Pen state detection circuit, system, and method
Provided are a pen state detection circuit, a pen state detection system, and a pen state detection method that can improve estimation accuracy for a pen state in an electronic pen including at least one electrode. A pen state detection circuit acquires, from a touch sensor, a first signal distribution indicating a change in capacitance associated with approach of a first electrode and uses a machine learning estimator to estimate an instruction position or an inclination angle of an electronic pen from first feature values related to the first signal distribution. The first feature values include first local feature values related to a first local distribution corresponding to sensor electrodes in a number fewer than the number of arranged sensor electrodes exhibiting the first signal distribution.
1. A pen state detection circuit that detects a state of an electronic pen including a first electrode, on a basis of a signal distribution detected by a capacitance touch sensor including a plurality of sensor electrodes arranged in a plane shape,
the pen state detection circuit executing:
an acquisition step of acquiring, from the touch sensor, a first signal distribution indicating a change in capacitance associated with approach of the first electrode, and
an estimation step of using a machine learning estimator to estimate an instruction position or an inclination angle of the electronic pen from first feature values related to the first signal distribution,
wherein
the first feature values include first local feature values related to a first local distribution corresponding to sensor electrodes in a number fewer than the number of arranged sensor electrodes exhibiting the first signal distribution and a reference position of the first local distribution in a sensor coordinate system defined on a detection surface of the touch sensor.
2. The pen state detection circuit according to claim 1 , wherein the first local feature values include a certain number of pieces of data regardless of the number of arranged sensor electrodes.
3. The pen state detection circuit according to claim 1 , wherein
the machine learning estimator is configured to be capable of executing position computation, with a relative position between the reference position and the instruction position as an output value, and
in the estimation step, the relative position is added to the reference position to estimate the instruction position.
4. The pen state detection circuit according to claim 1 , wherein
the electronic pen further includes a second electrode different from the first electrode,
in the acquisition step, a second signal distribution indicating a change in capacitance associated with approach of the second electrode is further acquired from the touch sensor,
in the estimation step, the instruction position or the inclination angle is estimated from the first feature values and second feature values related to the second signal distribution, and
the second feature values include second local feature values related to a second local distribution corresponding to sensor electrodes in a number fewer than the number of arranged sensor electrodes exhibiting the second signal distribution.
5. The pen state detection circuit according to claim 4 , wherein
the first electrode is a tip electrode that has a shape symmetrical with respect to an axis of the electronic pen and that is provided at a tip of the electronic pen, and
the second electrode is an upper electrode that has a shape symmetrical with respect to the axis of the electronic pen and that is provided on a base end side of the tip electrode.
6. The pen state detection circuit according to claim 5 , wherein the machine learning estimator is configured to be capable of sequentially executing:
angle computation with the second local feature values as input values and with the inclination angle as an output value, and
position computation with the first local feature values and the inclination angle as input values and with the relative position as an output value.
7. The pen state detection circuit according to claim 6 , wherein the machine learning estimator includes:
a first switch that is capable of switching and outputting one of a learning parameter group for the angle computation and a learning parameter group for the position computation,
a second switch that is capable of switching and outputting one of the input values for the angle computation and the input values for the position computation, and
a common computation element that is capable of selectively executing the angle computation or the position computation according to the switch of the first switch and the second switch.
8. The pen state detection circuit according to claim 4 , wherein
the machine learning estimator is configured to be capable of executing position computation with the first local feature values and the second local feature values as input values and with the relative position as an output value.
9. The pen state detection circuit according to claim 4 , wherein the machine learning estimator includes:
a combiner that combines the first feature values and the second feature values to output third feature values, and
a computation element that sets the third feature values as input values and sets the instruction position as an output value.
10. The pen state detection circuit according to claim 4 , wherein
the first local feature values include feature values indicating tilts of the first local distribution or absolute values of the tilts, and
the second local feature values include feature values indicating tilts of the second local distribution or absolute values of the tilts.
11. The pen state detection circuit according to claim 1 , wherein the machine learning is learning with training using training data obtained by actual measurement or calculation simulation.
12. The pen state detection circuit according to claim 1 , wherein
the pen state detection circuit further executes a processing step of sequentially applying a dimension compression process and a dimension restoration process to the first feature values to execute an autoencoding process of obtaining first feature values equivalent to the number of dimensions of input, and
in the estimation step, the machine learning estimator is used to estimate the instruction position or the inclination angle from the first feature values to which the autoencoding process is applied.
13. The pen state detection circuit according to claim 1 , wherein
in the estimation step, the instruction position or the inclination angle is estimated from the first feature values by following different computation rules according to a projection position of the first electrode on a detection surface of the touch sensor.
14. The pen state detection circuit according to claim 13 , wherein
the computation rules are rules for estimating the instruction position or the inclination angle, and
in the estimation step, the instruction position or the inclination angle is estimated by using the machine learning estimator in which different learning parameter groups are set according to whether or not the projection position of the first electrode interferes with a periphery of the touch sensor.
15. The pen state detection circuit according to claim 13 , wherein
the computation rules are rules for calculating the first local feature values, and
in the estimation step, the instruction position or the inclination angle is estimated from the first local feature values calculated by following different computation rules according to whether or not the projection position of the first electrode interferes with a periphery of the touch sensor.
16. The pen state detection circuit according to claim 1 , which is included in a pen state detection system including:
an electronic device including the pen state detection circuit;
the electronic pen used along with the electronic device; and
a server apparatus that is configured to be capable of performing two-way communication with the electronic device and storing learning parameter groups of the machine learning estimator constructed on the pen state detection circuit, wherein
the electronic device is triggered by detection of the electronic pen, to request the server apparatus to transmit a learning parameter group corresponding to the electronic pen.
17. A pen state detection method of detecting a state of an electronic pen including an electrode, on a basis of a signal distribution detected by a capacitance touch sensor including a plurality of sensor electrodes arranged in a plane shape, the pen state detection method comprising:
an acquisition step of acquiring, from the touch sensor, a signal distribution indicating a change in capacitance associated with approach of the electrode, and
an estimation step of using a machine learning estimator to estimate an instruction position of the electronic pen from feature values related to the signal distribution,
wherein
the feature values include local feature values related to a local distribution corresponding to sensor electrodes in a number fewer than the number of arranged sensor electrodes exhibiting the signal distribution and a reference position of the first local distribution in a sensor coordinate system defined on a detection surface of the touch sensor.
18. The pen state detection method according to claim 17 , wherein
the estimation step includes following different computation rules according to a projection position of the electrode on a detection surface of the touch sensor.