TWI552025B - Pattern matching method and touch integrated circuit - Google Patents

Description

Pattern matching method and touch integrated circuit

The present invention relates to a method for recognizing a user's gesture input for a user's appliance and related computer software and equipment.

When a portable user device such as a smartphone is used, when a finger or the like touches a screen and inputs a mode set in advance, the technique for recognizing this mode is immediately introduced. In order to identify such patterns, the touch locations are sampled and stored for a certain time interval, and a stored series of touch locations are compared with a previously stored standard mode to determine whether the input mode is similar to the stored standard mode.

According to the conventional technique, in order to accurately perform the recognition of the above modes, the above sampling must be performed promptly. At this time, since the processor speed of the user's electric appliance is relatively fast, the energy consumption of the user's electric appliance is relatively large.

On the other hand, portable consumer appliances are sensitive to power consumption because they use batteries. Therefore, in order to allow the user's appliance to minimize power consumption during the lock screen, the clock speed is reduced. If the above pattern recognition is performed in the above-described lock screen mode, a problem that the recognition rate is extremely low will occur due to a lower processor speed.

The present invention is directed to providing a pattern matching method that provides accurate pattern matching results even at lower processor speeds.

According to one aspect of the present invention, in the provided pattern matching method, not only the obtained coordinates but also the correlation values of the displacement vectors between the respective coordinates can be utilized during the completion of the input mode. The raw information obtained at a limited sampling speed is a coordinate about the user's input completion, but there will be many additional information obtained after processing it. In this additional information of the present invention, in particular, the displacement vector between the coordinates is used to increase the recognition rate, thereby providing a simple pattern recognition calculation program.

According to another aspect of the present invention, the provided pattern matching method includes: in a standard input mode, as a sampled N standard tuples, respectively including standard input coordinates and standard correlation vectors, and preparing phases of N standard tuples; In the comparison input mode, as the sampled M comparison tuples, respectively, including the comparison input coordinates and the comparison correlation vector, preparing the M comparison tuple stages (however, M>=N); presuming the above M comparison tuples a phase of N comparison tuples matching the above N standard tuples; respectively calculating a standard representative value of the standard correlation vector and a comparison representative value of the comparison correlation vector according to the matched standard tuple and the comparison tuple pair a phase of the distance; using the above N distances of the N standard tuples and the N comparison metablocks, using at least one of the distances to determine the similarity between the standard input mode and the comparison input mode The stage of the degree.

At this time, the above speculative phase includes: among the N comparison tuples belonging to the above The M comparison input coordinates are presumed to have a phase from the N standard input coordinates belonging to the N standard tuples to the N comparison tuples of the most recent input input coordinates.

In this case, the trend value in the up, down, left, and right directions of the standard correlation vector is the standard correlation vector, and the trend value in the up, down, left, and right directions of the comparison correlation vector may be the comparison correlation vector.

In this case, the comparison representative value is the comparison correlation vector, and a representative value of one or more other comparison correlation vectors sampled before the comparison correlation vector is sampled.

At this time, the phase of determining the similarity includes a phase, that is, if at least one of the N distances obtained exceeds a preset first threshold, the standard input mode and the comparison input may be determined. The patterns are not similar.

At this time, the phase of determining the similarity includes a phase, that is, if the representative value of the N distances obtained exceeds the second threshold set in advance, it can be determined that the standard input mode and the comparison input mode are not similar.

At this time, each stage included in the pattern matching method processes the analog signal input in the touch screen device to determine the touch input position; and transmits the relevant information of the determined touch input position to other processing devices at these stages in the touch product. The body circuit (touch IC) is completed.

In this case, the touch IC includes a first mode that operates at a first processor speed and a second mode that operates at a second processor speed that is faster than the first processor speed, and is included in the pattern matching method. Each stage can be above Executed in the first mode.

At this time, one of the standard input mode and the comparison input mode is a mode that is previously stored after the user inputs the first time, and the other is an input that is input at the second time after the first time. mode.

According to another aspect of the present invention, a touch IC is processed as an analog signal input to a touch screen device to determine a touch input position; and the related information of the determined touch input position is transmitted to a touch IC of another processing device, The included processor enables the following stages to proceed smoothly: as a sample of N standard tuples, including standard input coordinates and standard correlation vectors, respectively, to prepare N standard tuple stages; in the comparison input mode, as M samples Comparing tuples, respectively, including comparing input coordinates and comparing correlation vectors, preparing phases of M comparison tuples (but, M>=N); presuming that the above M comparison tuples match the above-mentioned N standard tuples Stages of N comparison tuples; stages of calculating the standard representative value of the standard correlation vector and the distance between the comparison representative values of the comparison correlation vectors according to the matched standard tuple and the comparison tuple pair; using the above N criteria The above distances of the N obtained by the tuple and the N comparison tuples, and using at least one of the distances to determine the standard input mode and Said phase comparing similarity between the input pattern.

According to another aspect of the present invention, a touch IC includes a processor that can smoothly perform the following steps: as the first information of the N samples sampled in the first input mode, respectively including the first coordinate and the first The preparation stage of the N first information of the vector; the M pieces of the second information sampled in the second input mode include the preparation stages of the M pieces of the second information of the second coordinate and the second vector (however, M >=N); presuming that the M pieces of the second information match the first information of the N pieces a stage of N pieces of second information; a stage of calculating a distance between a first representative value of the first vector and a second representative value of the second vector based on the first information and the second information pair that match each other; The N first pieces of information and the N distances obtained by the N pieces of second information are used to determine a degree of similarity between the first input mode and the second input mode by using at least one of the distances.

According to another aspect of the present invention, a pattern matching method includes: estimating, in a first input mode, a first estimation phase of a first displacement vector between one or more sampling first coordinates; and in a second input mode Predicting a second estimation stage of the second displacement vector between the second coordinates of one or more samples; a stage in which the first displacement vector and the second displacement vector match each other according to a predetermined rule; and the first displacement of the matching is completed The vector and the distance between the first representative value of the second displacement vector and the second representative value of the second displacement vector determine the similarity between the first input mode and the second input mode based on the distance stage.

In this case, the first displacement vector and the second displacement vector may represent displacements between a pair of coordinates in which the sampling time is adjacent to each other on the time axis.

In this case, the number of the first displacement vectors is smaller than the number of the second displacement vectors, and the first displacement vector may match only one of the second displacement vectors.

In this case, the number of the first displacement vectors is smaller than the number of the second displacement vectors, and at least one of the first displacement vectors may match a plurality of the second displacement vectors.

According to the present invention, it is provided that even if the processor operates at a slow speed, it can be provided Pattern matching method for accurate pattern matching results.

1‧‧‧ track

11‧‧‧Slow input mode (standard input mode)

12‧‧‧Quick input mode (comparison input mode)

100‧‧‧User appliances

110‧‧‧Frame

120‧‧‧Touch input device

R(1,1), R(2,2), R(3,3), R(4,4), R(5,5)‧‧‧ input coordinates (standard input coordinates)

S(1,1), S(2,2), S(3,3), S(4,4), S(5,5), S(6,6), S(7,7),S (8,8), S(9,9), S(10,10)‧‧‧ Input coordinates (comparison input coordinates)

S10, S20, S30, S40‧‧

T1~t10‧‧‧Time

TR(1,1), TR(2,2), TR(3,3), TR(4,4), TR(5,5)‧‧‧ tuples

VR(2,2), VR(3,3), VR(4,4), VR(5,5)‧‧‧ association vector (standard association vector)

VS(2,2), VS(3,3), VS(4,4), VS(5,5), VS(6,6), VS(7,7), VS(8,8),VS (9,9), VS(10,10)‧‧‧ association vector (comparison correlation vector)

Figure 1 shows an example of a touch input trajectory in a portable consumer appliance.

2 (including FIGS. 2A and 2B) is for explaining the sampling trend of the input coordinates of the input mode in the case where the gesture is completed.

3 (including FIGS. 3A, 3B, and 3C) is for explaining the method of defining "mode information" using the coordinates in accordance with the input coordinates obtained by the user gesture in order to perform pattern matching according to an embodiment of the present invention.

Figure 4 (comprising Figures 4A and 4B) is for explaining the interpretation of the defined association vector using the input mode in accordance with one embodiment of the present invention.

Figure 5 (comprising Figures 5A, 5B, 5C, and 5D) is for the purpose of explaining an advance processor for generating tuples in accordance with one embodiment of the present invention.

Figure 6 illustrates a comparison of standard tuples and comparison tuples as defined in accordance with one embodiment of the present invention.

Figure 7 illustrates a sequence diagram of an inter-input mode mode matching execution method performed in accordance with one embodiment of the present invention.

Hereinafter, the embodiments of the present invention will be explained and explained in detail with reference to the drawings. However, the present invention is not limited to the embodiments explained herein, and the present invention can be embodied in many different forms. The vocabulary used in the description of the present invention is to help This embodiment is not intended to limit the invention. In addition, the singular forms used hereinafter also include the plural forms.

Figure 1 shows an example of a touch input trajectory in a portable consumer appliance. The user appliance (100) of Figure 1 includes a frame (110) and a touch input device (120). The touch input device (120) may be in a state of overlapping with a screen not shown. Moreover, at this time, the touch input device (120) is actually in the visible light field, and can be made of transparent material according to the user. The track (1) is a path that causes a finger to touch the touch input device (120) and move. The above-described behavior of touching a finger to the touch input device (120) and moving it is referred to as a gesture. The above gestures can be completed quickly or slowly. For example, the same track (1), when drawing quickly, can be regarded as completing the "quick gesture", and when drawing slowly, it can be regarded as completing the "slow gesture". The above-mentioned user gestures, whether completed quickly or slowly, are referred to as "input modes." The input mode completed by the quick gesture is called "fast input mode", and the input mode completed by the slow gesture is called "slow input mode".

2A illustrates the input coordinate sampling trend of the input mode of the slow gesture completion, and FIG. 2B illustrates the input coordinate sampling trend of the input mode of the slow gesture completion.

In Figs. 2A and 2B, t1 to t10 refer to the sampling time of each input coordinate, and it is assumed that the operation is performed at the same processor speed.

Assuming that the user's appliance is operating at a certain processor speed, the same form of gesture is completed at different speeds, and different numbers of input coordinates are perceived during the completion of each gesture.

For example, Figure 2A illustrates the slow input mode (11) elapsed time t1~t10 In the case of completion, FIG. 2B illustrates the case where the fast input mode (11) is completed after the time t1 to t5. In Figure 2A, a slow input mode (11) can get a total of 10 input coordinates (input coordinate samples), and a fast input mode (12) in Figure 2B can get a total of 5 input coordinates.

The input coordinates obtained in the slow input mode (11) in Fig. 2A are S(i, i) (however, i = 1, 2, 3, ..., 10).

It is shown that the input coordinates obtained by the fast input mode (12) in Fig. 2B are represented by R(j, j) (however, j = 1, 2, 3, ..., 5).

<Information form presumed to perform pattern matching>

3 (including FIGS. 3A, 3B, and 3C) is for explaining the method of defining "mode information" using the coordinates in accordance with the input coordinates obtained by the user gesture in order to perform pattern matching according to an embodiment of the present invention.

The term "mode information" in the present specification refers to a special type of information recorded in the present invention in order to compare two input modes. In one embodiment of the invention, mode information referred to as a "tuple" can be defined and used in order to perform pattern matching. The above-mentioned "tuple" as the countable information can generate the same number as the number of input coordinates obtained from the input mode (11, 12). For example, in the fast input mode (12) of FIG. 2B, since five input coordinates R(j, j) are generated, five tuples can be generated.

The composition of each of the above tuples may include an input coordinate and an associated vector associated therewith. Here, the "association vector" can be obtained from the second input coordinate of the sample before the sampling of the first input coordinate. In particular, at this time, the first input coordinate and the associated correlation vector may be a displacement vector between the first input coordinate and the second input coordinate or a deformation value of the displacement vector. However, about the input mode The initial input coordinates of the initial sample cannot be used to define the input coordinates that existed before, so the displacement vector associated with the initial input coordinates described above does not exist. Therefore, it can be defined as a displacement vector that is not associated with the initial input coordinates described above. Hereinafter, the concept of the above tuple will be described by way of example.

3A illustrates that an associated vector VR(j,j) can be defined for each input coordinate R(j,j) of the fast input mode (12) of FIG. 2B (however, j=1, 2, 3,...,5). Moreover, FIG. 3B only represents the correlation vector VR(j, j).

The following [Table 1] shows an example of a tuple combination generated by the information obtained in Fig. 3A.

The above correlation vector may be an actual displacement value between two input coordinates, but the above actual displacement value may also be redefined by a "deformation vector" composed only of directivity. The generation method of the above deformation vector can be explained with reference to FIG. 3C.

Fig. 3C illustrates a standard direction for estimating the directivity of the above-described correlation vector. When the horizontal component of the correlation vector tends to the right side, it can assume a value of +1, and when it is inclined to the left side, it can take a value of -1. Further, the vertical component of the correlation vector may serve as the +1 value when it is inclined to the upper side, and may take the value of -1 when it is inclined to the lower side.

Therefore, the correlation vector of Table 1 can be redefined as the [Table 2] deformation vector.

In accordance with an embodiment of the invention, the above described correlation vector or deformation vector may be selected and used.

4A illustrates that an association vector VS(i,i) connected between respective input coordinates S(i,i) of the slow input mode (11) of FIG. 2A can be defined (however, i=1, 2) , 3,...,10). Moreover, FIG. 4A only shows the correlation vector VS(i, i). As with Figure 3, the information obtained in Figure 4A can define the same tuple combination as Table 3.

According to an embodiment of the present invention, in order to determine the similarity between the slow input mode (11) in FIG. 2A and the fast input mode (12) in FIG. 2B, FIG. 3 (including FIG. 3A, FIG. 3B and FIG. 3C) and FIG. 4 (including FIG. 4A and FIG. 4B) are related to the obtained different sets of tuples. Hereinafter, a method according to one embodiment of the present invention for judging the similarity of two input modes will be explained.

<Comparative method of speculative information in two modes>

Figure 5 (comprising Figures 5A, 5B, 5C, and 5D) is for the purpose of explaining an advance processor for generating tuples in accordance with one embodiment of the present invention.

As shown in Figure 5A, the user appliance (100) inputs two different gestures into the user input device. The first gesture is the slow input mode (11), and the second gesture is the fast input mode (12). The two gestures can be the same gesture that the same person intended to input, but it can be no. When the same gesture is intended, the overall image and size of the two gestures are similar, but the position of the center may be different.

Fig. 5B shows the input coordinates (S(i, i)) obtained by sampling from the slow input mode (11) in the slow input mode (11), and Fig. 5C is the sample obtained from the fast input mode (12). The input coordinates (R(j, j)) are indicated in the slow input mode (12) (i = 1 to 10, j = 1 to 5). Hereinafter, the input coordinates (S(i, i)) obtained by sampling from the slow input mode (11) are referred to as the first input coordinates, and the input coordinates obtained from the fast input mode (12) are sampled (R(j, j)) called the 2nd input coordinate

Figure 5D shows the position of the first input coordinate S(1,1) of the input coordinate (S(i,i)) and the first input coordinate R(1,1) of the input coordinate (R(j,j)). In the same state, the input coordinates (S(i, i)) and the input coordinates (R(j, j)) are arranged. View Looking at Figure 5D, it is enough to grasp the distance between the input coordinates (S(i, i)) and the input coordinates (R (j, j)).

In one embodiment of the invention, to determine the similarity of two input modes, one of the two input modes is treated as a "standard input mode" and the other is referred to as a "comparison input mode." at this time. The input mode with a small number of sampled input coordinates will be treated as the standard input mode and the other as the comparison input mode. In the embodiment of Fig. 5, the first input mode (11) is set to the comparison input mode, and the second input mode (12) is the standard input mode. Hereinafter, the tuple obtained from the standard input mode is referred to as a "standard tuple", and the tuple obtained from the comparison input mode is referred to as a "comparison tuple".

If the standard input mode has been determined in the two input modes, the respective standard tuple obtained from the standard input mode and the respective comparison tuple obtained in the comparison input mode will be set in advance according to an embodiment of the present invention. The "distance comparison rule" is compared. The above-described distance comparison rule is set in advance in order to find the distance between the input coordinates constituting each tuple.

Figure 6 is a graph showing the comparison results of the standard tuple and the comparison tuple in Figure 5D. The input coordinates constituting the standard tuple in Fig. 6, the correlation vector and the deformation vector are referred to as standard input coordinates, standard correlation vectors, and standard deformation vectors, respectively. In addition, the input coordinates constituting the comparison tuple, the correlation vector and the deformation vector are referred to as comparison input coordinates, respectively, and the correlation vector and the comparison deformation vector are compared.

Observed together with Figure 5D, and the standard input coordinates R (2, 2) belonging to TR(2,2), TR(3,3), TR(4,4), TR(5,5) in the standard tuple. ), R(3,3), R(4,4), R(5,5) are closest to the input coordinates S(3,3), S(6,6), S(8,8), S (10, 10). In addition, the comparison tuple with this comparison input coordinate is TS(3,3), TS (6, 6), TS (8, 8), TS (10, 10).

In one embodiment of the invention, a comparison tuple that is closest to a particular standard tuple may be defined as a comparison tuple having comparison input coordinates that are closest to the standard input coordinates to which the particular standard tuple belongs.

In one embodiment of the present invention, the comparison tuples that are closest to all standard tuples are extracted and then phased to correspond to each other, which is referred to as the "tuple matching phase" in the present specification. According to the above-mentioned tuple matching stage, when N standard tuples are given, N comparison tuples are extracted from M comparison tuples to match each other (however, M>=N).

<Method of generating representative value from vector for completion of pattern matching>

N pairs of standard tuple-comparison tuples are generated from the above N standard tuples, thus pattern matching in one embodiment according to the present invention is done as a standard.

In one embodiment of the present invention, the pattern matching may be performed using a correlation vector generated from a standard tuple and a comparison tuple or a representative value generated in a variable vector. At this time, N standard tuples can respectively define a "standard representative value", and N comparison tuples matching each standard tuple can also define a "comparative representative value". The above-mentioned standard representative value and comparison representative value may be collectively referred to as representative values.

At this time, the first sampled standard tuple and the comparison tuple can be assumed to exist in the same position, so that the two do not need to be compared. Therefore, the standard tuple of the first sample (for example: TR(1,1)) and the comparison tuple (for example, TS(1,1)) may also not define the standard representative value and the comparison representative value.

Table 4 shows the targets generated by the information obtained in Figure 3A. An example of a standard representative value of a standard association vector that is defined in a quasi-tuple. The standard representative value of the standard correlation vector is represented by RVR(,).

The representative value of the standard association vector can be the standard association vector itself. In addition, the representative value of the standard correlation vector may be the trend value of the standard correlation vector. At this time, the standard correlation vector itself is a special case of the trend value of the above-described standard correlation vector.

The "trend value" is a numerical value indicating the directivity of the standard correlation vector, and the specific example is the above-described deformation vector.

The above trend values can be defined by, for example, the manner in Table 5 below.

In Table 5, the directions of the arrows of the correlation vectors VR(j, j) and VS(i, i) appearing in FIG. 3B or FIG. 4B are separated by horizontal and vertical directions, and the directivity of the lateral portion is regarded as a trend value. The first value of the vertical portion is regarded as the second value of the trend value, and is performed according to this rule.

Tables 6, 7 and 8 show examples of comparative representative values of the comparison correlation vectors defined in the comparison tuple generated based on the information obtained in Fig. 4A. The standard representative value of the comparison correlation vector is represented by RVS(,). Examples of comparing the standard representative values of the correlation vectors can be given, for example, Examples 1, 2, 3, 4, etc. in Table 5, Table 6, and Table 7 below.

The first instance of the comparison representative value in Table 6 may be the comparison association vector that matches the standard association vector itself.

The second instance of the comparison representative value appearing in Table 6 may be the above-described trend value of the comparison correlation vector that matches the standard correlation vector.

The third example of the comparison representative value appearing in Table 7 is a comparison correlation vector matching the correlation standard correlation vector, and a representative value of the trend value of one or more comparison correlation vectors composed of the previously compared comparison correlation vectors.

In other words, the third instance of the comparison representative value shown in Table 7 is composed of 1 [comparison correlation vector matching the correlation standard association vector], and 2 [matching the standard correlation vector sampled before the correlation standard correlation vector] The correlation vector is compared, and then the comparison correlation vector of the sample is compared with the above-described correlation standard correlation vector, and the comparison correlation vector of the comparison correlation vector before sampling is compared with the representative value of the trend value of one or more comparison correlation vectors.

In the fourth example of the comparison representative value shown in Table 8, the correlation correlation vector matched by the correlation standard correlation vector and the value of the trend value of one or more comparison correlation vectors composed of the previously sampled comparison correlation vectors are all the same. In contrast, when all values are different, the above comparison represents a value of (0, 0).

An example of the standard representative values shown in Table 4 is used in the standard input mode shown in Figure 3A, as shown in Table 9.

An example of the comparison representative values shown in Table 6 is used in the comparison input mode shown in Figure 4A, as shown in Table 10.

An example of the comparison representative values shown in Table 7 is used in the comparison input mode shown in Figure 4A, as shown in Table 11.

Examples of comparative representative values shown in Table 8 are used in the Figure 4A. Compare the input modes as shown in Table 12.

In one embodiment of the present invention, the similarity between the standard input mode and the comparison input mode can be determined using the standard representative value and the comparative representative value described by Tables 4 to 12. There are two methods that can be used in general.

Method 1: As long as any one of the difference between the standard representative value and the comparison representative value (that is, the distance between the vectors) exceeds the previously set first critical value (TH1), the standard input mode and the comparison input can be judged. The patterns are not similar, otherwise they are judged to be similar.

Method 2: As long as any one of the difference between the standard representative value and the comparison representative value (that is, the distance between the vectors) exceeds the previously set second threshold value (TH2), the standard input mode and the comparison input can be judged. The patterns are not similar, otherwise they are judged to be similar.

The difference D() between the matching standard representative value and the comparative representative value shown in the above method 1 and the difference TD between the matching standard representative value and the comparative representative value shown in the method 2 are all sorted and shown in Table 13.

[Table 13]

The values in Table 13 are as shown in Table 14 when used in, for example, the standard input mode and the comparison input mode shown in FIGS. 3A and 4B.

Figure 7 illustrates a sequence diagram of an inter-input mode mode matching execution method performed in accordance with one embodiment of the present invention.

In one embodiment of the present invention, the following input-mode input matching method is accomplished using the idea of the present invention described above.

In the stage (S10), the standard input coordinates of the first time sampling and the standard input coordinates of the second time sampling from the first time before the first time, the standard correlation of the displacement vectors between the standard input coordinates obtained at the first time N standard tuples composed of values of the vector, during the input of the standard input mode, prepare N standard tuples obtained in N times different from each other.

In the stage (S20), the comparison input coordinate of the 1st time sampling and the comparison input coordinate of the 4th time sampling from the third time, the standard correlation vector of the displacement vector of the comparison input coordinate obtained at the third time The M comparison tuples composed of the values are input during the comparison input mode, and M comparison tuples obtained in M times different from each other are prepared.

In the stage (S30), according to the standard input coordinates indicating the N standard tuples, the distances of the comparison input coordinates belonging to the M comparison tuples are calculated, wherein the standard tuples and the comparison tuples that are closest to each other are performed. match.

In the stage (S40), the criterion is determined based on the distance between the standard representative value of the standard paired tuple and the comparison representative value of the standard paired vector in the pair of standard tuples and the comparison tuple in the phase (S30). The similarity between the input mode and the comparison input mode.

In the above stage (S40), the distance between the standard representative value of the standard paired tuple belonging to the matched N-pair standard tuple and the comparison tuple and the comparison representative value of the comparison-related vector is included, if at least one distance exceeds the preset The first critical value, it can be judged that the above standard input mode is not similar to the above comparative input mode. Stage (S41).

Further, in the above stage (S40), a representative value of the N distances between the standard representative value of the standard paired tuple belonging to the matched N-pair standard tuple and the comparison tuple and the comparison representative value of the comparison-related vector (ex) : Average value) If the second threshold value set in advance is exceeded, it is possible to determine a phase in which the standard input mode and the comparison input mode are not similar (S42).

The present invention can be variously modified and modified, and can be easily implemented without departing from the spirit and scope of the invention. The contents of each request item of the patent application scope can be combined with other request items having no reference relationship within the scope of the present specification.

11‧‧‧Slow input mode (standard input mode)

12‧‧‧Quick input mode (comparison input mode)

100‧‧‧User appliances

R(1,1), R(2,2), R(3,3), R(4,4), R(5,5)‧‧‧ input coordinates (standard input coordinates)

S(1,1), S(2,2), S(3,3), S(4,4), S(5,5), S(6,6), S(7,7),S (8,8), S(9,9), S(10,10)‧‧‧ Input coordinates (comparison input coordinates)

Claims (13)

A pattern matching method includes: in a standard input mode, as a sample of N standard tuples, respectively comprising a standard input coordinate and a standard correlation vector, preparing a phase of N standard tuples; in the comparison input mode, as a sampling M comparison tuples, respectively comprising a comparison input coordinate and a comparison correlation vector, preparing phases of M comparison tuples, where M>=N; presuming the M comparison metagroups and the N standard tuples a phase of matching N comparison tuples; a phase of calculating a standard representative value of the standard correlation vector and a distance between the comparison representative values of the comparison correlation vectors respectively according to the matched standard tuple and the comparison tuple pair; and using the said And determining the similarity between the standard input mode and the comparison input mode by using at least one of the distances of the N standard tuples and the N comparison tuples. a stage; wherein the comparison representative value is a representative value of the comparison correlation vector and one or more other comparison correlation vectors sampled before the comparison correlation vector is sampled. The pattern matching method of claim 1, wherein the speculating stage comprises: in a M comparison input coordinates belonging to the N comparison tuples, inferring to have a criterion subordinate to the N The N standard input coordinates in the tuple begin to the stage of the N comparison tuples of the most recent input input coordinates. A pattern matching method as described in claim 1, wherein The standard representative value is a vector indicating the up, down, left, and right direction trend values of the standard correlation vector; the comparison representative value is also a vector, and represents a trend value of the upper and lower directions of the comparison related vector. The pattern matching method according to claim 3, wherein the trend value of the up, down, left, and right directions of the standard correlation vector is the standard correlation vector, and the trend value of the up, down, left, and right directions of the comparison associative vector may be Compare the associated vectors. The mode matching method according to claim 1, wherein the phase of determining the similarity includes a phase, that is, if at least one of the obtained N distances exceeds a first threshold set in advance Value, it can be judged that the standard input mode and the comparison input mode are not similar. The mode matching method according to claim 1, wherein the phase of determining the similarity includes a phase, that is, if the representative value of the obtained N distances exceeds a second threshold value set in advance, It can be judged that the standard input mode and the comparison input mode are not similar. The pattern matching method according to claim 1, wherein each stage included in the pattern matching method processes an analog signal input in the touch screen device to determine a touch input position; and the determined touch input The transfer of location related information to other processing devices can be done in the touch integrated circuit. A pattern matching method as described in claim 7 of the patent application, wherein The touch integrated circuit includes a first mode that operates at a first processor speed and a second mode that operates at a second processor speed that is faster than the first processor speed, in the pattern matching method The various stages involved can be performed in the first mode. The pattern matching method according to claim 1, wherein in the standard input mode and the comparison input mode, one of them is a mode that is stored in advance according to a user input at the first time, and the other is The input mode that is input at the second time after the first time. A touch integrated circuit for processing an analog signal input in a touch screen device, determining a touch input position, and transmitting related information of the determined touch input position to another processing device, wherein the touch integrated circuit includes The processor is configured to perform the following steps: as a sample of N standard tuples, including standard input coordinates and standard correlation vectors, respectively, to prepare N standard tuple stages; in the comparison input mode, as comparison M groups of samples , respectively, including comparing the input coordinates and comparing the correlation vectors, preparing M comparison tuple stages, where M>=N; presuming N comparisons of the M comparison tuples matching the N standard tuples a stage of the tuple; a stage of calculating a standard representative value of the standard correlation vector and a distance between the comparison representative values of the comparison correlation vector according to the matched standard tuple and the comparison tuple pair; and using the N standard elements N obtained by the group and the N comparison tuples And the phase of determining the degree of similarity between the standard input mode and the comparison input mode using at least one of the distances; wherein the comparison representative value is the comparison correlation vector and the comparison correlation vector is A representative value of one or more other comparison correlation vectors sampled before sampling. A touch integrated circuit for processing an analog signal input in a touch screen device, determining a touch input position, and transmitting related information of the determined touch input position to other processing devices, wherein the touch integrated circuit further For performing, the N pieces of first information sampled in the first input mode include preparation stages of N pieces of first information of the first coordinate and the first vector, and Ms sampled in the second input mode. The second information includes the preparation stages of the M second information of the second coordinate and the second vector, where M>=N; and the first of the M second information and the N first a stage of N pieces of second information matching the information; and a stage of calculating a distance between the first representative value of the first vector and the second representative value of the second vector based on the first information and the second information pair that match each other; And determining the first input mode and the second input mode by using at least one of the N distances obtained by using the N first information and the N second information. Phase of similarity; wherein the second representative value The vector and the second vector is prior to said second samples are sampled, a representative value other than the second vector. A pattern matching method includes: estimating, in the first input mode, the first of the first coordinates of one or more samples a first estimation phase of the displacement vector; in the second input mode, a second estimation phase of the second displacement vector between the second coordinates of one or more samples is estimated; the first displacement vector and the second displacement vector are in accordance with a stage in which the rules set in advance match each other; a distance between the first displacement vector of the matching and the first representative value of the second displacement vector and the second representative value of the second displacement vector; a stage of determining a similarity between the first input mode and the second input mode; wherein the number of the first displacement vectors is smaller than the number of the second displacement vectors, and the first displacement At least one of the vectors matches a plurality of the second displacement vectors. The pattern matching method according to claim 12, wherein the first displacement vector and the second displacement vector represent displacements between a pair of coordinates whose sampling times are adjacent to each other on a time axis.