normalized in energy,

p0013D is a constant,

p0014| B (ω) | is the average of several noise spectra in the object forming an acoustic interface, acquired at different times;

p0015- During normal operation of the device: computing a product Pι (ω) equal to S '(ω) multiplied by the conjugate of R i' (ω) for references i = 1 ... n, then we normalize the products Pi ( ω). then performs the inverse Fourier transform of all Pi products (ω), and obtains time functions Xι (t) and the signal S is assigned (t) at an active area (10) based on said temporal functions Xι (t); - The signal S is assigned (t) to an active region based on the maximum values ​​of said temporal functions Xι (t).

p0016Furthermore, invention also relates to a device especially adapted to implement a method¹ interface as defined above.

p0017Other features and advantages of the invention appear from the following description of five of its embodiments, given by way of non-limiting examples in the accompanying drawings.

p0018In the drawings: Figure 1 is a schematic perspective view showing an exemplary device comprising an acoustic interface adapted to implement a method according to a first embodiment of 1 invention, Figure 2 is a block diagram of the device of Figure 1, Figure 3 represents a graph illustrating an exemplary method for associating an impact on the visible surface of the acoustic interface in Figure 1, with an active area of ​​this surface, Figure 4 schematically shows an acoustic interface usable in a device implementing a method according to a second embodiment of the invention, Figure 5 is a block diagram of an example device that can use the input interface of Figure 4, - and figures 6 to 9 show schematically acoustic interfaces for use in a device implementing a method according to the third, fourth and fifth embodiments of 1 invention. In the various figures, the same references designate identical or similar elements.

p0019FIG 1 shows a device 1 for implementing the present invention, which comprises for example: - a central processing unit 2 of the microcomputer, a display 3 connected to the central processing unit 2, and an acoustic input interface 4 which allows to communicate information to the central unit 2 in the example. The acoustic input interface 4 comprises a solid object 5, here constituted by a table in which acoustic waves propagate is made by generating impacts on its surface 9, as will be explained below.

p0020Note however that the object forming an acoustic interface may be constituted by any other object, homogeneous or heterogeneous, consisting of a single piece or several pieces assembled or simply in mutual contact, such as: glass, door, window, portable tablet computer screen, billboard, kiosk toy table board vehicle rear seat back before the motor vehicle or aircraft seat, wall, floor, bumper car (the information transmitted by the acoustic interface then being the position of an impact on the front -chocs), etc .. at least one acoustic sensor 6 (a single sensor 6 in the example shown) is attached to the object 5, this acoustic sensor 6 being connected for example to one input microphone 7 of the unit Central 2, via a cable 8 or by any other means of transmission (radio, infrared or other) to sense said acoustic waves and transmit them to the central unit 2.

p0021The acoustic sensor 6 may for example be a piezoelectric sensor, or other (e.g., a capacitive sensor, a magnetostrictive sensor, an electromagnetic sensor, a velocimeter acoustic, an optical sensor [laser interferometer laser vibrometer, ...] , etc.). It can be adapted to measure eg the amplitudes of displacements due to the propagation of sound waves in the object 5 forming an acoustic interface, or the speed or acceleration of such movements, or he may be of a pressure sensor measuring the pressure variations due to the propagation of acoustic waves in the object 5.

p0022On the outer surface 9 of the object 5 (in this case on the upper face of the table constituting said object 5 in the example shown in Figure 1), there are defined several active zones 10, which may be defined e.g. : by a physical marking, removable or not affixed to the surface 9 of the object 5, or by a luminous marking obtained by projecting an image on the surface 9.

p0023The surface 9 could also have parts where the ban to generate impact, for example by covering them with a flexible material or simply inaccessible to the user, including higher system reliability.

p0024The various active region 10 may simply be portions of the surface 9, identical to the rest of the surface 9. These active areas, however, differ from each other and the rest of the surface 9, to the extent that an impact on one of the zones 10 generates a different sound signal of the signal generated by an impact on another of the active zones 10 or on another part of the surface 9.

p0025Each of the active areas 10 is associated with a predetermined information a user may want to communicate to the CPU 2. The information in question may for example be a command, a number, a letter, a position on the surface 9, or any other information that can be usually transmitted to an electronic device such as a computer (or the CPU other electronic device) using conventional input interfaces such as keypads, buttons, mouse or others.

p0026The information in question may possibly be indicated in clear by 10a markings on the surface 9 (as pins zones 10, these markings may be affixed physically on the surface 9 of permanent or removable manner, or they may be projected under as bright images on said surface 9).

p0027Alternatively, the surface 9 of the object 5 may simply comprise marks (physically affixed or light) to distinguish the active areas each other. These marks may for example be numbers or colors and their meaning can possibly be returned by a display generated by the CPU 2 on the screen 3. Optionally, the surface 9 may also comprise no marking or to delineate active areas, or to identify the information to which they correspond, in which case the active areas 10 would be known only to authorized users of the device 1.

p0028the predetermined information is noted that associated with each active area 10 can be always the same, or vary depending on the execution of a program in the central unit 2, or else rely on previous actuations other active zones 10 (some active region 10 may for example be operable to change the function assigned to one or more area (s) active (s) 10 actuated (s) after it, so that, for example, to access specific functions, of characters. special, or to put capital letters, etc.).

p0029The various active zones 10 of the object 5 thus constitute a virtual keyboard which is operated by tapping the active areas, either with a fingernail, with the fingertips, with a subject as pen, stylus or other.

p0030Note that the surface 9 of the object 5 could where appropriate comprise a single active area 10 in the simplest case, this active zone 10 does s' but not extending over the whole of the surface 9 and preferably constituting a small portion of said surface 9.

p0031As shown in Figure 2, the sensor 6 (SENS.) May conventionally be connected by way of the inlet 7 to an amplifier 11, itself connected to an analog-digital converter 12 (A / D) which transmits the received signals to the processor 13 of the CPU 2

p0032(CPU) processor which is itself connected to one or more memories 14 (MEM.) And controls the aforementioned screen 3 (SCR.) Or any other output interface returning information to the user.

p0033the acoustic interface Note that 4 could serve as an information input interface to all other electronic devices as a microcomputer, such as a household appliance or electronic professional, digital code, an electronic central unit of the vehicle, etc. In all cases, the electrical signals generated by the sensor 6 may be treated either in this electronic device, or an external device digital signal processing (DSP). During use of the device 1 described above, when a user generates an impact on the surface 9 of the object 5, the impact generates an acoustic wave which propagates in the object 5 to the acoustic sensor 6. The sensor acoustic 6 then generates an electrical signal S (t) which, after scanning are processed by the processor 13 (or by another processor dedicated, internal or external to the central processing unit 2).

p0034The processor 13 then compares the received signal with different predetermined signals belonging to a library of signals previously stored in the memory 14, these predetermined signals corresponding respectively to impacts generated on different active zones 10 of the object 5.

p0035This comparison indicates whether the acoustic signal is from one of the active zones 10, and which, regardless of the mode of excitation of said active surface (impact of a nail, a toe end, a palm, an object such as a pen or stylus, etc.). The predetermined signals from the library signals may have been determined during an initial learning phase in which successively generates impacts on all active areas 10 of the object 5, recording the corresponding signals (preferably after normalization, for example that energy of each reference signal is equal to 1) received in the central unit 2 through the acoustic sensor 6.

p0036Alternatively, when the object 5 has a simple geometric shape and / or repetitive, it is possible that the signals from the library of predetermined signals are obtained by modeling or experimentally be determined once for all objects 5 of a series of identical objects: in both cases, so there would be no prior learning phase for the object 5 particularly connected to the CPU 2, but simply installing the library of signals in the memory 14 said central unit.

p0037Note that in some cases (particularly if the object 5 is made of wood), one can vary the predetermined signals from the library of signals depending on the ambient conditions including temperature and humidity. These variations can be calculated or be the result of a new learning phase. Comparing the signals received during use of the device 1 with the predetermined signals from the library of signals, may be made:

p0038- Directly on the S time signals (t) received from β sensor, or on the frequency spectrum of these signals (eg after Fourier transform of the time signals received from sensor 6), or on other data signal characteristics including its phase. Comparing the signals captured with the predetermined signals from the library of signals may be effected by any known means, including: by intercorrelation, - by methods known voice recognition, signal recognition or pattern recognition,

p0039- By use of neural networks, or others. As a more specific example, may especially be used to recognize the active zone 10 where does the sensed signal S (t), the following process:

p0040(1) After normalizing the sensed signal S (t) (for example, caliber S (t) so that its energy is equal to 1), one proceeds to a cross-correlation of the signal S (t) generated by the sensor 6 with the n predetermined signals also normalized library, denoted Ri (t) i = l..n. Are thus obtained functions Ci (t), which are the temporal results of the product of cross-correlation of the signal S (t) respectively with the signals Ri (t) of the library. From these calculations, it is determined a potentially activated active zone j corresponds to the result of intercorrelation Cj (t) having a maximum amplitude greater than those of the other results Ci (t). (2) It also determines the distribution D (i) of the maximum amplitude of the results of correlation: D (i) = ax ((C (t)) with i = l..n.

p0041(3) a second distribution function is calculated D '(i) identically obtained to calculate the function D (i) but replacing S (t) R (t).

p0042(4) is carried out a cross-correlation distributions of the amplitude maxima D (i) and D '(i). If the maximum amplitude E of the result of intercorrelation between D (i) and D '(i) is sufficient, then j is the considered area of ​​the activated number. Otherwise the signal generated by the sensor corresponds to a false alarm.

p0043During this step (4), one can simply calculate E and the maximum value of D (i), or Max (D (i)): if one considers these two values ​​as the coordinates of a point in a dimensional space axes x = Max (D (i)) and y = E, as shown in Figure 3, one can determine in advance (empirically or by calculation) a threshold curve L which defines an area D corresponding validated in points (this domain is finite and limited at x = l and y = l, absolute maximum values ​​of D (i) and E. the signals which give points outside the domain D, meanwhile, are eliminated as false alarms.

p0044In the example, the D line is a line that can pass for example through the points (SI, 0) and (0, S2). For example, Sl = S2 = 0.4 and 0.4 or 0.6.

p0045Note that in addition to identifying the active zone 10 where does the impact, it would be possible to measure the force of impact, for example to guide the user in how to use the interface acoustic, or to modulate the action that affects an active area 10, depending on the intensity of this impact.

p0046Note also that the recognition signals from the active zones 10 may optionally be done using only a part of the signals S (t) received or a portion of their frequency spectrum, or more generally a portion of their characteristics. In this case, during the recognition step, processing the sensed signal to extract data representative of certain characteristics of the sensed signal and comparing the data thus extracted to the reference data extracted from the signal that is sensed when an impact is generated on each active zone.

p0047Thus, it is for example possible to measure the amplitude and the signal phase for predetermined frequencies m (m being a natural integer at least equal to 1), and to compare these measured amplitudes al-am and the pl-pn measured phases with Garlic-Aim amplitudes and phases measured Pil-Pim said predetermined frequencies from the signals received during the learning phase (or determined by modeling) for the different active zones 10 of number i (i being between 1 and n, where n is the number of active areas 10).

p0048Alternatively, it is possible to determine a code from said data extracted from the sensed signal and compare this code with a table which gives a correspondence between at least certain codes and each active zone (the codes contained in this table then represent the signals the aforementioned predetermined signals from the library). By way of non-limiting example, one can determine a 16-bit code from the sensed signal S (t), as follows:

p0049- The first 8 bits of the code are determined from the frequency spectrum of the signal S (t) that is subdivided into 8 predetermined frequency tranches [f_k, ₊i] _t k = 1..8: the bit of rank k is 1 for example if the final energy value given by the spectrum at frequency f_{k +}ι is greater than the average energy value of the acoustic wave in the frequency range [f, f_{k +}i] _r and this bit is 0 otherwise;

p0050- The last 8 bits of the code are determined from temporal signal S (t) which is divided into 9 predetermined temporal tranches [t_k, t_{k +}ι], k = l ..9: the bit of rank k + 8 is equal to 1 for example if the average value of the signal power during the time interval [t_k, t_{k +}ι] is greater than the average value of the signal power during the time interval [t_{k +}χ, t_{k + 2}], Louse k = 1..8, and this bit is 0 otherwise.

p0051In this particular example, the codes of the correspondence table would be determined during phase learning, calculated as above indicated the codes that correspond to signals picked up by the acoustic sensor 6 when generating impacts on different active areas 10. Furthermore, as shown in Figures 4 and

p00525, it may be possible to use two acoustic sensors 6 (SENS.l and SENS.2), especially when the object 5 has symmetries such that it can be a likelihood of confusion between the signals from two active areas 10 different. If necessary, we could use more than two acoustic sensors 6, although the preferred solutions make use one or two sensors 6.

p0053When two or more sensors are used, there are two options: 1) mixture of the various sensor signals and processing the global signal according to the method described above.

p00542) or, preferably, individual treatment of various sensor signals with the method described above and cross-checking the results: if the active zones 10 determined from the various sensors have identical numbers then it is determined that the area that has received an impact is it, in other cases, it can be considered the received signal as a false alarm, or determine the area that received an impact for example by cross-correlation between the functions of correlation Ci (t) determined for each sensor, or by more complex methods such as neural networks or other. Note that the two acoustic sensors may be of different types and / or capture of different sizes and / or their signals may be processed differently to identify the active areas 10 receiving impacts. For example, one of the acoustic sensors can be used to record the signal S (t) received while the other can be used only to determine a time difference between the arrival of the acoustic wave on the two sensors.

p0055The second sensor could not otherwise capture the acoustic wave propagated in the solid object 5, but the acoustic wave propagated in the air when one impact.

p0056As shown in Figure 6, the object forming an acoustic interface may consist of a computer screen 3 or a television screen to which the sensor is fixed 6. The surface receiving the impacts may advantageously be the window 15 of the screen this may allow in particular to display the screen by 3 delimitation of the active zones 10 and their meanings. This variant would be used for example to program a VCR, particularly in the case where the screen 3 is a television monitor (the central processing unit 2 would be replaced by the VCR).

p0057As shown in Figure 7, the object forming an acoustic interface may also be constituted by a glass door 16 or the like. In the example shown in Figure 7, the surface 17 which carries the active areas 10 is constituted by the window area of ​​the door, and always in the particular example shown in this figure, the acoustic sensor 6 is attached to a wooden part of the door 16.

p0058In the example shown in FIG 8, the object forming an acoustic interface is a 18 tablet specifically designed to serve as an acoustic interface. This tablet may for example include a rigid frame 19 secured to a rigid 20 background also bringing the acoustic sensor 6.

p0059A flexible membrane 22, made for example of elastomer, is stretched over the frame 19 a short distance above the bottom 21. The flexible membrane 22 is provided rigid pins 23 under its lower face (it may be for example glass hemispheres that are glued under the membrane 22). Thus, when a user types on the membrane 22 and including an active area 10 carried by this membrane, this action generates an impact at least one pin 23 of the bottom 21 of the shelf frame 18. This alternative presents the advantage of producing relatively little impact dependent on how the user types on the membrane 22 (with finger or fingernail or tool, with more or less force, etc.).

p0060In the embodiments of Figures 6 to 8, the method used may be identical or similar to that described above and allow to match an impact generated on the surface of the object forming an acoustic interface, either with an active zone 10 or with no active zone.

p0061But it is also possible in all embodiments of the invention using multiple active surfaces (possibly one-off), to determine the position of the impact on the surface 9 of the object 5 forming an acoustic interface (see the example in Figure 9), even when this impact is not on active areas. an acoustic interface is thus obtained continuous or pseudo-continuous (for example by an operation similar to a computer mouse, a light pen, a touch screen or similar).

p0062In this case, during the recognition step:

p0063- Determining representative similarity values ​​of the resemblance between the sensed signal and the predetermined signals (including values ​​from functions intercorrelation Ci (t) above, for example their maxima D (i) defined above),

p0064- The impact is associated with a number p of at least two adjacent active zones corresponding to a maximum resemblance, called reference active zones R1-R4 (p may be advantageously 4 in order to reinforce the impact in two dimensions x, y, or where applicable less than 4 particularly if the impact should only be positioned as a single dimension x or y): it is for example first determining the RI area as the active area 10 having the resemblance value D (i) maximum, and then determine which of the adjacent active areas Ri, the three areas R2 R4 which give the highest values ​​of the resemblance value D (i)); then determines the position of the impact I on the surface 9 according to the resemblance values ​​D (i) assigned to the reference active zones R1-R4. During this last step, it is advantageous to determine the position of the impact on the surface so that the resemblance values ​​attributed to the reference active zones correspond as well as possible to theoretical resemblance values ​​computed for said active areas reference for an impact generated in said position on the surface.

p0065These theoretical resemblance values ​​may be functions of the position of impact on the surface, determined in advance for each possible set of reference active zones.

p0066These functions may be determined during the learning step, for example by adjusting a deviation function of the resemblance values ​​of the active areas between them. The function-type in question may be dependent on the shape of the object and be determined in advance, either theoretically, or experimentally.

p0067To take a concrete example, the theoretical resemblance functions Rth (X, Y) between two points X, Y of the surface 9 may correspond to the maximum of the function of cross-correlation between the signals Sx (t) and Sy (t) captured by the sensor 6 respectively when impacts are generated at these two points X, Y, and this function can for example be of the type Rth (X, Y) = (sin (a (β) · d)) / (a ​​(β ) .d), approximated for example by Rth (X, Y) = 1- [a (β) .d] ²/ 6, where: - d is the distance between X and Y, β is an angle between for example the x axis (or y axis) and the direction XY, and (β) is a coefficient depending on the angle β in an elliptical function: a (β) = al.cos (β + β0) + a2.sin ((β + βO) where βO is an angle representative of the orientation of one ellipse.

p0068One can determine the function Rth for each possible set of active zones of R1-R4 references, from the Ri predetermined signals (t) of the library of signals, sensed by generating impacts respectively on the reference active zones during the learning phase.

p0069To this end, for a given set of four reference zones R1-R4, one can calculate the maximum of the function of cross-correlation of the signal R (t) corresponding to

p0070RI, with each of the signals R2 (t), R3 (t), R 4 (t) corresponding to R 2 -R 4 areas. We deduce the values ​​al, a2 and βO. We can then do the same from the reference zones R2, R3 and R4, which gives every time al values, a2 and βO, then take the average of the four values ​​thus found respectively for al, a2 and βO: these mean values ​​then determine the function Rth for all of R1-R4 reference areas. Alternatively, the function Rth could be determined by an iterative optimization process (type method of least squares) to minimize an error function between the function of theoretical resemblance and the maxima functions of intercorrelation between the signals R (t), R2 (t), R 3 (t) and R 4 (t) taken in pairs.

p0071Once determined the theoretical resemblance functions Rth above, when trying to determine the position of an impact I between four adjacent active zones R1-R4 (preferably point), this position can for example be determined by an iterative process optimization by minimizing an error function between D (i) values ​​defined above (D (i) ≈Max (Ci (t)) i here being the number of the active zone of reference Ri considered) and the theoretical resemblance values rth (I, R). For example, it can minimize an error function E equal to the sum of the values ​​(D (i) -Rth (I, Ri))². The process just described does not of course limited to the examples described above; It has many applications, including: the use of glass or other surfaces as an input interface 4, in shops, museums, art galleries, or others to allow customers or visitors are introduced details on a screen or via a loudspeaker concerning exhibits or works,

p0072- Use of glass or other surfaces of display panels as input interfaces 4, allowing passers-by to be present for example, details on current display advertisements, or to be present general information about a city or another location (for example, news and practical information, such as a map of the place), or other such details or information being presented for example on a screen visible at the bottom of the display panel, the use of parts of the walls, floor, or any other object as an input interface 4 for example to control home automation systems (thus may notably enable the inhabitants of an apartment of themselves determine the locations of switches, simply consist of the active 10 areas mentioned above, positioned on the walls or other the desired locations), use parts of the walls, floor, or other object as input interface 4 for example to control industrial machinery especially in hostile environments (areas with explosive, high temperature places, places with high radioactivity, etc.) the use of smooth and easy to clean surfaces as input interface 4, to form - domestic objects input keyboards such as refrigerator, washing machine or other, the use of door panels of buildings as input interfaces 4, constituting such virtual keyboards to digital code, the use of land to locate the position of a person walking on the realization of keyboards or control panels unaffected by pollution, weather or other external factors in industrial, military and even domestic applications (or the acoustic sensors can optionally be fully integrated to the object which serves as an input interface, in particular in the case of an object at least partially molded in plastic material) ; when these input interfaces must control a device (eg, a microcomputer) comprising a display screen, keyboard or acoustic control panel may be formed by the screen itself or a transparent wall covering that screen. the realization of input interfaces in automobiles or other vehicles. Note also that the 4 previously described input interface could be provided with processing means for locally performing the recognition of acoustic signals S (t) from the active zones 10, the input interface 4 then sending directly to the CPU 2, or any other electronic device user, only coded signals indicating directly which active zone 10 has been touched by the user and, where appropriate information on the impact: the severity of impact and nature of 1 impact.

p0073Note that the method of the invention does not require that the object 5 has a uniform or predetermined structure, or be made with particular care, or is done with very precise dimensions, or specific surface finish. On the contrary, the object 5 is more heterogeneous and / or irregular, more acoustic signals emitted by the different active zones 10 will be different from each other, the better the recognition of acoustic signals. One can even in some cases intentionally create heterogeneity such as cavities or other in the object 5 to facilitate the recognition of acoustic signals from the active areas 10. Furthermore, when the predetermined signals from the library of signals are determined during a learning phase, it is possible to use a piezoelectric sensor connected by any means known to the central unit 2 and attached either to the finger of the user or to the object (or other stylus) used to generate impacts on the active areas of the object 5. in this case, the pulse signal generated by the piezoelectric sensor at each impact can be used to trigger the acquisition of the predetermined acoustic signal intended to feed the library signals, and / or to measure the intensity of the impact, the intensity measurement can be used for example to invalidate certain acquisitions of predetermined signals especially when the intensity is less than a predetermined threshold or when the intensity is not within a predefined interval.

p0074Moreover, when the predetermined signals from the library of signals are determined during a learning phase, it may be advantageous to retain only the sensed acoustic signals whose amplitude is above a first relatively high threshold reference . In this case, during normal operation of the device can then take into account the acoustic signals whose amplitude exceeds a second predetermined threshold significantly below the first threshold. The first predetermined threshold may thus be e equal to several times (at least two to three times) the time average value of the absolute magnitude of the ambient noise, measured, for example a few seconds, while the second predetermined threshold may for example be 1.5 times this mean value. In this way, it records only good quality reference signals during the learning phase, while maintaining high sensitivity of the device during normal operation. Where appropriate, the central processing unit 2 may be provided with an auxiliary programming keyboard (not shown) which may be used in particular during the learning phase to indicate for example the type of signal is generated. The type of generated signal may in particular be one of the following types:

p0075- New signal replacing one of the reference signals of the library of signals (identifying the replaced reference signal can thus be transmitted to the central processing unit 2 through the auxiliary keyboard), new reference signal (either for a preexisting but incomplete reference library or to a new reference library corresponding in particular to new conditions of temperature, humidity or state of the object 5)

p0076- New signal for checking an existing reference signal in a signal library.

p0077Moreover, when the predetermined signals from the library of signals are determined during a learning phase, it may be provided if necessary with validating the reference signals of this library when they are confirmed by generation of one or more impact (s) on the same active area, in a predetermined period of time following the generation of a first impact.

p0078When the predetermined signals from the library of signals are committed to being a learning phase, the impacts generated on the object 5 during this learning phase can be generated - either with a hard object such that a stylus, in which case the same stylus is preferably used during normal operation of the device,

p0079- Either with a damping object such as eg a hard plastic rubber attached to the end of a pen or similar (the inventors were able to get good results with a transparent hard plastic eraser to mark "Staedler") in which case the impact on the object 5 can then be generated as well with relatively hard objects with less hard objects (finger nail, finger pulp or other) during normal operation of the device.

p0080Moreover, alternatively the method described above to recognize the active zone 10 from which comes the sensed signal S (t), it is possible to use the following method: (1) during the learning phase, calculating the Fourier transform Rι (ω) of each acoustic signal R (t) generated by an impact on the active zone i, where i is an index lying between 1 and n: Ri (ω) = | Ri (ω) | . e^{j φi (ω)}

p0081It retains this Fourier transform only the phase component, only in the frequency bands ω in which the amplitude of the spectrum is greater than a predetermined threshold. The frequency shaping of the stored reference signal is thus expressed in the form R '(ω) = e^{φl j (ω>} for frequencies ω which | Ri (ω) | is above the predetermined threshold, and R 'i (ω) = ω 0 at the other frequencies.

p0082The predetermined threshold in question may for example be equal to the maximum of MAX / D and | B (ω) | , or :

p0083MAX can be either the maximum value of

p0084| Ri (ω) |, the maximum value of the modules | Ri (ω) | each normalized in energy, the maximum value of the envelope of the average of the modules | Ri (ω) | each normalized in energy,

p0085D is a constant, for example equal to 100, | B (ω) | is the average of several noise spectra in the object 5, acquired at different times.

p0086(2) During normal operation of the device, each sensed signal S (t) undergoes the same treatment as in step (1) above, so that a signal S 'is obtained (ω) = e^{jψ (ω)} for frequencies ω which | S (ω) | is higher than the aforesaid predetermined threshold, S '(ω) is equal to 0 at the other frequencies. (3) then calculates a product Pi (ω) equal to

p0087S '(ω) multiplied by the conjugate of R' (ω) for i = 1 ... n.

p0088(4) were normalized products Pi (ω) by dividing them by their integrals.

p0089(5) is then carried out the inverse Fourier transform of all the products P i (ω), and is obtained temporal functions Xi (t).

p0090According to the different functions Xi (t) and in particular of their maximum, it can then assign the signal S (t) to one of the active areas 10. For example, one can calculate the maximum value (relative value or absolute value), different functions Xi (t), and assign the signal S (t) to the active zone i which corresponds to the function Xi (t) with the highest maximum. Optionally, one can also compare the maximum of the Xi (t) retained with a threshold set in advance, for example equal to 0.6, and decide that the signal S (t) must be allocated to zone i when the maximum of Xi (t) is greater than this threshold (if several functions Xi (t) have their maximum greater than 0.6, it is then retains the Xi (t) of greatest maximum).

p0091It is possible optionally to check that the assignment of signal S (t) to the active zone i is correct, for example by calculating a value M = M i / M where M is the maximum of the absolute value of Xi (t) and M is the average value of all values ​​Mi. the assignment of the signal S (t) to the active zone i can then be considered correct if MMi value exceeds a certain limit, for example equal to 1.414.

p0092Note also that the above MMi values ​​can be calculated by replacing S '(ω) by R' (ω), so as to obtain information on the spatial resolution of active areas. In particular, one can thus verify that an active area of ​​index i is not likely to be confused with another, checking the corresponding MMi value exceeds a predetermined limit, for example greater than 1.414.

p0093Furthermore, it is also possible to take into account different environmental parameters (temperature, humidity, mechanical stresses, etc.) by modifying the predetermined signals from the library signals function or environmental parameters.

p0094For this purpose, one may use one of the following correction methods: linear expansion or contraction of time reference signals from the library of signals: in this case, the reference signals Ri (t) of the signal library are replaced by signals Ri (± o), or is a positive non-zero multiplier which is a function of environmental parameters, this coefficient may be determined theoretically or experimentally for a given material, or experimentally for each object 5 ; expansion or contraction of linear time^• Sensed signals S (t): in this case, the reference signals Ri (t) are left unchanged, but the sensed signal S (t) is replaced by S (αt), where is a coefficient as defined above; nonlinear dilation or contraction in frequency of the reference signals: in this case, replaces the frequency signals R '(ω) by R' (ω), where ω with_NOT is equal to half the <img id="imgf000030_0001" he="13" wi="49" file="imgf000030_0001.tif" img-format="tif" img-content="drawing" orientation="portrait" inline="no" /> sampling frequency of the processor, and β is a coefficient determined theoretically or experimentally; - Dilation or nonlinear contraction frequency of the sensed signal S (t): in this case, the library of signals of the reference signals are left unchanged, and the signal S '(ω) is replaced by S' (ω ') , ω 'being defined above. In both the above cases of nonlinear dilation or contraction in frequency, it is possible also to use an averaged phase correction, in which case the signals R (ω ') are replaced by Ri (ω') .M '( ω) / N '(ω) or the signals S (ω') are replaced by S (ω ') .M' (ω) / N '(ω). In either of these formulas, N '(ω) = M (ω) / Im (ω) I, and^' N '(ω) = (ω) / | N (ω) | , M (ω) is equal to the average of all Rι (ω) and N (ω) is equal to the average of all Ri (ω '). The various corrections mentioned above Ri reference signals (ω) or the sensed signal S (ω) can be made either automatically by the central unit 2, in particular as a function of information from one or more sensors (not shown) , or manually by the user.

p0095Furthermore, note that the CPU 2 may comprise several libraries of reference signals adapted to different values ​​of the environmental parameters.

p0096Furthermore, to adapt to the types of impacts generated during use of the device, and in particular to accommodate the use of either a user's finger or other object to generate the impacts, it may be advantageous to request the user to generate impacts on one or more predetermined active zones 10, for example two active zones of indices m and p. two time signals thus captures S_m(T) and S_p(T) whose Fourier transform is calculated S_m(Ω) and S_p(Ω), then the average is calculated Mi (ω) of the following two terms: - (R_m(Ω). | S_m(Ω) | ) / (| R_m(Ω) | .S_m(Ω)),

p0097- And (R_p(Ω). | S_p(Ω) |) / (| R_p(Ω) | .S_p(Ω)). This average Mχ (ω) is then used in step (3) defined above to replace the product Pi (ω) by i (ω). P ± (ω), this product is then used in place of Pi (ω) in step (4).

p0098Moreover, it is noted that the invention allows a user to define at will the active areas, and the central unit 2 may be adapted to maintain active the definition of the active areas only during the actual use of the object 5 as interface acoustic. In this case, the above definition of the active areas is cleared by the CPU 2 after a period of inactivity of the device.

p0099Note also that the function generated by an impact on an active area can be adjusted as appropriate depending on the intensity of impact.

p0100Note also that, when the object 5 has resonance phenomena which cause prolonged acoustic wave propagation at each impact on the active regions, it may be advantageous to raise the detection threshold of the acoustic signals S (t) ( for example up to 0.5 times the maximum permissible value by the electronic system of acquisition of signal S (t)) where a signal S (t) has been detected, then belittle this detection (including exponentially ) to its normal level; thus, we avoid multiple detections of the same impact.

p0101Note that in all of the invention embodiments, it might be possible to define a single active area on the object 5, in which case it is nevertheless possible to encode multiple functions on this single active region, for example following the number of impacts generated consecutively to the same area.

p0102Furthermore, the active region 10 may optionally not be defined in advance, but be simply defined in terms of successive impacts received when using the device. Thus the device can for example be designed to include three active zones, which are each defined simply on receipt of a first impact on each area, and then are recognized as "first area", "second area", and " third zone ", on receipt of the following impacts.

p0103Moreover, when the active regions are very numerous, it may be appropriate to use an automated device for generating the reference signals stored in the library of signals during the learning phase. The automated device may for example comprise a two-dimensional displacement system, comprising two stepper motors, to move for example an excitation stylus or the like on the surface of the object 5 and to generate impacts by means of this stylus, for example actuated by a solenoid, at the various active areas.

p0104Still in the case where the active zones 10 are very numerous, it may be possible to divide them into several groups of resemblance. In this case, when the current use of the device, when an impact generates a signal S (t), a first treatment is used to attach the signal S (t) to one of the similarity groups, then a refined processing to assign this signal S (t) to one of the active areas of the similarity group.

p0105Note also that the same central processing unit 2 could where appropriate process signals from multiple objects 5. Moreover, it is also possible to directly interface or the acoustic sensors with a particular IP network, so as to direct the signals captured to a single IP address from which these signals can be operated by n any computer connected to the IP network.