JP2016031672A - Input device, finger determination method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an input device with which it is possible to determine whether an object in contact with or close to an operation surface is a finger or not with high accuracy requiring a small amount of calculation, a method for the finger determination, and a program.SOLUTION: The two-dimensional distribution of detection data in an area of the operation surface of a sensor unit 10 which an object is in contact with or close to is assumed to be a two-dimensional normal distribution. The contour of the area of the operation surface which the object is in contact with or close to is approximated to a virtual ellipse that is a set of points on a two-dimensional plane that satisfy a condition that a probability density function f(x, y) representing the two-dimensional normal distribution equals a certain value. When the area of the operation surface which the object is in contact with or close to is specified, a first evaluation value E1 corresponding to a ratio between the long and short axes of the virtual ellipse is calculated on the basis of the detection data included in the specified area and its detection position. Depending on whether the first evaluation value E1 is within a prescribed range, determination is made as to whether the object in contact with or close to the area is a finger.SELECTED DRAWING: Figure 6

Description

  The present invention relates to an input device used for inputting information in various information devices such as a computer, and in particular, specifies an area where an object such as a finger or a pen is in contact with or close to an operation surface, and is based on the specified area. The present invention relates to an input device for inputting information.

  As input interfaces for information devices such as notebook PCs, tablet terminals, and smartphones, devices such as touch pads and touch panels equipped with sensors that detect contact positions of objects such as fingers and pens are widely used. There are various types of sensors that detect the contact position of an object, such as a resistive film type and a capacitance type, but in recent years, an electrostatic that can handle "multi-touch" that detects multiple contact points. The adoption of capacitive sensors is increasing.

  Patent Document 1 below describes a multi-touch digitizer system that detects a contact position of a finger or the like using a two-dimensional sensor matrix that detects a change in capacitance between conductive lines orthogonal to a grid.

JP 2011-501261 A

  By the way, in the multi-touch type sensor, a case where a fingertip touches the operation surface and a case where a part of the hand that is not the fingertip (such as a palm or a belly of a finger) or another object touches the operation surface are correctly identified. Is required. If it cannot be correctly determined whether or not the object touching the operation surface is a fingertip, an instruction unintended by the user is likely to be input, and usability is reduced.

  Japanese Patent Application Laid-Open No. 2004-228561 describes a method of using an aspect ratio of a touch area in order to identify whether or not the touch area detected by the two-dimensional sensor matrix is a user's fingertip. It is known that when the touch area is not due to the fingertip, the shape is likely to be a long and narrow ellipse compared to the touch area due to the fingertip. In the method of Patent Document 1, the feature as an elliptical shape is estimated based on the ratio (shape scale) of the number of conductive lines in the long axis to the number of conductive lines in the short axis of the touch region. However, when the major axis and the minor axis of the ellipse are obliquely inclined with respect to the direction of the conductive wire, there is a problem that the shape described as the ellipse cannot be correctly estimated by the method described in Patent Document 1. That is, even if the shape of the actual touch area is a long and narrow ellipse, if the inclination of the major axis and the minor axis with respect to the conductive line is close to 45 degrees, it is estimated as a shape close to a circle.

  In order to improve the inaccuracy of the estimation of the ellipse, for example, a method of deriving an ellipse equation that approximates the outline of the touch area by using the least square method, the Gauss-Newton method, or the like can be considered. However, these methods for deriving an elliptic equation have a problem that the calculation amount is large and the calculation time is long, and hardware with high calculation capability is required. In addition, since the elliptic equation is derived only from the data near the outline of the touch area, if the data near the outline is locally disturbed by noise or the like, a large error occurs in the estimation of the elliptic equation. There is also a problem.

  The present invention has been made in view of such circumstances, and an object thereof is an input device capable of accurately determining whether an object in contact with or close to an operation surface is a fingertip with a small amount of calculation, and a finger determination method thereof And providing a program.

  A first aspect of the present invention relates to an input device that inputs information according to contact or proximity of an object to an operation surface. The input device includes a sensor unit that detects the degree of contact or proximity of an object at a plurality of detection positions distributed on the operation surface, and a plurality of positions on the operation surface based on a detection result of the sensor unit. A two-dimensional data generation unit for generating two-dimensional data including a plurality of detection data indicating the degree of contact or proximity of the object, and an area on the operation surface where the object has contacted or approached based on the two-dimensional data An area specifying unit that specifies the first evaluation value according to a ratio between a major axis and a minor axis of a virtual ellipse corresponding to the area specified by the area specifying unit, and the first When the evaluation value is within the first range, it is determined that the object that is in contact with or close to the operation surface is the fingertip, and when the first evaluation value is not within the first range, the object is not the fingertip Judge And a determination unit. When the virtual ellipse is assumed that the two-dimensional distribution of the detection data in the region specified by the region specifying unit is a two-dimensional normal distribution, the probability density function representing the two-dimensional normal distribution is a constant value. A set of points on a two-dimensional plane that satisfy the same condition. The calculation unit calculates the first evaluation value based on the detection data included in the specified region and its detection position.

In the invention according to the first aspect, a two-dimensional distribution of the detection data in a region where an object is in contact with or close to the operation surface is assumed to be a two-dimensional normal distribution. The contour of the area where the object is in contact with or close to the operation surface is approximated by a virtual ellipse that is a set of points on a two-dimensional plane that satisfies the condition that the probability density function representing the two-dimensional normal distribution is equal to a constant value. The When an area on the operation surface that is in contact with or close to an object is specified, the major axis and the minor axis of the virtual ellipse are determined based on the detection data and the detection position included in the specified area. The first evaluation value corresponding to the ratio is calculated. It is determined whether or not the object is a fingertip depending on whether or not the first evaluation value is within the first range.
Therefore, the first evaluation value representing the feature of the shape of the ellipse is calculated without performing a complicated and computationally intensive process for deriving the equation of the ellipse.
In addition, the first evaluation value calculated based on the detection data included in the region on the operation surface that is in contact with or close to the object and the detection position is based only on information near the contour line of the region. Compared to the derived ellipse equation, the ellipse shape features are represented with high accuracy. Therefore, the fingertip determination accuracy is improved as compared with the method of deriving an elliptic equation.

Preferably, the calculation unit may calculate a predetermined parameter in the probability density function based on the detection data included in the specified region. The calculation unit may calculate the first evaluation value based on the calculated parameter.
According to the above configuration, since the first evaluation value is calculated based on a predetermined parameter in the probability density function representing a two-dimensional normal distribution, the amount of calculation is reduced as compared with the case where an elliptic equation is derived.

For example, the detection position of the detection data may be specified by the coordinate value of the first coordinate axis and the coordinate value of the second coordinate axis in the orthogonal coordinate system set on the operation surface. The calculation unit uses the coordinate value of the first coordinate axis as a first random variable of the probability density function, the coordinate value of the second coordinate axis as a second random variable of the probability density function, and the probability density function detects the detection In the case of having a value according to the data, the first parameter according to the variance for the first random variable, the second parameter according to the variance for the second random variable, and the first random variable and the first A third parameter corresponding to the covariance for the two random variables may be calculated. The calculation unit may calculate the first evaluation value based on the calculated first parameter, second parameter, and third parameter.
In this case, the calculation unit adds the value obtained by squaring the difference between the first parameter and the second parameter and a value four times the third parameter to obtain the first parameter and the second parameter. The first evaluation value according to the numerical value obtained by dividing the sum of the two by the squared value may be acquired.

Preferably, the calculation unit may calculate a second evaluation value corresponding to the area of the virtual ellipse based on the first parameter and the second parameter. The determination unit determines that an object that is in contact with or close to the operation surface is a fingertip when the first evaluation value is within the first range and the second evaluation value is within the second range. If the first evaluation value is not within the first range or the second evaluation value is not within the second range, it may be determined that the object is not a fingertip.
In this case, the calculation unit may calculate the second evaluation value corresponding to the sum of the first parameter and the second parameter.
According to said structure, since the area of the said virtual ellipse is further considered in determination of a fingertip, the precision of the said determination improves. In addition, since the second evaluation value corresponding to the area of the virtual ellipse is calculated based on the first parameter and the second parameter, an increase in calculation amount accompanying the calculation of the second evaluation value is small. Can be suppressed.

  According to a second aspect of the present invention, a computer that has input detection results of sensors that respectively detect the degree of contact or proximity of an object at a plurality of detection positions distributed on the operation surface is based on the detection result. The present invention relates to a finger determination method for determining whether or not is a fingertip. The finger determination method generates two-dimensional data including a plurality of detection data indicating the degree of contact or proximity of an object at a plurality of positions on the operation surface based on a detection result of the sensor; Based on the two-dimensional data, a step of identifying a region on the operation surface that is in contact with or close to an object, and a ratio corresponding to a ratio between a major axis and a minor axis of a virtual ellipse corresponding to the identified region. If the first evaluation value is within the first range and the first evaluation value is within the first range, it is determined that the object in contact with or close to the operation surface is a fingertip, and the first evaluation value is the first evaluation value. And determining that the object is not a fingertip. The virtual ellipse has a constant probability density function representing the two-dimensional normal distribution when the two-dimensional distribution of the detection data in the region specified in the step of specifying the region is assumed to be a two-dimensional normal distribution. A set of points on the two-dimensional plane that satisfy the condition equal to the value. In the step of calculating the first evaluation value, the first evaluation value is calculated based on the detection data included in the specified region and its detection position.

  The invention according to the third aspect of the present invention is a program for causing a computer to execute each step in the finger determination method according to the second aspect.

  According to the present invention, it is possible to accurately determine whether or not an object that is in contact with or close to the operation surface is a fingertip with a small amount of calculation, as compared with a method of deriving an equation of a contour line of a region in contact with or close to an object. .

It is a figure which shows an example of a structure of the input device which concerns on embodiment of this invention. It is a figure which shows an example of the graph of the probability density function showing a two-dimensional normal distribution. It is a flowchart for demonstrating operation | movement of an input device. It is a figure which shows the example of binarization of two-dimensional data. FIG. 4A shows two-dimensional data before binarization, and FIG. 4B shows two-dimensional data after binarization. It is a figure which shows the example in which the contact / proximity | contact area | region of an object is specified based on the binarized two-dimensional data. FIG. 5A shows binarized two-dimensional data, and FIG. 5B shows a state where the contact / proximity region of the object is specified. It is a flowchart for demonstrating a finger determination process. It is a flowchart for demonstrating the other example of a finger determination process.

FIG. 1 is a diagram illustrating an example of a configuration of an input device according to an embodiment of the present invention.
The input device illustrated in FIG. 1 includes a sensor unit 10, a processing unit 20, a storage unit 30, and an interface unit 40. The input device according to the present embodiment is a device that inputs information according to the position of contact or proximity by bringing an object such as a finger or pen into contact or proximity with an operation surface provided with a sensor. Note that “proximity” in the present specification includes both being close in a contact state and being close in a non-contact state.

[Sensor unit 10]
The sensor unit 10 detects the degree of contact or proximity of an object such as a finger or a pen at a plurality of detection positions distributed on the operation surface. For example, the sensor unit 10 includes a sensor matrix 11 in which capacitors (capacitive sensor elements) 12 whose capacitance changes according to the proximity of an object are formed in a matrix, and detection data corresponding to the capacitance of the capacitor 12. A detection data generation unit 13 to generate and a drive unit 14 to apply a drive voltage to the capacitor 12 are included.

  The sensor matrix 11 includes a plurality of drive electrodes Lx extending in the vertical direction and a plurality of detection electrodes Ly extending in the horizontal direction. The plurality of drive electrodes Lx are arranged in parallel in the horizontal direction, and the plurality of detection electrodes Ly are arranged in parallel in the vertical direction. The plurality of drive electrodes Lx and the plurality of detection electrodes Ly intersect in a lattice pattern and are insulated from each other. A capacitor 12 as a capacitive sensor element is formed near the intersection of the drive electrode Lx and the detection electrode Ly. In the example of FIG. 1, the shape of the electrodes (Lx, Ly) is drawn in a strip shape, but other arbitrary shapes (such as a diamond pattern) may be used.

  The drive unit 14 is a circuit that applies a drive voltage to each capacitor 12 of the sensor matrix 11. Specifically, the drive unit 14 selects one drive electrode Lx in order from the plurality of drive electrodes Lx according to the control of the processing unit 20, and periodically changes the potential of the selected one drive electrode Lx. . When the potential of the drive electrode Lx changes within a predetermined range, the drive voltage applied to the capacitor 12 formed near the intersection of the drive electrode Lx and the detection electrode Ly changes within the predetermined range. Charging or discharging occurs.

  The detection data generation unit 13 generates detection data corresponding to the charge transmitted through each detection electrode Ly when the capacitor 12 is charged or discharged with the application of the drive voltage by the drive unit 14. That is, the detection data generation unit 13 samples the charge transmitted through each detection electrode Ly at a timing synchronized with the periodic change of the drive voltage of the drive unit 14, and generates detection data according to the sampling result. To do.

For example, the detection data generation unit 13 converts the output signal of the capacitance-voltage conversion circuit (CV conversion circuit) that outputs a voltage corresponding to the capacitance of the capacitor 12 into a digital signal. An analog-digital conversion circuit (AD conversion circuit) that outputs data is included.
The CV conversion circuit samples the charge transmitted through the detection electrode Ly according to the control of the processing unit 20 every time the driving voltage of the driving unit 14 periodically changes and the capacitor 12 is charged or discharged. Specifically, every time positive or negative charge is transmitted at the detection electrode Ly, the CV conversion circuit transfers this charge or a charge proportional thereto to the reference capacitor and generates it in the reference capacitor. A signal corresponding to the voltage is output. For example, the CV conversion circuit outputs a signal corresponding to an integrated value or an average value of charges periodically transmitted through the detection electrode Ly or charges proportional thereto. The AD conversion circuit converts the output signal of the CV conversion circuit into a digital signal at a predetermined period according to the control of the processing unit 20 and outputs it as detection data.

  In addition, although the sensor part 10 shown in the above-mentioned example detects the proximity | contact of an object by the change of the electrostatic capacitance (mutual capacitance) which arises between electrodes (Lx, Ly), it is not restricted to this example, others The proximity of the object may be detected by various methods. For example, the sensor unit 10 may be a system that detects a capacitance (self-capacitance) generated between the electrode and the ground due to the approach of an object. In the case of a method for detecting self-capacitance, a drive voltage is applied to the detection electrode. Further, the sensor unit 10 is not limited to the electrostatic capacity method, and may be, for example, a resistance film method or an electromagnetic induction method.

[Processing unit 20]
The processing unit 20 is a circuit that controls the overall operation of the input device, and includes, for example, a computer that performs processing according to an instruction code of a program stored in the storage unit 30, and a logic circuit that implements a specific function. Composed. All of the processing of the processing unit 20 may be realized based on a program in a computer, or part or all of the processing may be realized by a dedicated logic circuit.

  In the example of FIG. 1, the processing unit 20 includes a timing control unit 21, a two-dimensional data generation unit 22, a region specifying unit 23, a calculation unit 24, a determination unit 25, and a coordinate calculation unit 26.

  The timing control unit 21 controls the detection timing in the sensor unit 10. Specifically, the timing control unit 21 selects the drive electrode and generates the pulse voltage in the drive unit 14, and selects the detection electrode and generates the detection data in the detection data generation unit 13 at appropriate timing. In addition, these circuits are controlled.

  The two-dimensional data generation unit 22 generates two-dimensional data including a plurality of detection data indicating the degree of contact or proximity of an object at a plurality of positions on the operation surface based on the detection result of the sensor unit 10, and a storage unit It is stored in 30 two-dimensional data memory 31.

  For example, the two-dimensional data generation unit 22 stores the detection data output from the sensor unit 10 in a storage area (current value memory) of the storage unit 30 as two-dimensional data in a matrix format. The two-dimensional data generation unit 22 calculates a difference between each detection data of the two-dimensional data stored in the current value memory and each detection data of the two-dimensional data stored in advance in another storage area (reference value memory) of the storage unit 30. Are calculated for the detection data corresponding to each other. The two-dimensional data generation unit 22 stores these calculation results in the two-dimensional data memory 31 of the storage unit 30 in a matrix format.

  In the reference value memory, detection data output from the sensor unit 10 in a state where nothing touches or is close to the operation surface is stored in advance. Therefore, the two-dimensional data written to the two-dimensional data memory 31 by the two-dimensional data generation unit 22 represents the amount of change from a state where the object is not in contact with or close to the operation surface.

  Note that the two-dimensional data generated by the two-dimensional data generation unit 22 is not limited to data representing the amount of change from the non-contact state as described above, and may be the same as the detection data output by the sensor unit 10.

  Based on the two-dimensional data written in the two-dimensional data memory 31 by the two-dimensional data generation unit 22, the region specifying unit 23 specifies a region on the operation surface that is in contact with or close to the object, and stores information on the specified region. The data is stored in the area information memory 33 of the storage unit 30.

  For example, the area specifying unit 23 converts each detection data included in the two-dimensional data stored in the two-dimensional data memory 31 of the storage unit 30 into binarized data (ON Data or off-data), and the conversion result is stored in the two-dimensional data memory 32 of the storage unit 30. Specifically, the region specifying unit 23 compares the detection data in the two-dimensional data memory 31 with a predetermined threshold value, and converts it into binary data having a value corresponding to the comparison result.

  When binarized two-dimensional data is obtained, the region specifying unit 23 specifies a contact region or a close region of each object on the operation surface based on the two-dimensional data. Specifically, the area specifying unit 23 scans the binarized two-dimensional data in order from the end, and searches for binarized data (on data) indicating that the object is in contact with or close to the area. When the on-data is found by scanning, the area specifying unit 23 tracks the outline of the area where the on-data is gathered, starting from the on-data. When returning to the starting point again as a result of the contour tracking, the region specifying unit 23 specifies the region surrounded by the closed contour as the contact / proximity region of the object, and assigns a unique label to the region. Then, the region specifying unit 23 changes the binarized data belonging to the region to which the unique label is assigned to off-data, and scans the binarized two-dimensional data again sequentially from the end. When the on-data is found again by scanning, the region specifying unit 23 performs contour tracking in the same manner as described above, specifies a collection region of on-data, and assigns a unique label. By repeating this process, the area specifying unit 23 specifies the contact / proximity area of the object on the operation surface, and assigns a unique label to each area.

  The calculation unit 24 calculates a first evaluation value E1 corresponding to the ratio between the major axis and the minor axis of a virtual ellipse corresponding to the region identified by the region identifying unit 23. This virtual ellipse is a condition that the probability density function representing the two-dimensional normal distribution is equal to a constant value when the two-dimensional distribution of detection data in the region specified by the region specifying unit 23 is assumed to be a two-dimensional normal distribution. A set of points (contour lines) on a two-dimensional plane that satisfies The calculation unit 24 calculates a first evaluation value E1 representing the feature of this virtual ellipse shape based on the detection data included in the region specified by the region specifying unit 23 and its detection position.

  FIG. 2 is a diagram illustrating an example of a graph of a probability density function representing a two-dimensional normal distribution. “F (x, y)” indicates a probability density function of a two-dimensional normal distribution, and “x” and “y” indicate random variables. In the example of FIG. 2, the coordinate values of the x-axis and y-axis constituting the orthogonal coordinate system correspond to the random variables x and y, and the coordinate value of the coordinate axis perpendicular to the x-axis and y-axis is the probability density function f ( x, y).

  The distribution of detection data (data indicating the contact degree or proximity detection result of an object) in a region where the fingertip is in contact with or in proximity tends to approximate a two-dimensional normal distribution as shown in FIG. That is, when the coordinate value of the x axis in the orthogonal coordinate system set on the operation surface of the sensor unit 10 is the random variable x and the coordinate value of the y axis in the orthogonal coordinate system is the random variable y, the orthogonal coordinate system The detected data at the coordinates (x, y) of the above has a tendency to have a value corresponding to the probability density function f (x, y) of the two-dimensional normal distribution.

  Hereinafter, it will be described that the contour line of the probability density function of the two-dimensional normal distribution becomes an ellipse as shown in the lower side of FIG. The ratio between the major axis and the minor axis of the ellipse can be calculated by a parameter (parameter) that defines the probability density function f (x, y) of the two-dimensional normal distribution.

  The probability density function f (X) of the two-dimensional normal distribution is expressed by the following equation.

  In Expression (1), “X” represents a random variable matrix, “μ” represents an average value matrix, and “S” represents a variance-covariance matrix. This random variable matrix X, mean value matrix μ, and variance-covariance matrix S are expressed by the following equations.

In formula (3), “μ _x ” indicates an average value for the random variable x, and “μ _y ” indicates an average value for the random variable y. Further, “σ _x ² ” in Equation (4) indicates the variance for the random variable x, “σ _y ² ” indicates the variance for the random variable y, and “σ _xy ” indicates the common for the random variables x and y. Indicates variance. The variance σ _x ² , the variance σ _y ² , and the covariance σ _xy are each expressed by the following equations.

  In Expressions (5) to (7), “Z (x, y)” indicates each sample value of two-dimensional data approximated to a two-dimensional normal distribution, and is stored in the two-dimensional data memory 31 in this embodiment. Corresponds to each detection data of the two-dimensional data.

The inverse matrix S ⁻¹ and the determinant | S | of the variance-covariance matrix S in Equation (1) are represented by the following equations, respectively.

  When the equations (8) and (9) are substituted into the equation (1) and rearranged, the probability density function f (x, y) is expressed by the following equation.

  Equation (10) can be further transformed into the following equation.

  The constant “ρ” in Expression (11) is expressed by the following expression.

  Assuming that the probability density function f (x, y) shown in the equation (11) is equal to a constant value c, the equation (11) can be transformed as the following equation.

Here, the random variables x and y are replaced with variables x _e and y _e shown in the following equation.

  Also, since the left side in equation (13) is a constant, it is set as “C”. Then, the expression (11) can be further modified as follows.

  Expression (16) represents an ellipse equation in which the major axis and the minor axis are inclined with respect to the coordinate axes (x axis, y axis) of the orthogonal coordinate system. Accordingly, it can be seen that the contour line of the probability density function of the two-dimensional normal distribution is an ellipse as shown in the lower two-dimensional plane of FIG.

When the covariance σ _xy is zero, the constant “ρ” is zero from the relationship of the equation (12). If “ρ” is zero in the equation (11), the elliptic equation shown in the equation (11) is as follows.

Equation (17) is an elliptic equation in which the major axis and the minor axis are not inclined with respect to the coordinate axes (x axis, y axis) of the orthogonal coordinate system (the major axis and the minor axis are parallel or orthogonal to the coordinate axes). Represent. Therefore, the covariance σ _xy is considered to be a parameter related to the inclination of the ellipse with respect to the coordinate axis of the orthogonal coordinate system.

  Here, it is assumed that an ellipse that is tilted with respect to the coordinate axis is formed by rotating the coordinate axis with respect to an ellipse that is not tilted with respect to the coordinate axis, as represented by Expression (16). If the variables in the coordinate system without the inclination of the ellipse are “u” and “v”, this rotational transformation is expressed as follows.

  Further, the conversion between the random variable X and the random variable Y can be expressed as follows.

  In Expression (19), “a” represents a constant vector, and “B” represents a coefficient matrix.

  The average value matrix μ for the random variable X is expressed by the following equation.

  The variance-covariance matrix Σ for the random variable X is expressed by the following equation.

Mean value matrix mu _Y for random variable Y is expressed by the following equation.

Variance-covariance matrix sigma _Y for random variable Y is expressed by the following equation.

The coefficient matrix B on the right side in Expression (23) can be regarded as a coefficient matrix for rotational transformation on the right side of Expression (18). On the other hand, the variance-covariance matrix Σ in the two-variable normal distribution is expressed using variance and covariance as shown in Equation (4). If the variance for the variable u in the coordinate system without the inclination of the ellipse is “σ _u ² ”, the variance for the variable v is “σ _v ² ”, and the covariance for the variables u and v is “σ _uv ”, Since it is considered that the covariance “σ _uv ” becomes zero when there is no inclination, the equation (23) becomes as follows.

  From the equation (24), the following equation is obtained.

  When the equation (26) is subtracted from the equation (25), the following equation is obtained.

  Further, from the equation (24), the following equation is obtained.

  When formula (25) and formula (26) are added, the following formula is obtained.

  When the equation (27) is squared, the following equation is obtained.

  When the equation (28) is squared, the following equation is obtained.

  When the square roots of both sides are obtained by adding Expression (30) and Expression (31), the following expression is obtained.

  From the equations (29) and (32), the following equation is obtained.

  From the equations (33) and (34), the ratio R between the major axis and the minor axis of the ellipse is expressed as the following equation.

As shown in Expression (35), the ratio R between the major axis and the minor axis of the ellipse is a parameter (σ _x ² , σ _y ² , σ) that defines the probability density function f (x, y) of the two-dimensional normal distribution. _xy ).

  In addition, although calculation of a square root is required in Formula (35), calculation of a square root can be avoided by simplifying this formula as follows.

  Assuming that the square root in Expression (36) is the first evaluation value E1, the first evaluation value E1 is expressed by the following expression.

  The first evaluation value E1 represented by Expression (37) also has a value corresponding to the ratio R between the major axis and the minor axis of the ellipse.

Based on the detection data included in the region specified by the region specifying unit 23 and its detection position, the calculation unit 24 uses the parameters σ _x ² , σ _y ² , σ represented by the equations (5) to (7). _xy is calculated. And the calculation part 24 calculates the 1st evaluation value E1 represented by Formula (37) based on these calculated parameters.

Since the denominator term (the total value of the detection data included in the region) in Equations (5) to (7) is reduced in Equation (37), when calculating Equation (37), No term calculation is required. Therefore, the calculation unit 24 may calculate only the numerator term in the equations (5) to (7) and calculate the first evaluation value E1 using the calculation result.
The above is the description of the calculation unit 24.

Returning to FIG.
When the first evaluation value E1 calculated by the calculation unit 24 is within a predetermined range (within the first range), the determination unit 25 determines that an object that is in contact with or close to the operation surface is a fingertip. If the 1 evaluation value E1 is not within this range, it is determined that the object is not a fingertip.

Based on the contact area or proximity area of the object specified by the area specifying unit 23, the coordinate calculation unit 26 calculates coordinates on the operation surface where the object has contacted or approached.
For example, the coordinate calculation unit 26 creates profile data for each of the horizontal direction (direction in which the electrode Ly is arranged) and the vertical direction (direction in which the electrode Ly is arranged) of the region specified by the region specifying unit 23. The profile data in the horizontal direction is obtained by calculating the sum of a group of detection data in the vertical direction of the operation surface for each column and arranging the sum of the detection data in the horizontal direction of the operation surface. The profile data in the vertical direction is obtained by calculating the sum of a group of detection data in the horizontal direction of the operation surface for each row and arranging the sum of the detection data in the order of the vertical direction of the operation surface. The coordinate calculation unit 28 calculates the peak position and the center of gravity position of the detection data for each of the horizontal profile data and the vertical profile data. The position in the horizontal direction and the position in the vertical direction obtained by this calculation represent coordinates at which the object is in contact with or close to the operation surface. The coordinate calculation unit 28 stores the coordinate data obtained by such calculation in the object coordinate memory 34 of the storage unit 30.

  Note that the coordinate calculation unit 26 may calculate only the coordinates of the region determined as the fingertip by the determination unit 25 among the regions specified by the region specifying unit 23.

[Storage unit 30]
The storage unit 30 stores constant data and variable data used for processing in the processing unit 20. When the processing unit 20 includes a computer, the storage unit 30 may store a program executed on the computer. The storage unit 30 includes, for example, a volatile memory such as a DRAM or SRAM, a nonvolatile memory such as a flash memory, a hard disk, or the like.

[Interface unit 40]
The interface unit 40 is a circuit for exchanging data between the input device and another control device (such as a control IC for an information device equipped with the input device). The processing unit 20 outputs information (such as object coordinate information and the number of objects) stored in the storage unit 30 from the interface unit 40 to a control device (not shown). Further, the interface unit 40 acquires a program executed in the computer of the processing unit 20 from a disk drive device (not shown) (device that reads a program recorded in a non-temporary recording medium), a server, or the like, and the storage unit 30 You may load it.

  Here, the operation of the input device shown in FIG. 1 having the above-described configuration will be described with reference to the flowchart of FIG. For example, the input device repeats the operation shown in the flowchart of FIG. 3 at regular intervals, and acquires information on the contact / proximity position of the object on the operation surface.

ST100:
The timing control unit 21 of the processing unit 20 controls the sensor unit 10 so that detection data is obtained at a plurality of detection positions distributed over the entire operation surface. The driving unit 14 of the sensor unit 10 sequentially selects a plurality of driving electrodes of the sensor matrix 11 according to the control of the timing control unit 21 and applies a pulse voltage. The detection data generation unit 13 selects a plurality of detection electrodes in the sensor matrix 11 in order each time one drive electrode is selected and driven, and the capacitive sensor element 12 at the intersection of the drive electrode and the detection electrode. A voltage signal corresponding to the electrostatic capacity is generated, and the voltage signal is converted into detection data having a predetermined bit length and output to the processing unit 20.

ST105:
The two-dimensional data generation unit 22 of the processing unit 20 stores the detection data sequentially output from the sensor unit 10 in a predetermined storage area (current value memory) of the storage unit 30 as two-dimensional data in a matrix format. The two-dimensional data generation unit 22 calculates a difference between each detection data of the two-dimensional data stored in the current value and each detection data of the two-dimensional data stored in advance in another storage area (reference value memory) of the storage unit 30. The calculation results are stored in the two-dimensional data memory 31 as two-dimensional data.

ST110:
The region specifying unit 23 of the processing unit 20 compares each detection data in the two-dimensional data memory 31 with a predetermined threshold value, converts the detected data into binary data having a value corresponding to the comparison result, and converts the two-dimensional data Store in the memory 32. FIG. 4 is a diagram illustrating an example of binarization of two-dimensional data. In the two-dimensional data shown in FIG. 4A, each detection data has a numerical value of 0 or more. In the two-dimensional data after binarization shown in FIG. 4B, detection data exceeding the threshold 50 is converted to “1” (on data), and detection data smaller than the threshold 50 is “0” (off data). ).
When the binarized two-dimensional data is obtained in this way, the region specifying unit 23 next performs contour tracking and labeling on the binarized two-dimensional data. FIG. 5 is a diagram illustrating an example in which a contact / proximity region of an object is specified based on binarized two-dimensional data. FIG. 5A shows binarized two-dimensional data, and FIG. 5B shows a state where the contact / proximity region of the object is specified. The region specifying unit 23 specifies each region included in the closed contour by contour tracking, and assigns a unique label to each specified region. In the example of FIG. 5B, numerical values “1” to “3” are assigned to the three specified areas.

ST115:
When the contact / proximity position of one or more objects is identified in step ST110, the processing unit 20 proceeds to step ST120, and when no contact / proximity position is identified, the process ends.

ST120:
The processing unit 20 performs a finger determination process for determining whether or not the region identified in step ST110 is due to fingertip contact or proximity.

FIG. 6 is a flowchart for explaining the finger determination process.
The calculation unit 24 of the processing unit 20 sequentially selects the regions (FIG. 5B) specified by the region specifying unit 24 (ST200), and based on the detection data included in the selected region and its detection position (FIG. 4A). Then, each parameter (σ _x ² , σ _y ² , σ _xy ) defining the probability density function f (x, y) of the two-dimensional normal distribution is calculated (ST205). Furthermore, the calculation unit 24 calculates the first evaluation value E1 represented by the equation (29) based on these parameters (ST210). The determination unit 25 of the processing unit 20 checks whether or not the first evaluation value E1 calculated by the calculation unit 24 is within a predetermined range (ST220). When the first evaluation value E1 is within the predetermined range, the determination unit 25 determines that the region specified by the region specifying unit 24 is due to contact or proximity of the fingertip (ST230), and the first evaluation value If E1 is not within the predetermined range, it is determined that the area is not due to fingertip contact or proximity (ST235). The processing unit 20 performs the processes after step ST205 for each area specified by the area specifying unit 24, and determines whether each area is a fingertip.

As described above, according to the input device according to the present embodiment, the two-dimensional distribution of detection data in a region where an object is in contact with or close to the operation surface of the sensor unit 10 is assumed to be a two-dimensional normal distribution. The contour of the region where the object is in contact with or close to the operation surface is a virtual set of points on a two-dimensional plane that satisfies the condition that the probability density function f (x, y) representing the two-dimensional normal distribution is equal to a constant value. Approximate ellipse (formula (16)). When the region on the operation surface that is in contact with or close to the object is specified, based on the detection data included in the specified region and its detection position (Equations (5) to (7)), this virtual A first evaluation value E1 (Expression (37)) corresponding to the ratio between the major axis and the minor axis of the ellipse is calculated. Depending on whether or not the first evaluation value E1 is within a predetermined range, it is determined whether or not the object in contact with or close to this area is the fingertip.
That is, the parameters of the probability density function f (x, y) by the relatively simple calculation (formulas (5) to (7)) using the detection data included in the contact / proximity region of the object and the information of the detection position ( σ _x ² , σ _y ² , σ _xy ) are calculated, and the first evaluation value E1 representing the feature of the ellipse shape can be easily calculated based on these parameters. Therefore, the calculation process is simplified and the amount of calculation can be greatly reduced as compared with the conventional method for deriving an elliptic equation that approximates the outline of the region.

  Further, the first evaluation value E1 calculated based on the detection data included in the region in contact with or close to the object and the detection position is an elliptic equation derived based only on information near the outline of the region. Compared to, the feature of the ellipse shape can be expressed with high accuracy. Therefore, the fingertip determination accuracy can be improved as compared with the conventional method of deriving an elliptic equation approximate to the contour line.

  In addition, this invention is not limited only to embodiment mentioned above, Various modifications are included.

  In the embodiment described above, in determining whether an object that is in contact with or close to the region on the operation surface of the sensor unit 10 is a fingertip, the evaluation value according to the ratio between the major axis and the minor axis of an ellipse that approximates the region However, the present invention is not limited to this. In another embodiment of the present invention, in addition to the evaluation value according to the ratio between the major axis and the minor axis of the ellipse, for example, an evaluation value according to the area of the ellipse may be used for fingertip determination.

FIG. 7 is a flowchart for explaining an example of finger determination processing according to another embodiment of the present invention. The flowchart shown in FIG. 7 is obtained by adding steps ST215 and ST225 to the flowchart shown in FIG. 6, and the other steps are the same as those shown in FIG. In the finger determination process shown in FIG. 7, the calculation unit 24 calculates a second evaluation value E2 corresponding to the area of the ellipse in addition to the already described first evaluation value E1 (ST215). The second evaluation value E2 is expressed by the following equation using the variance σ _x ² for the variable _x and the variance σ _y ² for the variable y.

When the first evaluation value E1 is within a predetermined range (within the first range) and the second evaluation value E2 is within the predetermined range (within the second range), the determination unit 25 determines whether the sensor unit 10 It is determined that the object in contact with or close to the operation surface is the fingertip, and the first evaluation value E is not within the predetermined range (first range) or the second evaluation value E2 is within the predetermined range (second If it is not within the range, it is determined that the object is not a fingertip (ST220 to ST235).
For example, even if the ratio of the major axis to the minor axis of the ellipse is a value that is acceptable as the shape of the fingertip, if the area of the region is extremely large or extremely small to an unacceptable level for the fingertip, It is determined that the object in contact with or close to the operation surface is not a fingertip.
Thus, by adding the size (area of the contact / proximity region) in addition to the shape of the ellipse, the accuracy of fingertip determination can be further increased. In addition, since the same parameters (σ _x ² , σ _y ² ) as those used for calculating the first evaluation value E1 can be used for calculating the second evaluation value E2, calculation by calculating the second evaluation value E2 is also possible. The increase in the amount can be suppressed to a minute.

DESCRIPTION OF SYMBOLS 10 ... Sensor part, 11 ... Sensor matrix, 12 ... Capacitive sensor element, 13 ... Detection data generation part, 14 ... Drive part, 20 ... Processing part, 21 ... Timing control part, 22 ... Two-dimensional data generation part, 23 ... Area specifying unit, 24 ... calculating unit, 25 ... determining unit, 26 ... coordinate calculating unit, 30 ... storage unit, 31, 32 ... two-dimensional data memory, 33 ... region information memory, 34 ... object coordinate memory, 40 ... interface unit , E1 ... first evaluation value, E2 ... second evaluation value.

Claims (8)

An input device for inputting information according to contact or proximity of an object to an operation surface,
A sensor unit for detecting the degree of contact or proximity of an object at a plurality of detection positions distributed on the operation surface;
A two-dimensional data generation unit that generates two-dimensional data including a plurality of detection data indicating the degree of contact or proximity of an object at a plurality of positions on the operation surface based on the detection result of the sensor unit;
Based on the two-dimensional data, an area specifying unit that specifies an area on the operation surface that is in contact with or close to an object;
A calculation unit that calculates a first evaluation value according to a ratio between a major axis and a minor axis of a virtual ellipse corresponding to the region identified by the region identifying unit;
When the first evaluation value is within the first range, it is determined that the object that is in contact with or close to the operation surface is a fingertip, and when the first evaluation value is not within the first range, the object And a determination unit that determines that is not a fingertip,
When the virtual ellipse is assumed that the two-dimensional distribution of the detection data in the region specified by the region specifying unit is a two-dimensional normal distribution, the probability density function representing the two-dimensional normal distribution is a constant value. A set of points on a two-dimensional plane that satisfy the same condition,
The input unit, wherein the calculation unit calculates the first evaluation value based on the detection data included in the specified region and its detection position. The calculation unit calculates a predetermined parameter in the probability density function based on the detection data included in the specified region, and calculates the first evaluation value based on the calculated parameter. The input device according to claim 1. The detection position of the detection data is specified by the coordinate value of the first coordinate axis and the coordinate value of the second coordinate axis in the orthogonal coordinate system set on the operation surface,
The calculation unit uses the coordinate value of the first coordinate axis as a first random variable of the probability density function, the coordinate value of the second coordinate axis as a second random variable of the probability density function, and the probability density function detects the detection In the case of having a value according to the data, the first parameter according to the variance for the first random variable, the second parameter according to the variance for the second random variable, and the first random variable and the first The third parameter according to the covariance for two random variables is calculated, and the first evaluation value is calculated based on the calculated first parameter, second parameter, and third parameter. The input device described in 1. The calculation unit adds a value obtained by squaring a difference between the first parameter and the second parameter and a value that is four times the third parameter to obtain a sum of the first parameter and the second parameter. The input device according to claim 3, wherein the first evaluation value corresponding to a numerical value obtained by dividing by a squared value is acquired. The calculation unit calculates a second evaluation value corresponding to the area of the virtual ellipse based on the first parameter and the second parameter;
The determination unit determines that an object that is in contact with or close to the operation surface is a fingertip when the first evaluation value is within the first range and the second evaluation value is within the second range. When the first evaluation value is not within the first range or the second evaluation value is not within the second range, it is determined that the object is not a fingertip. The input device described in 1. The input device according to claim 5, wherein the calculation unit calculates the second evaluation value according to a sum of the first parameter and the second parameter. A computer that has input detection results of sensors that detect the degree of contact or proximity of an object at a plurality of detection positions distributed on the operation surface determines whether the object is a fingertip based on the detection result. A finger determination method for
Generating two-dimensional data including a plurality of detection data indicating the degree of contact or proximity of an object at a plurality of positions on the operation surface based on a detection result of the sensor;
Identifying a region on the operation surface in contact with or close to an object based on the two-dimensional data;
Calculating a first evaluation value according to a ratio between a major axis and a minor axis of a virtual ellipse corresponding to the identified region;
When the first evaluation value is within the first range, it is determined that the object that is in contact with or close to the operation surface is a fingertip, and when the first evaluation value is not within the first range, the object Determining that is not a fingertip,
The virtual ellipse has a constant probability density function representing the two-dimensional normal distribution when the two-dimensional distribution of the detection data in the region specified in the step of specifying the region is assumed to be a two-dimensional normal distribution. A set of points on a two-dimensional plane that satisfy a condition equal to a value,
In the step of calculating the first evaluation value, the first evaluation value is calculated based on the detection data included in the specified region and its detection position. A program for causing a computer to execute each step in the finger determination method according to claim 7.