CN101604212B - Writing action and trajectory recognition method, writing device and recognition system thereof - Google Patents

Abstract

The invention discloses a writing action identification method, which mainly captures a voice message generated by a writing tool during writing and knows that the writing tool is in a plane to write or leaves the writing surface by judging whether the voice message exists or not. In another embodiment, the present invention further provides a writing track recognition method, which combines the writing motion recognition method with a character recognition technology to convert the writing track into characters. In addition, the present invention further provides a writing device and an identification system, wherein the writing device is provided with a device capable of collecting voice and information, and the processing unit with the writing action and track identification method is used for processing to identify corresponding characters.

Description

Writing action and trajectory recognition method, writing device and recognition system thereof

technical field

The present invention relates to a writing recognition technology, in particular to a writing motion and track recognition method, a writing device, and a recognition system that uses sound signals to judge writing motions.

Background technique

With the advancement of science and technology, the life of modern people has become inseparable from information technology products. Among information technology products, especially technology products such as computers and mobile phones, they are closely related to the lives of most people. And users can use these computer technology products to transmit the operation commands to be executed to the computer, and after processing and conversion, the computer will present the processed results on the screen. In such an operation process, the way of communication between the user and the computer needs to use an input interface as an interface for man-machine dialogue.

In the past, the most common human-machine interfaces of computer products were keyboards and mice. Generally speaking, the mouse can only change the input mode of movement and selection, so its functions are limited and it is difficult to issue complex and diverse commands. As for keyboard input, because there are many keys, it needs to be studied and familiarized with for a certain period of time, so there is a higher learning threshold. In addition, it is usually difficult to draw and process delicate graphics using a keyboard or mouse, or it takes more time to process. Based on these limitations, other input methods, such as: handwriting or voice man-machine input mode will emerge accordingly.

In the commonly used handwriting input mode, such as a handwriting input trajectory recognition system and method disclosed in United States Published Patent Application No. US2004/0184659 discloses a handwriting trajectory recognition system, including: a motion detection unit for outputting based on the system ontology The electronic signal of the acceleration change in space and a control unit are used to detect the non-handwriting segment interval when the system body is temporarily stopped, which defines the writing time range by comparing the standard deviation of the acceleration value with a threshold value to obtain handwritten electronic signals. In this technology, writing between characters requires a deliberate pause to detect breakpoints.

In addition, another example is a writing recognition system disclosed in U.S. Published Patent Application No. US 2006/0088215, which uses a pressure sensor at the front end of the writing tool to determine whether there is contact with the writing surface and separate the trajectory of the word and the pen-lifting signal, and then Then use the acceleration calculation to write the trajectory and add a pressure gauge as a handwriting signal breakpoint detection. Another example is US Pat. No. 6,628,847, which discloses a pressure sensor at the tip of the pen to determine whether there is contact with the writing surface, separate the track of the word and the signal for lifting the pen.

Contents of the invention

The present invention provides a writing action recognition method, which can detect the sound generated by the writing tool when writing on a plane, and determine whether the writing tool is writing on the plane or away from the writing surface by judging whether the sound is present or not. .

The present invention provides a writing track identification method and its writing device and identification system, which can use the audio information to correspond to the writing track of the writing tool to identify the track signal belonging to the writing state, and then identify the track signal to generate a pair The text when the state should be written.

The invention provides a writing device and a recognition system. A device capable of collecting audio information is provided in the writing device. The system combines the writing device and a processing unit with a writing action and trajectory recognition method to recognize corresponding characters.

In one embodiment, the present invention provides a writing action recognition method, which includes the following steps: capturing a sound signal generated by a writing tool when writing; and judging the state of the writing tool according to the sound signal.

In another embodiment, the present invention further provides a writing action recognition method, which includes the following steps: extracting a sound signal generated by a writing tool when writing; and judging the sound signal of the writing tool according to the sound signal. state.

In another embodiment, the present invention further provides a method for recognizing writing traces, which includes the following steps: extracting a sound signal generated by a writing tool when writing; a continuous electronic signal; converting the sound signal into an identification sound signal including at least one written message; comparing the identification sound signal with the continuous electronic signal; and corresponding to the position of the at least one written message in the continuous electronic signal The possessed signals undergo an identification and combination process.

In another embodiment, the present invention further provides a writing device, including: a writing body, which can generate a writing action; a sound collecting device, which is arranged on the writing body, and the sound collecting device can pick up the writing body A sound signal generated when writing is generated; and an audio judging unit, which is connected to the sound receiving device by telecommunication, the audio judging unit can receive the audio signal and process it to generate an identification audio signal corresponding to the state of the writing body.

In another embodiment, the present invention further provides a writing recognition system, including: a writing body, which can generate a writing action; a sound receiving device, which is arranged on the writing body, and the sound receiving device can capture the writing A sound signal generated when the main body produces a writing action; an audio judging unit, which is connected to the sound receiving device by telecommunication, and the audio judging unit can receive the audio signal and process it to generate an identification audio signal corresponding to the state of the writing body; A motion detection unit, which can detect the motion of the writing body to generate a continuous electronic signal; a trajectory processing unit, which can compare the recognition sound signal with the continuous electronic signal, and in order to correspond to at least one writing audio in the continuous electronic signal Identifying and combining the signals of the position; and a character recognition unit, which can generate corresponding written content according to the result of the identification and combination.

Description of drawings

FIG. 1A is a schematic flow chart of an embodiment of the writing action recognition method of the present invention;

Fig. 1B is a schematic flow chart of an embodiment of judging the state of the writing instrument of the present invention;

2A is a schematic diagram of an initially measured sound signal;

Fig. 2B is a schematic diagram of the processed sound signal;

FIG. 3 is a schematic flow chart of another embodiment of the writing action recognition method of the present invention;

FIG. 4 is a schematic flow chart of an embodiment of the method for recognizing writing traces in the present invention;

Fig. 5 is a schematic flow chart of the identification combination program of the present invention;

Figure 6A and Figure 6B are schematic diagrams of the comparison between the sound signal and the continuous electronic signal of the present invention;

7A to 7C are schematic diagrams of character recognition;

Fig. 8 is a text schematic diagram with a turning point;

Fig. 9 is a schematic diagram of an embodiment of the writing device of the present invention;

10A is a schematic diagram of the writing recognition system of the present invention;

FIG. 10B is a schematic diagram of an embodiment of a trajectory processing unit of the present invention;

Fig. 11 is a schematic diagram of another embodiment of the writing body of the present invention.

Wherein reference sign is:

1-Writing Action Recognition Method

10～11-steps

120～121-steps

2-Writing Action Recognition Method

20～23-steps

3-Writing track identification method

30～34-steps

340～342-steps

4- Writing device

40-Writing Ontology

41-radio device

410, 411-Microphone

42 - noise processing unit

421-Sonic Directional Control Unit

43-Message judging unit

5- Writing recognition system

50-writing ontology

51 - radio device

510, 511-Microphone

52 - noise processing unit

53-Message judging unit

54-Motion detection unit

55- Trajectory processing unit

550-signal division operation unit

551-Track computing unit

56-Text Recognition Unit

90-threshold value

91, 910, 911, 912 - sound signal

913 - turning point

92 - Continuous electronic signal

920, 921, 922-acceleration sensing signal

924-area

925-track signal

926-action signal

927, 928-area

9270～9273-track

9280～9282-relative position

93-area

94-track

95-track direction

Detailed ways

In order to further recognize and understand the features, purposes and functions of the present invention, the relevant detailed structure and design concept of the device of the present invention will be described below in detail as follows:

Please refer to FIG. 1A , which is a schematic flowchart of an embodiment of the writing action recognition method of the present invention. In this embodiment, the method 1 includes a step 10 of capturing a sound signal generated by a writing tool when writing. In this step, the writing tool may be a pen or a tool that can be held by the user, or even a finger of the user, etc., but not limited thereto. When the writing instrument is writing, the sound will be generated due to friction, so the sound signal generated during writing can be collected by means of a sound collecting device or component. As for the method of capturing sound in step 10, there are many ways. The receiver can be set on the writing device or the writing board, or even an external receiver can be used to receive audio messages. There is no specific limitation. Then go to step 12, judge the state of the writing instrument according to the sound signal. Please refer to FIG. 2A , which is a schematic diagram of an audio signal. Since the collected sound signals basically cover the sound generated by friction when the writing tool is writing, and also include the sound generated when the writing tool moves, so in this step, the signal processing method can be used to distinguish Isolate the information generated during writing and the information generated during non-writing (writing instrument movement).

As shown in FIG. 1B , this figure is a schematic flow chart of an embodiment of judging the state of the writing instrument according to the present invention. First, step 120 is performed, and a threshold value is first set. The threshold value can be determined according to the actual situation, and there is no certain limit. Please refer to FIG. 2A to understand the meaning represented by the threshold value of 90, that is, signals within the range of the threshold value of 90 will be regarded as non-writing information. Shown in Fig. 1B again, proceed to step 121, when the sound signal exceeds the threshold value, it is judged that the writing tool is in the writing state, if the sound signal is smaller than the threshold value, then it is judged that the writing tool is in the moving state . After step 121 is completed, the signal shown in FIG. 2A can be formed as shown in FIG. 2B , leaving only the sound signal of the judged writing instrument when writing.

Please refer to FIG. 3 , which is a schematic flowchart of another embodiment of the writing action recognition method of the present invention. In this embodiment, the method 2 includes a step 20 of capturing a sound signal generated by a writing tool when writing. Then proceed to step 21 , using a beamforming method to direct the sound collection direction to the direction in which the sound is generated during writing to form an enhanced sound signal. Because when writing, although there will be the sound produced by the friction between the writing tool and the plane, there will still be interference from other sounds in the external environment. Therefore, in this embodiment, beamforming can be used to direct the direction of sound reception to the direction of sound generation during writing, so as to reduce errors in subsequent judgments and increase the accuracy of judgments. As for the method of beamforming, it is mainly to digitally convert and weight the sound signals received by the array of radio devices to perform calculation processing on the sound signals to obtain an output result that can adjust the radio beam direction. The way of beamforming belongs to the prior art, and it can be used for signal processing such as sound waves and radar waves, and will not be described in detail here. Then proceed to step 22 , performing a noise filtering step on the enhanced sound signal. In this step, in order to reduce noise interference, the enhanced sound signal can be processed by a band-pass filter to filter out unnecessary noise. Then, go to step 23, judge the state of the writing instrument according to the sound signal. This state includes writing state or non-writing movement. The procedure of step 23 is the same as that described in step 12 above, and will not be repeated here.

Please refer to FIG. 4 , which is a schematic flowchart of an embodiment of the method for identifying writing traces of the present invention. In this embodiment, the written characters are mainly determined by combining the above-mentioned writing action recognition method with the writing track recognition of the writing device. The method 3 includes the following steps. First, step 30 is performed to capture a sound signal generated by a writing instrument when writing. This step is as described in step 10 above, and will not be repeated here. Then proceed to step 31 , to capture a continuous electronic signal generated by the writing device due to movement changes. Since the sound signal in step 30 can only be used as a basis for whether there is writing or not, it is necessary to sense the moving state of the writing device during writing in step 31 to facilitate subsequent character recognition procedures.

In step 31, there are many ways to sense the movement change of the writing device, for example: setting a gyroscope or an inertial sensing component of an acceleration sensor in the writing device or setting an image sensing device (CCD), or even using infrared rays or Existing technologies such as ultrasound can sense the movement changes of the writing device. Then proceed to step 32, converting the audio signal into an identification audio signal, which includes at least one written audio message. Because the sound signal obtained in step 30 includes the audio information of the writing device when it is writing, the audio signal of the writing device while it is moving, and even the interference information or noise of the external environment. Therefore, in step 32, the same as the aforementioned steps 21 to 23, the sound signal can be processed first by using the beamforming procedure mentioned in steps 21 and 22 in FIG. 2 to filter out noise, and then Go to step 23 again, judge the state of the writing instrument according to the sound signal. This state includes a writing state or a non-writing moving state. The result of step 32 can refer to the signal status shown in FIG. 2B.

After step 32, proceed to step 33 to compare the identified sound signal with the continuous electronic signal. Because the sound signal is consistent with the time of the continuous electronic signal, after determining the sound signal of the writing state in step 32, it can be compared with the continuous electronic signal to determine the time of the continuous electronic signal. Which segment of the electronic signal in the signal belongs to the electronic signal sensed during writing. Please refer to FIG. 6A and FIG. 6B , which are schematic diagrams of the comparison between the sound signal and the continuous electronic signal according to the present invention. In FIG. 6A , the sound signal 91 can distinguish between a writing sound signal 910 and a non-writing sound signal 911 . The continuous electronic signal 92 is mainly composed of three-axis acceleration sensing signals 920 , 921 and 922 , but not limited thereto. When the sound signal 91 is compared with the continuous electronic signal 92 , the continuous electronic signal in the area 93 corresponding to the non-writing sound signal 911 is judged as meaningless electronic signal. Conversely, the continuous electronic signal 92 corresponding to the sound signal 910 during writing is a meaningful electronic signal. After the comparison in step 33 is completed, the state shown in FIG. 6B is formed.

Returning to FIG. 4 , proceed to step 34 , performing an identification and combination procedure on the signal corresponding to the position of at least one written audio message in the continuous electronic signal. The so-called signal in the continuous electronic signal corresponding to the position of the at least one written audio message in step 34 is the continuous electronic signal covered by the area 924 in FIG. 6B . As shown in FIG. 5 , this figure is a schematic flow chart of the identification and combination program of the present invention. First, step 340 is performed to divide the continuous electronic signal into a track signal and an action signal, wherein the track signal corresponds to the signal of the position of at least one written audio message, and the action signal corresponds to the position of the non-written audio message. As shown in FIG. 6A , the continuous electronic signal 92 is an electronic signal sensed by an acceleration sensor. After being divided, the result is shown in FIGS. 7A and 7B . 7A shows that the continuous electronic signal corresponds to the track signal 925 of the writing sound signal 910 , while FIG. 7B shows that the continuous electronic signal corresponds to the action signal 926 of the non-writing sound signal 911 .

Referring back to FIG. 5 , proceed to step 341 , and determine at least one writing track according to the track signal. In step 341, as shown in FIG. 7A , the track restoration step is performed on the track signal 925 to understand what track and direction each electronic signal corresponding to the writing sound signal represents. 9270 to 9273. Next, step 342 shown in FIG. 5 may be performed to determine the relative position between the at least one writing track according to the action signal. As shown in FIG. 7B , the track recovery step is performed on the motion signal 926 to know the motion track between adjacent writing tracks, and then determine the relative position between adjacent writing tracks. For example, in Fig. 7B, the motion signal 9260 represents the inertial signal generated by the hand movement between the writing tracks before and after it, so the direction and distance of the movement can be determined according to the signal, and then the relative position and distance between adjacent writing tracks can be understood. Relative positions 9280 to 9282 determined according to step 342 are represented in area 928 in FIG. 7B .

Referring back to FIG. 5 , step 343 is performed to combine the at least one writing track and its corresponding relative position. This step is mainly to use the trajectories 9270 to 9273 and the relative positions 9280 to 9282 determined in FIG. 7A and FIG. 7B to combine according to the sequence of time to obtain the pattern as shown in FIG. 7C . Since the pattern has no text meaning before it is recognized, it is necessary to perform text recognition according to the combination result of step 343 through step 344 of FIG. 5 , and display the recognition result on the screen.

For example, as shown in FIG. 7A and FIG. 7B , the figure is a schematic diagram of character recognition. When step 341 is used, the trajectory 94 and the trajectory direction 95 corresponding to each piece of writing sound signal as shown in FIG. 7A can be obtained. The trajectories 94 identified in FIG. 7A are then combined into a pattern as shown in FIG. 7C using step 342 . And through step 344, it is recognized that the pattern is the word "王" in Chinese. In addition, since there may be a turning point during the writing process, for example, as shown in Figure 8, there will be a turning point at the position 913 when writing the character "2", so the sound signal of the present invention can also be used to assist in judging The position where the turning point occurs is used for subsequent text judgment. As shown in FIG. 6B , the position of the mark 912 in the sound signal represents that a turning point occurs during writing.

Please refer to FIG. 9 , which is a schematic diagram of an embodiment of the writing device of the present invention. In this embodiment, the writing device 4 has a writing body 40 , a sound collecting device 41 , a noise processing unit 42 and an audio message judging unit 43 . The writing body 40 can generate a writing action. In this embodiment, the writing body 40 is in the shape of a pen, which is convenient for the user to hold. In another embodiment, as shown in FIG. 11 , the writing body 40 is in the shape of a sleeve to accommodate the user's fingers. In the embodiment of FIG. 11 , the user can directly use the sleeve-shaped writing body 40 to write directly on the writing board. The sound collecting device 41 is arranged on the writing body 40, and the sound collecting device 41 can pick up the sound signal generated by the writing body 40 during the writing action. The sound collecting device 41 can be composed of at least one sound collecting component (such as a microphone or a directional microphone). In this embodiment, the sound collecting device is composed of a plurality of sound collecting microphones 410 and 411 . The plurality of sound-receiving microphones can be arranged in groups on the writing body 40 .

The sound receiving device 41 is further connected with a noise processing unit 42, the noise processing unit 42 can receive the sound signal generated by the sound receiving device 41, and filter out the noise of the sound signal. In addition, the noise processing unit 42 further has a sound wave direction control unit 421, which has a beamforming calculation mechanism, and can guide the sound collecting direction of the sound collecting device 41 to the direction in which the writing sound occurs. As for the calculation mechanism of the beam shaping belongs to the prior art, it will not be repeated here. The audio judging unit 43 is telecommunication connected with the noise processing unit 42. The audio judging unit 43 can receive the noise-processed audio signal and process it to generate an identification audio signal corresponding to the state of the writing instrument. The audio judging unit 43 mainly processes the audio signal into the identification audio signal shown in FIG. 2B through the procedure of step 12 in FIG. 1A or step 23 in FIG. 3 . The noise processing unit 42 in the present embodiment can be set as required, although the noise processing unit 42 is taught to be connected between the sound receiving device 41 and the audio signal processing unit 43 in Fig. 9, but the sound receiving device 41 can also be It is directly connected to the audio processing unit 43 .

Please refer to FIG. 10A , which is a schematic diagram of the writing recognition system of the present invention. The writing recognition system 5 has a writing body 50 , a sound receiving device 51 , a noise processing unit 52 , an audio message judging unit 53 , an action detection unit 54 , a trace processing unit 55 and a character recognition unit 56 . The writing body 50 is, in this embodiment, a pen-like element for the user to hold and write on the writing board. Of course, the writing body 50 can also have the structure shown in FIG. 11 . The sound-receiving device 51 has a plurality of sound-receiving components 510 and 511 arranged on the writing body, which can be arranged in groups or ring-shaped on the writing body 50 . The sound collecting components 510 and 511 can be selected as components such as directional microphones or microphones. The sound receiving device 51 can pick up the sound signal generated by the writing body 50 during the writing action.

The noise processing unit 52 is electrically connected with the sound receiving device 51 to perform noise filtering and sound guiding processing on the sound signal captured by the sound receiving device 51 . The function of the noise processing unit 52 in this embodiment is the same as that of the noise processing unit 42 in the embodiment of FIG. 9 , so details are not described here. The audio judging unit 53 is electrically connected with the noise processing unit 52, and the audio judging unit 53 can receive the sound signal and process it to generate a recognition sound signal corresponding to the state of the writing body (the sound signal shown in Figure 2B ). The motion detection unit 54 can detect the motion of the writing body to generate a continuous electronic signal. In this embodiment, the motion detection unit 54 can be composed of an inertial sensing component (such as a gyroscope and an acceleration sensor), or an image sensing component (CCD or CMOS), or an ultrasonic or infrared sensor. The continuous electronic signal detected by the motion detection unit 54 can be used to restore the writing track.

The track processing unit 55 can compare the recognized sound signal with the continuous electronic signal, so as to identify and combine the signal of the position corresponding to at least one written audio message in the continuous electronic signal. As shown in FIG. 10B , which is a schematic diagram of an embodiment of the trajectory processing unit of the present invention. In this embodiment, the trajectory processing unit 55 further has a signal division operation unit 550 and a trajectory operation unit 551 . The signal division operation unit 550 compares the recognized sound signal with the continuous electronic signal and divides the continuous electronic signal corresponding to the at least one written audio message into at least one sub-electronic signal (result shown in FIG. 6B ). The trajectory calculation unit 551 combines the at least one electronic sub-signal. The results processed by the trajectory computing unit 551 can be referred to as shown in FIG. 7A and FIG. 7B . After the trajectory computing unit 551 combines the trajectories, the result of the combination can be transmitted to the character recognition unit 56 . Generally speaking, the character recognition unit 56 can be disposed in a computer system 6 . The character recognition unit 56 can generate corresponding written content according to the result of recognition and combination. For example, after the trajectory calculation unit 551 combines the trajectory of FIG. 7A into the state of FIG. 7B, the result of the combination is transmitted to the character recognition unit 56, and the character recognition unit 56 finds the matching character from the character database it has. text, and display it on the monitor in the computer system 6 .

Referring back to FIG. 9 , the path processing unit 55 transmits the processed signal to the character recognition unit 56 in a wireless or wired manner. On the other hand, although the trajectory processing unit 55 of the present invention is arranged on the writing body 50, for people familiar with this technology, it can be easily changed and arranged in the computer system 6, and directly by a message The judging unit 53 and the motion detecting unit 54 transmit the sensed signals to the trajectory processing unit 55 in a wireless or wired manner.

The above descriptions are merely examples of the present invention, and should not be used to limit the scope of the present invention. That is, all equivalent changes and modifications made according to the claims of the present invention will still not lose the gist of the present invention, nor depart from the spirit and scope of the present invention, so all should be regarded as further implementation status of the present invention.

To sum up the above, the writing action and track identification method and its writing device and identification system provided by the present invention can identify whether the writing tool is in the writing state, and further identify the writing track when writing, so as to form a corresponding writing track. Word. Therefore, it has been able to improve the competitiveness of this industry and promote the development of surrounding industries. Sincerely, it has met the requirements required for an invention application as stipulated in the Invention Patent Law, so the application for an invention patent is submitted according to law.

Claims (10)

1. A writing track identification method is characterized in that, comprising the following steps: capturing a sound signal generated by a writing tool when writing; Capture a continuous electronic signal generated by the writing instrument due to changes in motion; converting the sound signal into a recognized sound signal comprising at least one written message; comparing the identified sound signal with the continuous electronic signal; and performing an identification and combination procedure on the signal corresponding to the position of at least one written message in the continuous electronic signal; The identification and combination procedure further includes the following steps: The continuous electronic signal is divided into a trajectory signal and an action signal, characterized in that the trajectory signal corresponds to the signal of the position of the at least one written audio message, and the action signal corresponds to the position of the non-written audio message; determining at least one writing track according to the track signal; determining the relative position between the at least one writing track according to the action signal; combining the at least one writing track and its corresponding relative position; and Character recognition is performed according to the result of the combination. 2. The method for recognizing writing traces as claimed in claim 1, further comprising a step of filtering the sound signal to remove noise. 3. The writing track recognition method according to claim 1, wherein converting the sound signal further comprises the following steps: set a threshold; and When the sound signal exceeds the threshold signal, it is judged to be a writing message when the writing tool is in a writing state, and if the sound signal is smaller than the threshold value, it is judged to be a non-writing message when the writing tool is in a moving state. 4. A writing recognition system, characterized in that it comprises: a writing body, which produces a writing action; A sound receiving device, which is arranged on the writing body, and the sound receiving device picks up the sound signal generated when the writing body produces a writing action; An audio message judging unit, which is connected to the sound receiving device by telecommunication, the audio message judging unit receives the audio signal and processes it to generate an identification audio signal corresponding to the state of the writing body; a motion detection unit, which detects the motion of the writing body to generate a continuous electronic signal; a track processing unit, which compares the identified sound signal with the continuous electronic signal, so as to identify and combine the signal of the position corresponding to at least one written audio message in the continuous electronic signal; and A character recognition unit, which generates corresponding writing content according to the result of the recognition and combination; The trajectory processing unit further has: A signal division operation unit, which compares the identification sound signal with the continuous electronic signal and divides the continuous electronic signal corresponding to the at least one written audio message into at least one sub-electronic signal; and A trajectory operation unit, which combines the at least one electronic sub-signal. 5. The writing recognition system according to claim 4, wherein the state of the writing body includes one of a writing state and a non-writing moving state. 6 . The handwriting recognition system according to claim 4 , further comprising a noise processing unit connected to the sound receiving device and the audio judging unit to filter out the noise of the audio signal. 7 . 7. The writing recognition system according to claim 6, wherein the noise processing unit further has a sound wave direction control unit, which directs the sound collection direction of the sound collection device to the direction in which the writing sound occurs. 8. The writing recognition system according to claim 4, wherein the sound collecting device further has at least one sound collecting microphone. 9. The handwriting recognition system according to claim 4, wherein the motion detection unit is selected as one of an inertial sensing unit and an image capturing unit. 10. The writing recognition system according to claim 9, wherein the inertial sensing unit is selected as one of an accelerometer, a gyroscope, and a combination thereof.

Images