DE10241328A1 - Biometric acoustic writing system, e.g. for computer input, has data processing unit for computing biometric data depending on detected sound signals caused by hand-guided pen movement on substrate - Google Patents

Abstract

The biometric acoustic writing system (1) has a pen housing (31) for carrying out hand-guided movements on a substrate (4), at least one microphone (5) integrated into the housing for acoustic detection of sound signals caused by the movements of the pen and a data processing unit (11) for computing biometric data depending on the detected sound signals. Independent claims are also included for the following: (a) a method of generating person-specific biometric reference data (b) a method of verification and identification of a person (c) and a method of handwriting recognition.

Description

The invention relates to a biometric writing system, in the case of the microphone integrated in a cursive casing with the hand-held Sound signals caused by movement of the stylus acoustically captured and then processed into biometric data. The invention relates also a procedure for personal identification and personal verification using a biometric writing system and a method for Handwriting recognition and / or hand sketch recognition.

There is currently a large selection of competing biometric systems. The biggest part of biometric systems currently make up systems, the finger lines as recognition features, followed by facial recognition systems and so-called iris scanners.

Due to the increasing electronic Networking in all areas of business life takes on importance electronic password and signature recognition. contracts are in more conventional Way closed by the two contractors by a signature is performed for certification. The signature is used for this Evidence purposes and is for the one who makes the signature specifically. When concluding contracts in In the field of e-commerce there is a need for the contractor provide an electronic signature to identify them have to, on their validity must be verified. Prerequisite for the widespread use of a electronic signature to verify the identity of the contractual partner is a high security against falsification the electronic signature. Furthermore, the performance Signature recognized with a high recognition rate and sent to one remote computers become. imperative is also that the entry of the electronic signature for a huge People in a simple manner without special knowledge can.

The previously known writing systems fulfill these requirements do not exist and some have serious disadvantages on.

They are pens with accelerometers known (e.g. accelerometer-based pens from British Telecommunications) to absorb acceleration forces when writing with the Pen. The disadvantage of this known pen is in a comparatively low recognition rate for signature verification due to the only two-dimensional data space. Furthermore the pen must always be aligned when writing, i.e. the angle between the pen and the base must be be largely constant, so that a natural writing behavior when Providing a signature is not guaranteed. Furthermore an acceptable recognition rate sets a minimum writing speed ahead, otherwise the measured acceleration forces are too low are.

A variety of writing systems recognize the handwritten text or characters using the desk pad. All such writing systems have the disadvantage that special writing pads with appropriate signal processing required are.

From the US 4,513,437 a data entry pen for signature verification is known. The pen contains a writing pen to visualize what has been written on a base and piezoelectric pressure and acceleration sensors for detecting acceleration forces which are caused by transverse writing movements of the pen on the base. In addition, the pressure on the pen is recorded in the axially parallel direction of the pen.

The disadvantage of that in the US 4,513,437 described data entry pen is that pronounced acceleration forces are required when writing. Strong changes in direction and / or changes in speed of a straight-line write movement for signal detection are therefore necessary for the detection. Also the US 5,181,661 describes a pen with acceleration sensors for detecting the acceleration of the pen when writing in three directions in space.

The so-called Anoto digital pen describes a writing system in which written text is transferred to a computer can be. The pen consists of a ballpoint pen an integrated digital camera. A force sensor detects that Put the writing on a special paper. With the special paper it is a paper with a printed dot pattern. The integrated digital camera of the Anoto pen takes that Writing movements of the pen over the dot pattern true. The camera takes a variety of pictures per second. Each picture taken is followed by the next Image compared and an integrated processor calculates from it considering the direction of the pen and the direction of movement of the writing pen integrated in the pen. On the Anoto special paper there are a lot of printed dots with ink are applied on a carbon basis and a distance of less than 1mm to each other. Each Rnoto special paper has a different dot pattern, so that the individual pages can be distinguished from one another. A light emitting diode (LED) illuminates the carbon-containing ink dots, which absorb long-wave infrared light and thus for the integrated Digital camera are recognizable within the Anoto pen. Thanks to the special Anoto paper can the relative movement of the pen as well as the absolute position of the Pen can be determined on paper.

The disadvantage of the Anoto pen is that it requires special paper and that Pen posture is specified. In addition, the security against tampering with signature verification is too low, since no person-specific biometric data are determined. The Anoto pen does not record the individual writing pressure on the writing surface.

In addition to electronic signature verification there is an increasing need for electronic capture of handwritten Texts or characters. With conventional Handwriting recognition methods are hand-guided writing movements optically recorded and the determined typeface data are then by means of OCR procedure converted into ASCI code data. traditionally, become several possible word candidates when a written word is recognized selected. The number of possible Word candidates is the higher, the more indistinct the handwriting is and the more similar words exist. Out the different possible Word candidate is then through a statistical evaluation of the most likely word candidate determined and output. The statistical evaluation is done like in machine speech recognition using so-called word sequence statistics programs, the statistical probability based on the previously recognized words for the Calculate the occurrence of the respective word candidate. The most likely to occur Word candidate is selected and then output. The disadvantage such more conventional The handwriting recognition process is based on the number the possible Word candidates in many cases is very high, so the detection rate is often very low.

The German patent application with the official file number 102 04 962.9 with the filing date of February 6th 2002 describes an optomechanical writing system and a process for handwriting recognition using biometric data. With this optomechanical writing system is in a pen housing Pen to carry out hand-held Movements integrated on a base. The pen housing contains an optical one Sensor device for detecting the movements carried out with the pen as movement or position data. In addition, one is in the pen housing further sensor device for detecting the movements carried out forces occurring in the pen integrated. A data processing unit is used to calculate biometric data depending on the data collected Movement data and the recorded writing pressure data.

Writing systems for signature verification, for personal identification or verification as well as for manuscript and speech recognition need a very high security against falsification and have low error rates. In addition, the technical Realization not too complex to enable a wide range of applications.

It is therefore the object of the present invention to create a biometric writing system that has low error rates has and a very high security against falsification offers.

This object is achieved by a biometric writing system with the specified in claim 1 Features resolved.

The invention relates to a biometric writing system with a pen housing to carry out hand-held Movements on a base, at least one integrated in the pen housing Microphone for acoustic detection of sound signals through the hand-held Movements are caused, and with a data processing unit for the calculation of biometric data depending on the recorded sound signals.

The biometric writing system according to the invention allows the generation of person-specific biometric reference data, the for the person are unique. This makes it possible to identify a person Verify writing person and by comparison with saved Even identify reference data. The biometric according to the invention Writing system enables also a handwriting recognition or hand sketch recognition for Identification and output of handwritten characters in digital Shape.

The biometric writing system according to the invention contains preferably a data processing unit for the reconstruction of handwritten Signs and texts from the recorded sound signals.

In the pen housing is preferably a pen is provided, which is placed on the base and performed on the pad becomes.

This offers the particular advantage that the person is a natural writing feel has, since the biometric Writing system behaves like a normal pen.

This is due to the friction of the pen on the pad during the hand-held Acoustic writing noise is caused by the writing movement preferably as a structure-borne sound signal over the Pen and / or additional as an airborne sound signal the ambient air through the integrated in the pen housing Transfer microphone.

Since the writing noise is not just about the ambient air, but also transmitted as structure-borne noise the writing system according to the invention is also suitable for areas of application in acoustically noisy areas.

The microphone is preferably for transmission of the structure-borne noise signal mechanically coupled to the pen.

The microphone is at one particularly preferred embodiment for transmission of the structure-borne noise signal mechanically coupled to an orchestra connected to the pen.

The sound body is preferably as a Resonator for certain natural frequencies designed.

In a preferred embodiment of the biometric according to the invention Writing system, the microphone is provided in the pen housing Airborne sound chamber arranged.

The airborne chamber is preferred as a resonator for certain natural frequencies designed.

The microphone is preferably from a soundproofing surround that for damping of ambient noise is provided and sound signals from the sound body and the airborne sound chamber pass through.

The airborne sound chamber is over a in the pen housing existing housing opening with the ambient air can be coupled.

The housing opening is preferably by means of a mechanical closure device to suppress External noise can be closed.

This gives the user the opportunity to go through Conclude the opening Noise to suppress the environment.

The microphone detects when open Housing opening preferably through the hand-held writing movement caused internal and external writing noise as a structure and airborne sound signal and / or a speech signal from a person.

Through the acoustic detection of the Speech signal additionally The error rate can be used for acoustic recording of the writing noise be significantly reduced when the person is recognized. Through the Combination of the evaluation of the speech signal and the writing noise a very high recognition reliability of the biometric writing system reached.

The microphone converts the captured acoustic sound signals preferably in an electrical sound signal around.

The electrical sound signal will preferably by an analog / digital converter in sound signal data for digital data processing by the data processing unit converted.

The writing noise and the speech signal simultaneously or in succession captured the microphone.

The sound signal data generated are preferably in a storage unit of the biometric writing system stored.

The writing pad preferably consists of any paper.

The writing pad is preferably one solid base, which has a certain impressed surface roughness and hardness.

In a particularly preferred embodiment of the biometric writing system is in the pen housing Loudspeaker for playing back recorded microphone signals, for playing back stored ones biometric reference data and for the reproduction of spoken information intended.

In a preferred embodiment of the biometric writing system is an additional optical sensor device provided the position data of the hand-held pen movement via image signals from the surface of the writing pad.

In a further preferred embodiment of the biometric writing system is also a pressure sensor device provided the static and dynamic pressure of the on the pad patch and hand-guided Pen captured.

The microphone is preferably a Electret microphone, a piezoelectric microphone, a piezoresistive Microphone or a capacitive microphone.

In the pen of the biometric writing system an exchangeable pen lead is preferably provided.

This pen lead is preferred extendable from the pen by means of a switching device.

In a preferred embodiment of the biometric according to the invention Writing system, the optical sensor device has imaging optics for mapping the document surface and a converter device for converting the optical imaging signal into an electrical one Signal on.

The imaging optics exist preferably made of optical lenses and / or glass fibers.

Acting in the converter device it is preferably a CCD camera or a photodiode array.

Furthermore, a diode light source is preferred provided for illuminating the surface of the underlay.

In a particularly preferred embodiment is additional a laser diode for illuminating the underlay surface and of an optical grating integrated in the pen.

The optical sensor device detects preferably the movements carried out with the pen by comparison the temporally offset recorded image sequences of the document surface.

In an alternative embodiment the optical sensor device detects that with the pen conducted Movements by comparing the staggered, recorded Image sequences of the interferences of the reflected on the grating and the underlay surface Laser light.

In a preferred embodiment, the pen lead, in an extended state, releases a writing liquid onto the base when the hand-guided movements are carried out, which increases the structuring of the surface of the document, so that the optical detection of the hand-guided movements by means of the optical sensors direction is facilitated and a natural spelling is created via the ink typeface on the base.

The data processing unit is preferably in the pen housing or in an external one Integrated receiver unit.

With the external receiver unit it is preferably a computer, a microcontroller, a cell phone, a credit card reader, a fax machine or around a printer.

The recorded sound signals, the optical movement data and the recorded multidimensional print data are preferably about a data transmission link from the pen housing transmitted to the data processing unit integrated in the external receiving unit.

The biometric writing system according to the invention preferably has an encryption unit contained in the pen housing for encryption of sound signal data, optical movement data and print data.

The data transmission path is included preferably wired or wireless.

The biometric writing system according to the invention preferably has a data memory for storing biometric Reference data, position data of the writing movement and spoken Information on.

The biometric reference data is preferred through the data processing unit from when writing and speaking sound signal data recorded from a word, from optical movement data and mechanical pressure data are calculated and stored in the reference data memory stored.

The reference data store is concerned it is preferably a microchip of an identity card, a credit card, an authorization magnetic card or an Storage unit of a computer or the biometric writing system.

For the written and spoken words and Characters are preferably pin codes, passwords, names or about texts.

The position data of the write movement will preferably from the sound signal data recorded when writing a word, optical movement data and mechanical pressure data are calculated and stored in the data memory for handwriting recognition.

The reference data storage is included preferably integrated in the pen housing.

The in the external receiving unit integrated data processing unit is preferably to one Reference data memory connected.

The external receiving unit points in a preferred embodiment of the biometric writing system a reading unit for reading a portable of biometric reference data.

The external receiving unit is preferably via a Data network to a database with a reference data store connected.

The data network is preferably the Internet.

In a preferred embodiment of the biometric writing system compares the data processing unit the calculated current biometric data with the saved biometric reference data for their verification and identification.

The data processing unit generates if there is broad agreement the current biometric data with the stored reference data Identification or verification display signal.

In a particularly preferred embodiment of the biometric writing system according to the invention recognizes the Data processing unit with complete agreement of the current one biometric data with the stored biometric reference data the current biometric data as a pirated copy of the saved Reference data and generates a corresponding warning signal.

The data processing unit generates if the current biometric data deviate from the saved data biometric reference data a deviation indication signal.

In a preferred embodiment of the biometric writing system, this has at least one actuator after generating the identification or verification display signal is operated.

The invention also provides a method for generating person-specific biometric reference data with the following steps:
Acoustic detection of hand-held writing movements, which are carried out by a person when writing a character, a word or a sequence of words with a pen on a base, and generating corresponding sound signal data,
Storing the generated sound signal data as a digital sound-time signal,
Calculating an associated frequency spectrum from the sound signal data by means of Fourier transformation, determining first biometric reference data from the sound intensity of the digital sound-time signal by means of feature extraction,
Determination of second biometric reference data from the calculated frequency spectrum of the sound-time signal by means of feature extraction.

In a preferred embodiment of the procedure for the generation of person-specific biometric The procedural steps are carried out several times and reference data statistically evaluated the respectively determined reference data, whereby the evaluated data is encrypted as person-specific reference data and subsequently can be saved.

In a preferred embodiment, the statistically evaluated, person-specific stored reference data in a microchip of an identity card, in a microchip of a credit card, in an authorization magnetic card or in the storage unit of a computer or a writing system.

Preferably, an additional speech signal originating from the person is recorded acoustically and corresponding Sound signal data generated.

In addition, at a preferred embodiment of the method according to the invention for the generation of person-specific, biometric reference data the carried out at the Writing movement of the pen in at least one spatial direction occurring typists detected by pressure sensors and correspondingly at least one-dimensional Write print data generated.

Also preferred are those when writing a character, a word or a character / word sequence with the Pen on the surface of the hand-held writing movements optically recorded and generates corresponding write movement data.

The biometric reference data is preferred from the sound intensity the digital time sound signal by means of feature extraction, from which calculated frequency spectrum of the sound time signal by means of feature extraction as well as those generated when writing and speaking a word optical movement data and writing pressure data determined and in the Reference data memory is stored in encrypted form as person-specific biometric reference data.

The data is preferably reduced static and dynamic write movement data, the write pressure data, the optical movement data and the sound signal data by means of cluster formation.

From the data-reduced cluster data feature parameters are preferably extracted.

The generated biometric reference data are preferably inside or outside the pen stored.

The invention also provides a method for verifying a person with the following steps, namely acoustic detection of hand-held writing movements which are carried out with a pen on a base when the person is writing a character, a word or a sequence of words, and generation of corresponding sound signal data, Storing the generated sound signal data as a digital sound-time signal,
Calculating an associated frequency spectrum from the sound-time signal by means of Fourier transformation,
Determining first current biometric data from the sound intensity of the digital time-sound signal by means of feature extraction,
Determining second current biometric data from the calculated frequency spectrum of the sound-time signal by means of feature extraction, and
Comparing the current biometric data with stored biometric reference data of the person to verify whether the current biometric data largely agree with the stored biometric reference data of the person.

In this case, one of the speech signal originating from the person is recorded acoustically and corresponding Sound signal data generated.

The invention furthermore provides a method for identifying a person with the following steps, namely acoustic detection of hand-guided writing movements which are carried out by a person with a pen on a base when writing a character, a word or a sequence of words, and generating corresponding sound signal data,
Storing the generated sound signal data as a digital sound-time signal,
Calculating an associated frequency spectrum from the sound-time signal by means of Fourier transformation,
Determining first current biometric data from the sound intensity of the digital sound-time signal by means of feature extraction,
Determining current second biometric data from the calculated frequency spectrum of the sound-time signal by means of feature extraction,
Comparing the current biometric data with stored biometric reference data of a large number of people to identify whether the current biometric data largely correspond to the stored biometric reference data of one of the stored people.

It is preferably an additional speech signal originating from the person is recorded acoustically and corresponding Sound signal data generated.

The invention also provides a method for recognizing handwriting and / or recognizing sketches with the following steps, namely
acoustic detection of hand-held writing movements which are carried out with a pen on a base when a person is writing or sketching a character, a word or a sequence of words, and generating corresponding sound signal data,
Storing the generated sound signal data as a digital sound-time signal,
Calculating an associated frequency spectrum from the sound signal data by means of Fourier transformation, determining first current acoustic data from the sound intensity of the digital sound-time signal by means of feature extraction for data reduction,
Determining current second acoustic data for calculating the frequency spectrum of the sound-time signal by means of feature extraction for data reduction,
Compute a sequence of feature vectors the acoustic data,
Comparing the sequence of feature vectors with stored person-specific reference feature vectors, selecting a sequence of reference feature vectors which have the smallest deviation from the respective feature vectors,
Determination of the character, the word or the word sequence from the sequence of selected reference feature vectors using statistical, connectionistic and knowledge-based methods.

Preferably, an additional speech signal from the person acoustically detected, and it corresponding sound signal data are generated.

The one when writing or sketching a character, a word or a character / word sequence with the Writing pen on the pad will be hand-guided writing movements preferably additionally optically recorded and corresponding position data generated.

Preferably, the at the performed Writing movement of the pen in at least one spatial direction forces occurring detected by pressure sensors and corresponding, at least one-dimensional Write print data generated.

In a preferred embodiment of the method according to the invention for handwriting recognition and / or hand sketch recognition, the character the word or the word sequence from the sequence of selected reference feature vectors, from the position data and from the writing pressure data by means of statistical, connectionistic and knowledge-based procedures.

In a particularly preferred embodiment of the method according to the invention become the ascertained symbol of the person carrying out the control optically and / or acoustically reported.

The generated person-specific Reference data are preferably used to determine psychological and / or physiological characteristics of the person are evaluated.

The biometric writing system according to the invention can be used in many ways.

In a preferred embodiment becomes the biometric writing system as a computer input device used.

With an alternative use becomes the writing system according to the invention used as a voice input device, in particular as a dictation machine.

The biometric writing system according to the invention is preferably used to detect neuromotor movement disorders Person used.

The biometric writing system according to the invention is also preferably used as a therapy system for remedying neuromotor movement disorders used by one person.

The biometric writing system according to the invention is also preferably used as a graphological system for determination psychological / physiological characteristics of a person.

The biometric writing system according to the invention is preferably used as a training system for learning to write used.

The biometric writing system according to the invention is preferably used as a training system for learning a language used.

Furthermore, preferred embodiments of the biometric writing system according to the invention and the method according to the invention for personal identification, for personal verification, for generation biometric reference data as well as for handwriting and sketch recognition with reference to the attached Figures for explanation Features essential to the invention are described.

Show it:

1 is a schematic illustration showing a preferred embodiment of the biometric writing system according to the invention;

2 a sectional view through a pen housing of the biometric writing system according to the invention;

3 a sectional view through a pen housing with an optical sensor device according to a preferred embodiment of the biometric writing system according to the invention;

4 a flow chart for explaining the handwriting recognition by means of the biometric writing system according to the invention;

5 a sketch to explain an optical or acoustic feedback for the person writing;

6 acoustically recorded writing noises at different writing speeds on cardboard and writing paper;

7 an acoustically detected sound-time signal when the letter "a" is written three times;

8th an acoustically detected sound-time signal when the letter "b" is written three times;

9 Frequency signals when writing the letters "a" or "b" over time;

10 an amplitude frequency signal when writing an associated letter "a";

11 dynamic behavior of some frequency lines when writing the letter "a";

12 dynamic behavior of some frequency lines when writing the letter "b";

1 shows a schematic block diagram of the biometric writing system according to the invention 1 , The biometric writing system according to the invention 1 includes a pen 2 that is in a pen housing 3 is integrated. The pen 2 is used to perform hand-held movements on a surface 4 , With the pad 4 can be any document with be act any surface. It is preferably the pad 4 around paper. The desk pad 4 is preferably a solid base, which has a certain impressed surface roughness and hardness.

The pen 2 preferably contains an interchangeable pen lead, which uses a mechanical switching device, not shown, for the pen 2 is extendable. The pen case 3 is used to perform hand-guided movements on the surface 4 , It is used to convey a normal feeling of writing through the pen 2 and its pen lead a writing liquid on the pad 4 for optical feedback for the person writing. When putting on the pen 2 There are writing noises on the surface and its movement on the surface. In the pen housing 3 is in the writing system according to the invention 1 at least one microphone 5 provided for acoustic detection of sound signals. The microphone 5 detects the sound signals that are caused by the hand-held movements, ie the writing noise caused by the writing movement. Due to the friction of the pen 2 on the pad 4 Acoustic writing noises produced during the hand-held writing movement are, on the one hand, a structure-borne sound signal via the pen 2 and on the other hand as an airborne sound signal via the ambient air to the microphone 5 transfer. Here is the microphone 5 to transmit the structure-borne noise signal mechanically with the stylus 2 coupled. At the in 1 The preferred embodiment shown is the microphone 5 to transmit the structure-borne sound signal mechanically using a pen 2 connected orchestra 6 coupled. The orchestra 6 is preferably designed as a resonator for certain natural frequencies.

As in 1 shown, the microphone is located 5 in one in the pen housing 3 provided airborne sound chamber 7 , The airborne sound chamber 7 is preferably designed as a resonator for certain natural frequencies. The microphone 5 is from sound insulation 8a . 8b surrounded, which is intended to dampen ambient noise and only sound signals from the sound body 6 and the air switch chamber 7 pass through. The airborne sound chamber 7 is about one in the pen housing 3 Existing housing opening can be coupled with the ambient air. The housing opening is preferably by means of a mechanical closure device 9 closable to suppress outside noise. The microphone 5 detects the internal and external writing noise caused by the hand-held writing movement as a body and airborne sound signal and additionally a speech signal from a person when the housing opening is open. The microphone 5 converts the recorded acoustic sound signals into an electrical sound signal. The electrical sound signal is converted into sound signal data for digital data processing by an analog / digital converter. The digital signal data are sent via a signal line 10 a data processing unit 11 fed for further data processing. The writing noise generated during writing and the additional speech signal that may be received by the person are either simultaneously or in succession by the microphone 5 recorded and converted into corresponding sound signal data. That of the data processing unit 11 Sound signal data supplied are preferably stored in a storage unit provided for this purpose 12 that with the data processing unit 11 over lines 13 connected, saved.

Next to the microphone 5 as an acoustic sensor, this indicates in

1 shown, biometric writing system according to the invention 1 preferably additionally an optical sensor device 14 on, the position data of the hand-held pen movement via image signals from the surface of the writing pad 4 detected. The optical sensor device is via a signal line 15 also to the data processing unit 11 connected.

With the microphone 5 it can be, for example, an electret microphone, a piezoelectric microphone, a piezoresistive microphone or a capacitive microphone. The optical sensor device 14 preferably contains imaging optics for imaging the document surface and a converter device for converting the optical imaging signal into an electrical signal. The electrical signal is converted into position data or movement data by an analog / digital converter and via the signal line 15 to the data processing unit 11 issued. The imaging optics of the optical sensor device 14 consists preferably of optical lenses and / or glass fibers. The converter device can be a CCD camera or a photodiode array. A diode light source is preferably additionally provided for illuminating the underlay surface. A laser diode is also provided for illuminating an optical grating integrated in the pen. The optical sensor device 14 captured that with the pen 2 performed movements by comparing the staggered image sequences that show the surface of the base. Alternatively, the optical sensor device 14 also be designed in such a way that it is the one with the pen 2 performed movements by comparing the temporally staggered recorded image sequences of the interference of the laser light reflected on the grating and the surface of the document.

The pen lead is in the extended state on the surface 4 a writing fluid when performing the hand-held movements ab, which increases the structuring of the surface, so that the optical detection of the hand-guided movements by means of the optical sensor device 14 is facilitated. Furthermore, a natural spelling is brought about by the person writing on the ink typeface on the base.

The data processing unit 11 is like in 1 shown in the pen housing 3 integrated. In an alternative embodiment, the data processing unit can also be integrated in an external receiving unit. Such a receiving unit is, for example, a computer, a microcontroller, a cell phone, a credit card reader, a fax machine or a printer.

The through the microphone 5 recorded sound signal data and that by the optical sensor device 14 Provided that the data processing unit is located externally in a local computer, motion data are recorded via a data transmission link 18 from the pen housing 3 to this local computer 20 transfer. The data processing unit 11 is like in 1 to recognize through a line 16 to a downstream encryption unit 17 connected. The encryption unit 17 is used to encrypt the data processing unit 11 inside the pen housing 1 emitted sound signal data and the detected optical movement data. The encrypted data is transmitted over the data link 18 , which is wireless or wired, to a data processing unit 19 within the local machine 20 transfer. The local computer 20 has in addition to the data processing unit 19 through a store 21 and an ad 22 , each via signal lines 23 . 24 with the data processing unit 19 are connected. The data processing unit 19 of the local computer 20 is also on lines 25 with a reading unit 26 connected. In addition, the data processing unit controls 19 of the local computer 20 via control lines 27 an actuator 28 , for example a security door. The data processing unit 19 is over a line 29 with a data network 30 connected to an external database 31 connected. With the data network 30 it is preferably the Internet.

This in 1 illustrated biometric writing system 1 contains an acoustic sensor device 5 and an optical sensor device 14 , In a preferred embodiment, the biometric writing system according to the invention 1 additionally a pressure sensor device that measures the static and dynamic pressure of the surface 4 attached hand-held pen and corresponding print data to the data processing unit 11 emits. The recorded sound signal data, the optical movement data and the experienced multidimensional print data are processed by the data processing unit 11 after appropriate encryption by the encryption unit 17 over the data transmission link 18 from the pen housing 3 to those in the local machine 20 integrated data processing unit 19 transfer. The data processing unit converts the recorded sound signal data, the optical movement data and the mechanical writing pressure data into or outside the pen housing 3 located, biometric reference data is calculated and in a reference data memory, for example in the memory 12 or memory 21 , filed. In alternative embodiments, the reference data memory can be a memory within an identity card, a credit card, an authorization magnetic card or a storage unit of any computer. The written words and characters, the typing noises of which are recorded acoustically by the microphone and emitted as sound signal data, are in particular written pin codes, passwords, names or handwritten texts. The character written by the person can be spoken simultaneously by the person and as a speech signal also by the microphone 5 be recorded.

The one in the local machine 20 provided reading unit 26 is used, for example, to read a portable storage medium for biometric reference data. In the data processing unit 31 calculated, current biometric data can be compared with stored biometric reference data for verification and identification. If the current biometric data largely matches the stored reference data, the data processing unit generates 11 or 19 an identification or verification display signal. This is done on the data processing unit 19 after generating the identification or verification display signal, an actuator, for example a security door, is actuated.

Detects the data processing unit 11 or 19 a complete match of the current biometric data with stored biometric reference data, the current biometric reference data by the data processing unit 11 . 19 recognized as a pirated copy of the stored reference data and a corresponding warning signal is generated.

The stored person-specific biometric reference data are first saved by the biometric writing system according to the invention 1 generated for later comparisons. The person-specific biometric reference data is generated as follows. First, the hand-held writing movements are made with the stylus 2 performed on the pad when writing a character, a word or a sequence of words by a person, acoustically recorded, ie the writing noise is heard by the microphone 5 acoustically recorded. The microphone 5 generated according to Ana log / digital conversion of sound signal data to the data processing unit 11 be delivered. The sound signal data is stored as a digital sound time signal and an associated frequency spectrum is calculated from the stored sound time signal by means of a Fourier transformation. Characteristic extraction is used to determine first biometric reference data from the sound intensity of the stored digital sound time signal. Characteristic extraction is used to determine second biometric reference data from the calculated frequency spectrum of the sound time signal. These method steps are preferably carried out several times, and the respectively determined reference data are statistically evaluated, the evaluated data being stored in encrypted form as person-specific reference data. The statistically evaluated, person-specific reference data are preferably stored in a microchip of an identity card, a microchip of a credit card, an authorization magnetic card or the storage unit of a computer or writing system. In addition to the typing noise, the microphone 5 additionally acoustically record a speech signal originating from the person and corresponding sound signal data to the data processing unit 11 submit. Preferably, the optical sensor device additionally 14 the hand-held write movements are optically recorded and corresponding write movement data generated, which is sent to the data processing unit 11 be delivered. In addition, it is preferred to perform the writing movement of the pen 2 occurring typists via pressure sensors, not shown, and correspondingly at least one-dimensional writing pressure data to the data processing unit 11 issued. The data processing unit 11 determines the biometric reference data from the sound intensity of the digital time sound signal by means of feature extraction, from the calculated frequency spectrum of the sound time signal by means of feature extraction as well as from the optical movement data and writing pressure data generated when writing and speaking a word. Subsequently, the static and dynamic write movement data, the write pressure data, the optical movement data and the sound signal data are preferably reduced by clustering. Characteristic parameters are then extracted from the data-reduced cluster data.

2 shows a preferred embodiment of a pen of the biometric writing system according to the invention 1 , This in 2 pen housing shown 3 is an ordinary ballpoint pen case for example. In the pen housing 3 there is a pen 2 with integrated pen refill. The pen 2 is mechanical with a sound body 6 connected for the transmission of structure-borne noise signals. The recording of the writing noise through the microphone 5 is done using the stylus 2 and the sound body 6 , In an alternative embodiment, the pen 2 serve as a sound body itself. In addition to the typing noise, the microphone can 5 acoustically detect a speech signal originating from a person when the air switch chamber 7 or the acoustic window is connected to the surroundings by opening a closure device. The recording noise and the speech signal can be recorded simultaneously or in succession by the microphone 5 respectively. At the in 2 The preferred embodiment shown is the microphone 5 an electret microphone. At the in 2 The embodiment shown is additionally in the pen housing 3 a speaker 32 intended. The loudspeaker is used to reproduce recorded microphone signals, and the loudspeaker can also be used 32 be used to reproduce stored biometric reference data and to reproduce spoken information. With the speaker 32 it is preferably a miniaturized commercial loudspeaker.

When writing or moving the pen on the surface 4 there is an acoustic write signal or a write noise. The pen 2 can consist of a ballpoint pen refill, an ink pen or a stick with a defined tip. The underlay 4 consists, for example, of paper or a pad with natural and additionally imprinted surface roughness and hardness. The surface roughness of the underlay 4 leads when writing the pen 2 to statistically stimulated, forced vibrations of the base and the pen, ie statistical scratching and scraping noises occur. Depending on the design of the writing instrument and the documents, airborne sound waves and / or structure-borne sound waves are produced from this and onto the microphone 5 within the cursive case 3 transfer. The pencil 2 with the housing 3 as a sound body and the airborne sound chamber 7 in the pen housing 3 can be optimally interpreted according to the laws of acoustics. The orchestra 6 and the airborne sound room 7 are preferably set as resonators with certain frequencies. Frequencies are used that have the greatest dynamic range in amplitude change when writing.

When writing the pen, the microphone gives 5 a sound time signal that is clearly above the noise signal and whose sound intensity is sensitive to the writing speed and to a very small extent to the different roughness of the writing pad and the different mean contact pressure of the writing lead. The sound intensity of the recorded sound time signal increases with the writing speed. At the same writing speed The intensity differences on different writing paper surfaces are negligible. The frequency spectrum of the recorded sound time signal consists of a continuous noise spectrum that is clearly overlaid by characteristic frequency lines. The frequency lines mainly occur at frequencies below 2 kHz. With the same frequency position, the amplitudes of these frequency lines increase significantly with the writing speed. The amplitudes of the frequency lines decrease slightly with increasing surface roughness of the writing surface 4 to. The frequency positions remain unchanged. With the same writing speed, the amplitudes are almost independent of the writing pad used 4 , The airborne microphone 5 is insensitive to surrounding sound events. If the pen case 3 is completely closed, has the airborne sound chamber 7 no connection with the ambient air. With the air chamber closed 7 Surrounding noises, especially spoken words or engine noises, have no influence on the recorded sound time signal when writing. In this case, only the structure-borne noise generated via the writing lead is emitted by the microphone 5 added.

The biometric characteristics of the person's writing dynamics are determined from the change in time of the acoustic microphone signals and / or the amplitudes of selected spectral lines. Since only the dynamics of the recorded sound signals are evaluated for the generation of biometric reference data, the influence of different roughnesses on the base is 4 and the influence of manufacturing variation in the manufacture of the writing instrument is negligible. Variations in writing dynamics are therefore person-specific or person-specific and can therefore be evaluated to identify and verify the person. That through the microphone 5 acoustically recorded sound time signal is in contrast to writing pressure signals, which show the writing lead pressure in different directions, regardless of how the pen on the surface 4 is held. That through the microphone 5 generated sound signal does not change when the pen housing 3 is held differently rotated about the longitudinal axis. This in 2 illustrated writing instrument can with respect to the geometry, the material, the layer structure, the damping and the acoustic coupling between the sound source and the microphone 5 be dimensioned in such a way that optimal, acoustically less susceptible to interference biometric features can be generated for authentication methods. In the pen housing 3 there is an acoustic window or opening that can be opened by the person for voice recording at the push of a button. A relatively small opening in the pen housing is sufficient for this 3 to the airborne sound when speaking through the microphone 5 to record with. To detect the pure acoustic write signal or write noise, the housing window is closed by the person.

3 shows a further preferred embodiment of the biometric writing system according to the invention 1 , At the in 3 illustrated, particularly preferred embodiment are in the pen housing 3 next to the microphone 5 Additional sensor devices are additionally provided for acoustic detection of writing cases. The in 3 The biometric pen shown also contains an optical sensor device 33 , The optical sensor device 33 it captures the position data of the hand-held pen movement via image signals from the surface of the writing pad 4 , The optical sensor device 33 contains imaging optics for imaging the document surface and a converter device for converting the optical imaging signal into an electrical signal.

The imaging optics include an optical lens 33a , The lens optics can be glass fiber bundles or GRIN lenses. The converter device 33c the optical sensor device 33 is for example a CCD camera with an integrated microprocessor. Alternatively, a photodiode array can be used. A diode light source is preferably additionally provided for illuminating the underlay surface. In an alternative embodiment, the laser diode also serves to illuminate an optical grating integrated in the pen.

The optical sensor device 33 captured that with the pen 2 performed movements by comparing the staggered image sequences of the document surface. In an alternative embodiment, the optical sensor device detects that with the stylus 2 performed movements by comparing the temporally offset recorded image sequences of the interference of the laser light reflected on the grating and the surface of the underlay. The pen refill 2 gives on the mat when extended 4 when carrying out the hand-guided movements from a writing fluid that increases the structure of the document surface. As a result, the optical detection of the hand-guided movements by means of the optical sensor device 33 lighter, and a person receives feedback via the ink typeface, which facilitates the natural spelling.

In the 3 illustrated, particularly preferred embodiment of the biometric writing system according to the invention 1 contains next to the microphone 5 for detecting an acoustic writing noise and the optical sensor device 33 To generate position or movement data, there is also a pressure sensor system for recording the writing pressure. The pressure sensor system consists, for example, of strain gauges DMS or force sensitive resistors FSR as well as piezo sensors 34a . 34b . 34c . 34d that the static and dynamic pressure of the on the pad 4 patch and hand-held pen 2 capture sensory. The through the microphone 5 emitted sound signal data by the optical sensor device 33 emitted movement data and the recorded multidimensional writing pressure data are an integrated data processing unit 11a . 11b inside the pen housing 3 fed and processed to biometric data. The determined biometric data are preferably stored in a data memory 12 inside the pen housing 3 cached. The currently determined biometric data are then preferably encrypted by an encryption unit and transmitted to a remote computer via a data transmission link.

At the in 3 illustrated embodiment can by moving a soundproofing 8b inside the pen housing 3 the acoustic window 7 can be switched between voice and writing sound recording. In a noisy environment, for example, by closing the acoustic window 7 the disturbing outside noise can be suppressed. To increase recognition security, the person can open the acoustic window and, for example, additionally speak a written password at the same time or at different times.

That in the 1 to 3 illustrated biometric writing system 1 is very versatile. The biometric writing system 1 can be used as a computer input device. The biometric writing system is also suitable 1 as a voice input device, in particular as a dictation machine. Another possible application is the identification and verification of a person writing by comparing currently obtained biometric data with stored biometric reference data of persons. When a person is verified, it is checked whether, for example, the given signature actually comes from the right person. For this purpose, the currently generated biometric data are compared with stored reference data of this person. The prerequisite for this is that the person-specific biometric reference data of this person are first generated. For this purpose, hand-held writing movements that are carried out by this person under supervision are first recorded acoustically. The person doing the signing, for example, under supervision with their signature, and the microphone 5 within the writing device, the sound signals of the writing noise that occur are recorded. The recorded writing noises are sent to the data processing unit as sound signal data 11 issued. The sound signal data is stored in a memory 12 temporarily stored as digital sound time signals. This is then carried out by the data processing unit 11 a calculation of the associated frequency spectrum, for example by means of a Fourier transformation. First biometric reference data are determined from the stored digital switching time signal by means of feature extraction. Second biometric reference data are likewise determined from the calculated and stored frequency spectrum of the switching time signal by means of feature extraction. These method steps can preferably also be carried out several times, and the biometric reference data determined in each case are statistically evaluated. The evaluated data is then encrypted and saved as person-specific reference data. The generated biometric reference data can then be used for person verification and person identification in a variety of applications. The generated person-specific reference data are stored, for example, in a microchip of an identity card, in a microchip of a credit card, in an authorization magnetic card or in any storage unit of a computer. When generating the biometric reference data, a speech signal originating from the person is preferably also acoustically recorded and corresponding sound signal data is generated. For example, the person concerned can sign a supervision or supervision at a bank or an authority and speak his or her own name simultaneously or subsequently. To further increase security, the typists who appear when signing can also be recorded and evaluated using the pressure sensor system. A further increase in security results from additional optical detection of the hand-guided writing movement, ie optical detection of the position data of the signature made. The in 3 The pen shown senses a variety of person-specific features, in particular the typeface of the signature, the static and dynamic writing pressure that occurs when the signature is written, the typing noise generated when the signature is written, as well as the word spoken by the person writing it, e.g. their name or a password , All of this data is processed and stored as biometric reference data. The detection reliability and the protection against tampering are very high, since a wide variety of person-specific characteristics are sensed and evaluated. To verify a person, the current biometric data generated by the writing system are compared with the previously generated and stored biometric reference data of the person in order to determine whether the current biometric data largely correspond to the stored biometric reference data of the person. At a 100 percent agreement, however, is probably a pirated copy of the stored reference data, and the biometric writing system 1 according to the invention emits a corresponding warning signal.

When identifying people the current biometric data with stored biometric Reference data compared to a large number of people and that Person selected whose biometric reference data with the current best match biometric data.

In addition to personal verification and personal identification, the biometric writing system according to the invention can also be used for handwriting recognition and / or hand sketch recognition. 4 shows schematically the steps for handwriting recognition. The one when writing the pen 2 on the pad 3 When writing or sketching a character, a word or a sequence of sounds, the microphone will notify you 5 acoustically recorded and converted into corresponding sound signal data. After acoustic signal processing, coding takes place. The generated sound signal data are temporarily stored as a digital sound time signal, and then the associated frequency spectrum is calculated from the sound signal data by means of a Fourier transformation.

In a further step first current acoustic data from the sound intensity of the digital Sound time signal by means of feature extraction for data reduction certainly. From the calculated frequency spectrum of the sound time signal are also current second acoustic by means of feature extraction Data determined. This step also primarily serves to reduce data. The acoustic data finally become a sequence of feature vectors calculated, and these calculated feature vectors are also stored person-specific reference feature vectors in a pattern comparison compared. There will then be a sequence of reference feature vectors, the smallest deviation from the respective feature vectors have selected. From the sequence of selected reference feature vectors using statistical, connectionistic and knowledge-based methods the written character, the written word or the written one Word order determined.

The statistical procedure is concerned it is, for example, hidden Markow models for recognizing a larger vocabulary. The connectionist methods are primarily neuronal Network models used to recognize a limited vocabulary. The knowledge-based processes are, for example around fuzzy logic models. The reference feature vectors are at the position of the writing system, system-specific and before use the system person-specific by repeated handwritten Writing a training text generated and saved. Over longer periods of use the person's reference feature vectors become adaptive and automatically new written characters and words updated. All processes are based on acoustic models from Vocabulary one characters, letters, syllables and words Speech in speech. The phonetic transcription of a language's vocabulary through sound elements, i.e. short, phonetic units (allophones) the writing noise educated. Every word of the vocabulary is used as a sequence of allophones saved. The acoustic models determine from the vocabulary in phonetic transcription several word candidates who are most likely belong to the acoustic symbol sequence. The probability for each Word is about the HMM (Hidden Markow models) or the neuro-fuzzy systems. The most likely word candidate is through a threshold decision or about a Word order statistical model or language model determined. That determined Character then becomes, for example output as formatted text or handwritten text. To increase The written character is preferably added to the security of recognition or word pronounced by the person and by the microphone acoustically recorded. In addition, the position data of the hand-held writing movement optically recorded. A further increase in security is achieved through Acquisition of the writing pressure data occurring when writing the word.

The through the inventive writing system 1 Generated person-specific reference data can be evaluated to determine psychological and or physiological characteristics of the person. In particular, by means of the biometric writing system according to the invention 1 the occurrence of neuromotor movement disorders of a person can be recognized. The biometric writing system according to the invention is therefore suitable for diagnosing psychological and / or physiological features and for diagnosing neuromotor movement disorders in the person.

5 shows a particularly preferred embodiment of the biometric writing system according to the invention 1 with optoacoustic feedback for the person writing. From the pen 3 the biometric data obtained are sent to a local computer 20 transmitted over lines 35a . 35b of the person writing on a screen 36a and a speaker 36b gives an optoacoustic feedback. The biometric writing system according to the invention is suitable in this configuration 1 as a system for the therapy of movement disorders of the person. Thanks to the optoacoustic feedback, the person who may have a neuromotor movement disorder can learn to write again.

Another interesting area of application is the use of the biometric writing system 1 as it is in 5 is shown as Training system for learning to write. The biometric writing system according to the invention is particularly suitable 1 as a training system for learning a language, especially a foreign language. For example, the learner can put a word on the pad 4 write, and that in the pen housing 3 Integrated loudspeaker reproduces the word in the correct pronunciation of the foreign language. Through a tutorial for students or people with coordination difficulties, such as stroke patients, predetermined numbers or characters can be found on the pad 4 are attached, can be traced by the person with the pen. For example, a student can have a predetermined number or character on the pad 4 trace and receives optoacoustic feedback. Documentation of their learning progress is possible for people with coordination difficulties. For children it is by means of the biometric writing system according to the invention 1 as it is in 5 is shown, it is possible to learn to write the alphabet in a simple manner. In addition, manufacturing exercises, awareness exercises and imitation exercises are possible.

6 shows the behavior of the writing system according to the invention 1 when writing on different documents. The 6a . 6b show the recorded time sound signal at different writing speeds on a cardboard surface and on a writing paper block. In the example shown, a line is drawn with increasing speed on the cardboard and the writing paper block. The 6c . 6d show the associated frequency spectra.

The 7 shows a time sound signal that when writing the letter "a" three times when writing the pen on a writing pad by means of the biometric writing system according to the invention 1 is generated. How one 7 can be seen, the sound time signal is very easily repeatable or reproducible, so that the letter "a" can be specifically recognized.

8th also shows a time sound signal which is generated when a letter "b" is written three times. The letter "b" also has a characteristic sound time signal curve.

The 9a shows amplitude frequency signals over time for the letter "a", and the 9b shows amplitude frequency signals for writing the letter "b" over time.

The 10a . 10b principally show how the amplitude frequency lines change dynamically over time when writing a letter, for example a letter "a". At time t0, the person writing puts the pen on (starting point AP) and carries out the writing movement. The spectrum existing at time t0 changes dynamically over time. 10b shows the change in the spectrum at time t0 at time t1.

The 11 show the dynamic behavior of some frequency lines when writing a letter "a". In the example shown, the best reproduction of the letter "a" results at the frequency line f = 86.133 hertz.

The 12 show the dynamic behavior of some frequency lines when writing a letter "b". The best representation of the letter "b" is provided by the frequency line f = 86.133 hertz.

The writing system according to the invention 1 can be designed in such a way that the frequency lines capture the handwritten characters in the largest possible dynamic range with large amplitude values.

The biometric writing system according to the invention 1 has a variety of applications. On the one hand, it can be used as a biometric authentication system for personal identification and verification. The writing system according to the invention 1 is suitable as a computer input device as a replacement for a computer mouse or keyboard commands, in particular for password entry. Furthermore, the biometric writing system according to the invention 1 for electronic speech and handwriting recognition. Another extensive area of application is in the field of biometric diagnosis and therapy. Finally, the biometric writing system according to the invention 1 be used for training purposes, for example to learn to write, learn foreign languages or to solve coordination problems. Important areas of application of the writing system according to the invention arise in the field of e-commerce, in particular on the Internet, for example in banks, trade and in the credit card system. The biometric writing system according to the invention 1 can also be used as a biometric system for personal access and security control in all security areas with high security against falsification and low error rates. Other possible uses are forensic use in criminal statistics, border guards or the police. In addition, the biometric writing system according to the invention 1 used for communication between man and machine, especially as a dictation machine, or for verbal and handwritten machine control.

The biometric writing system according to the invention 1 as it is in 1 to 3 is shown, can be produced inexpensively with relatively little technical effort, so that a wide range of applications can be opened up.

1

biometric writing system

2

pen

3

Pen housing

4

document

5

microphone

6

orchestras

7

Airborne sound chamber

8th

soundproofing

9

closure device

10

management

11

Data processing unit

12

Storage

13

cables

14

optical sensor device

15

cables

16

cables

17

encryption unit

18

Data transmission link

19

Data processing unit

20

local computer

21

Storage

22

display

23

management

24

management

25

management

26

reader

27

control line

28

actuator

29

management

30

network

31

Database

32

speaker

33

optical sensor device

34

sensor device

35

management

36

acoustic Feedback devices

Claims (82)

Biometric writing system ( 1 ) with: (a) a pen housing ( 3 ) to perform hand-held movements on a surface ( 4 ); (b) at least one in the pen housing ( 3 ) integrated microphone ( 5 ) for acoustic detection of sound signals caused by hand-held movements; (c) and a data processing unit ( 11 ) for the calculation of biometric data depending on the recorded sound signals. Biometric writing system according to claim 1, characterized in that the data processing unit ( 11 ) is intended for the reconstruction of handwritten characters and texts from the recorded sound signals. Biometric writing systems according to claim 1, characterized in that in the pen housing ( 3 ) a pen ( 2 ) is provided, which is on the base ( 4 ) placed on the mat ( 4 ) to be led. Biometric writing system according to one of claims 1 to 3, characterized in that the friction of the pen ( 2 ) an acoustic writing noise is generated on the surface during the hand-held writing movement, which is transmitted as a structure-borne sound signal via the pen ( 2 ) and / or as an airborne sound signal via the ambient air to the microphone ( 5 ) is transmitted. Biometric writing system according to claim 4, characterized in that the microphone ( 5 ) to transmit the structure-borne noise signal mechanically with the stylus ( 2 ) is coupled. Biometric writing system according to claim 4, characterized in that the microphone ( 5 ) for the transmission of the structure-borne sound signal mechanically using a pen ( 2 ) connected sound body ( 6 ) is coupled. Biometric writing system according to claim 6, characterized in that the sound body ( 6 ) is designed as a resonator for certain natural frequencies. Biometric writing system according to one of the preceding claims, characterized in that the microphone ( 5 ) in one in the pen housing ( 3 ) provided airborne sound chamber ( 7 ) is arranged. Biometric writing system according to claim 8, characterized in that the airborne sound chamber ( 7 ) is designed as a resonator for certain natural frequencies. Biometric writing system according to claim 1, characterized in that the microphone ( 5 ) from sound insulation ( 8th ) is surrounded, which is provided for damping ambient noise and sound signals from the sound body ( 6 ) and the airborne sound chamber ( 7 ) lets through. Biometric writing system according to claim 8, characterized in that the airborne sound chamber ( 7 ) can be coupled to the ambient air via a housing opening in the pen housing. Biometric writing system according to claim 11, characterized in that the housing opening by means of a mechanical closure device ( 9 ) can be closed to suppress external noise. Biometric writing system according to claim 11, characterized in that the microphone ( 5 ) when the housing opening is open, the internal and external writing noise caused by the hand-held writing movement is acoustically detected as a structure-borne and airborne sound signal and / or a speech signal originating from a person. Biometric writing system according to claim 1, characterized in that the microphone ( 5 ) converts the recorded acoustic sound signals into an electrical sound signal. Biometric writing system according to claim 14, characterized in that the electrical sound signal by an analog / digital converter in sound signal data for digital data processing by the data processing unit ( 11 ) being transformed. Biometric writing system according to claim 13, characterized in that the writing noise and the speech signal simultaneously or in succession by the microphone ( 5 ) are recorded. Biometric writing system according to claim 13, characterized in that the sound signal data in a storage unit ( 12 ) can be saved. Biometric writing system according to claim 1, characterized in that the writing pad ( 4 ) consists of any paper. Biometric writing system according to claim 1, characterized in that the writing pad ( 4 ) is a solid base that has a specific surface roughness and hardness. Biometric writing system according to claim 1, characterized in that in the pen housing ( 3 ) a speaker ( 32 ) is provided for reproducing recorded microphone signals, for reproducing stored biometric reference data and for reproducing spoken information. Biometric writing system according to claim 1, characterized in that an optical sensor device ( 33 ) is provided, the position data of the hand-held pen movement is captured via image signals from the surface of the writing pad. Biometric writing system according to claim 1, characterized in that a pressure sensor device ( 34 ) is provided, which the static and dynamic pressure of the on the base ( 4 ) attached and hand-held pen ( 2 ) detected. Biometric writing system according to claim 1, characterized in that the microphone ( 5 ) is an electret microphone, a piezoelectric microphone, a piezoresistive microphone or a capacitive microphone. Biometric writing system according to claim 3, characterized in that in the pen ( 2 ) a replaceable pen lead is provided. Biometric writing system according to claim 24, characterized in that the pen lead by means of a switching device from the pen is extendable. Biometric writing system according to claim 21, characterized in that the optical sensor device ( 33 ) has imaging optics for imaging the document surface and a converter device for converting the optical imaging signal into an electrical signal. Biometric writing system according to claim 26, characterized in that the imaging optics from optical lenses and / or glass fibers consists. Biometric writing system according to claim 26, characterized in that the converter device is a CCD camera or a photodiode array is. Biometric writing system according to claim 21, characterized in that a diode light source is provided for illuminating the underlay surface is. Biometric writing system according to claim 21, characterized in that a laser diode for illuminating the underlay surface and an optical grating integrated in the pin is provided. Biometric writing system according to claim 21, characterized in that the optical sensor device ( 33 ) the movements carried out with the stylus by comparing the staggered image sequences of the document surface. Biometric writing system according to claim 21, characterized in that the optical sensor device ( 33 ) the movements carried out with the stylus by comparison of the temporally offset recorded image sequences of the interferences of the laser light reflected on the grating and the surface of the document. Biometric writing system according to claim 24, characterized in that the pen lead is in an extended state on the Document at implementation the hand-held Movements a writing fluid which increases the structuring of the underlay surface, so that the optical detection of hand-held movements using the optical sensor device facilitated and on the ink typeface the pad is a natural one Spelling is caused. Biometric writing system according to one of the preceding claims, characterized in that the data processing unit ( 11 ) in the pen housing ( 3 ) or is integrated in an external receiving unit. Biometric writing system according to claim 34, characterized in that that the external receiving unit is a computer, a microcontroller, a cell phone, a credit card reader, a fax machine or a printer. Biometric writing system according to one of the preceding claims, characterized in that the recorded data, in particular the sound signal data, the optical movement data and / or the recorded multidimensional print data via a data transmission path ( 18 ) from the pen housing ( 3 ) to the data processing unit integrated in the external receiving unit ( 19 ) are at least partially transferred. Biometric writing system according to one of the preceding claims, characterized in that in the pen housing ( 3 ) an encryption unit ( 17 ) is provided for the encryption of sound signal, optical movement data and / or print data. Biometric writing system according to claim 36, characterized in that the data transmission path ( 18 ) is wired or wireless. Biometric writing system according to one of the preceding claims, characterized in that data storage ( 12 . 21 ) are provided for storing biometric reference data, position data of the writing movement and spoken information. Biometric writing system according to one of the preceding claims, characterized in that the biometric reference data by the data processing unit ( 11 ) are calculated from the sound signal data recorded when writing and / or speaking a word, from optical movement data and / or mechanical pressure data and are stored in the reference data memory. Biometric writing system according to one of the preceding claims, characterized characterized in that the reference data memory microchips one Identity card, a credit card, an authorization magnetic card or storage unit of a computer or a writing system. Biometric writing system according to one of the preceding claims, characterized characterized that the written and / or spoken words and Characters pin codes, passwords, names or are texts. Biometric writing system according to one of the preceding claims, characterized in that the position data of the writing movement is calculated from the sound signal data, optical movement data and / or mechanical pressure data recorded when writing a word and is stored in the data memory ( 12 . 21 )) for handwriting recognition. Biometric writing system according to claim 39, characterized in that the reference data memory ( 12 ) in the pen housing ( 3 ) is integrated. Biometric writing system according to claim 36, characterized in that the external receiving unit ( 20 ) integrated data processing unit ( 19 ) to a reference data storage ( 21 ) has a connection. Biometric writing system according to one of the preceding claims, characterized in that the external receiving unit ( 20 ) a device ( 26 ) for reading a portable storage medium of biometric reference data. Biometric writing system according to one of the preceding claims, characterized in that the external receiving unit ( 20 ) via a data network ( 30 ) to a database ( 31 ) is connected to a reference data memory or can establish a connection to a reference data memory. Biometric writing system according to claim 47, characterized in that the data network ( 30 ) the internet is. Biometric writing system according to one of the preceding claims, characterized in that the data processing unit ( 11 . 19 ) compares the calculated current biometric data with the stored biometric reference data for their verification and identification. Biometric writing system according to claim 49, characterized in that the data processing unit ( 11 . 19 ) If the current biometric data largely matches the stored reference data, an identification or verification display signal is generated. Biometric writing system according to claim 49, characterized in that the data processing unit ( 11 . 19 ) if the current biometric data completely matches the stored biometric reference data, recognizes the current biometric data as a pirated copy of the stored reference data and generates a warning signal. Biometric writing system according to claim 48, characterized in that the data processing unit ( 11 . 19 ) If the current biometric data deviates from the stored biometric reference data, a deviation display signal is generated. Biometric writing system according to claim 50, characterized in that at least one actuator ( 28 ) is provided, which is actuated after generating the identification or verification display signal. Process for the generation of person-specific biometric Reference data with the following steps: (a) Acoustic Capture of hand-held writing movements, the one with a pen on a pad when writing a Character, word or character / phrase by one person carried out and generating corresponding sound signal data; (B) Storing the generated sound signal data as a digital sound-time signal; (C) Calculate an associated one Frequency spectrum from the sound signal data by means of Fourier transformation; (D) Determination of first biometric reference data from the sound intensity of the digital sound-time signal by means of feature extraction; (e) Determine second biometric Reference data from the calculated frequency spectrum of the sound-time signal by means of feature extraction. A method according to claim 54, characterized in that the Process steps are carried out several times and each certain reference data are statistically evaluated, the evaluated data stored as person-specific reference data encrypted become. 55. The method according to claim 55, characterized in that the statically evaluated person-specific reference data in one Microchip of an identity card, in a microchip of a credit card, into an authorization magnetic card or into the storage unit of a Computers or a writing system can be saved. A method according to claim 54, characterized in that an additional speech signal originating from the person is recorded acoustically and corresponding Sound signal data are generated. A method according to claim 57, characterized in that at the carried out Writing movement of the pen in at least one spatial direction forces occurring via pressure sensors are recorded and correspondingly at least one-dimensional write-print data be generated. A method according to claim 54, characterized in that the when writing a character, a word or a character / word sequence with the pen on the surface of hand-guided writing movements optically are recorded and corresponding write movement data are generated. A method according to claim 54, characterized in that the biometric reference data from the sound intensity of the digital Time sound signal by means of feature extraction, from the calculated Frequency spectrum of the sound time signal by means of feature extraction as well as those generated when writing and speaking a word optical movement data and write-print data determined and in the reference data storage as person-specific biometric reference data can be saved. A method according to claim 60, characterized in that a Data reduction of the static and dynamic write movement data, the Writing print data, optical movement data and sound signal data by means of cluster formation. A method according to claim 61, characterized in that feature parameters are extracted from the data-reduced cluster data. A method according to claim 62, characterized in that the generated biometric reference data inside or outside of the pen can be saved. A method for verifying a person, comprising the following steps: (a) acoustic detection of hand-held writing movements which are carried out by the person with a pen on a base when writing a character, a word or a sequence of words and generating a corresponding sound signal Data; (b) storing the generated sound signal data as a digital sound time signal; (c) calculating an associated frequency spectrum from the sound-time signal by means of Fourier transformation; (d) determining first current biometric data from the sound intensity of the digital time sound signal by means of feature extraction; (e) determining second current biometric data from the calculated frequency spectrum of the sound-time signal by means of feature extraction and; (f) comparing the current biometric data with stored biometric reference data of the person to verify whether the current biometric data largely agree with the stored biometric reference data of the person. A method according to claim 64, characterized in that an additional speech signal originating from the person is recorded acoustically and corresponding Sound signal data are generated. Procedure for identifying a person with the following steps: (a) acoustic detection of hand-held writing movements, the one with a pen on a pad when writing a Character, a word or a character / phrase can be carried out by one person and generating corresponding sound signal data; (b) Save the sound signal data generated as a digital sound-time signal; (C) Calculate an associated one Frequency spectrum from the sound-time signal by means of Fourier transformation; (D) Determining the first current biometric data from the sound intensity of the digital sound-time signal by means of feature extraction; (e) determining current second biometric data from the calculated frequency spectrum of the sound-time signal by means of feature extraction; (f) Compare the current biometric Data with stored biometric reference data of a large number of people to identify whether the current biometric Data with the stored biometric reference data one of the saved people largely match. A method according to claim 66, characterized in that an additional speech signal originating from the person is recorded acoustically and corresponding Sound signal data are generated. Process for recognizing handwriting and / or recognizing sketches with the following steps: (a) acoustic detection of hand-held writing movements, the with a pen on a pad when writing or Outline a character, word, or phrase one person running and generating corresponding sound signal data; (B) Storing the generated sound signal data as a digital sound-time signal; (C) Calculate an associated one Frequency spectrum from the sound signal data by means of Fourier transformation; (D) Determining the first current acoustic data from the sound intensity of the digital Sound-time signal by means of feature extraction for data reduction; (E) Determining current second acoustic data from the calculated Frequency spectrum of the sound-time signal by means of feature extraction for data reduction; (f) computing a sequence of feature vectors from the acoustic data; (g) comparing the sequence of feature vectors with stored person-specific reference feature vectors; (H) Select a sequence of reference feature vectors that are from the respective feature vectors have the smallest deviation; (I) Find the character, word, or phrase from the sequence selected reference feature vectors using statistical, connectionistic and knowledge-based processes. Process for recognizing handwriting and / or recognizing sketches according to claim 68, characterized in that an additional speech signal originating from the person is recorded acoustically and corresponding Sound signal data are generated. A method according to claim 68, characterized in that the when writing or sketching a character, a word or a string of characters / words with the pen on the pad hand-held Write movements are recorded optically and corresponding position data be generated. A method according to claim 68, characterized in that at the carried out Writing movement of the pen in at least one spatial direction forces occurring detected by pressure sensors and corresponding at least one-dimensional Write print data are generated. Method according to one of the preceding claims 68 to 71, characterized in that that the character, word, or string of words / phrases from the sequence selected reference feature vectors, from the position data and from the write-print data using statistical, connectionistic and knowledge-based processes. A method according to claim 68, characterized in that the identified signs of the executing Person optically and / or acoustically confirmed for his control becomes. A method according to claim 54, characterized in that the generated person-specific reference data from the determination psychological and / or physiological characteristics of the person evaluated become. Use of the biometric writing system according to one of claims 1 to 53 as a computer input device. Use of the biometric writing system according to one of claims 13 to 53 as a voice input device, in particular as a dictation machine. Use of the biometric writing system according to one of claims 1 to 53 for detecting a person's neuromotor movement disorders. Use of the biometric writing system according to one of claims 1 to 53 as a therapy system to remedy neuromotor movement disorders Person. Use of the biometric writing system according to one of claims 1 to 53 as a graphological system for determining psychological / physiological Characteristics of a person. Use of the biometric writing system according to one of the preceding Expectations 1 to 53 as a training system for learning to write. Use of the biometric writing system according to one of claims 13 to 53 as a training system for learning a language. A method according to claim 54, characterized in that the biometric reference data can be encrypted and saved.