Method and apparatus for determining shape and orientation of a touch object on handheld devices

• US 9,684,407 B2
• Filed: 04/22/2015
• Issued: 06/20/2017
• Est. Priority Date: 04/22/2015
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Vanga, Sandeep
• Primary Examiner(s): Park, Sanghyuk
• Application Number:
  US14/693,859
• Publication Number:
  US 20160313843A1
• Time in Patent Office: 790 Days
• CPC Class Codes: G06F 2203/04103 Manufacturing, i.e. details related to manufacturing processes specially suited for touch sensitive devices
• CPC Class Codes: G06F 3/0412 Digitisers structurally integrated in a display
• CPC Class Codes: G06F 3/04166 Details of scanning methods, e.g. sampling time, grouping of sub areas or time sharing with display driving Synchronisation with the driving of the display or the backlighting unit to avoid interferences generated internally G06F3/04184
• CPC Class Codes: G06F 3/0446 using a grid-like structure of electrodes in at least two directions, e.g. using row and column electrodes

Abstract

A method and apparatus are disclosed for determining the shape and orientation of an object touching a touch panel on, for example, a handheld device. The method comprises extracting and storing in a training phase a plurality of touch attributes corresponding to a plurality of predefined objects touching the touch panel, receiving in a test phase an input data for a test object touching the touch panel, wherein the input data comprises a plurality of values corresponding to difference mutual capacitance at each node of the touch panel, determining a plurality of touch attributes associated with the test object by comparing the input data with the plurality of features extracted during the training phase and approximating input data of the test object with the plurality of touch attributes obtained in the training phase to determine the shape and orientation of the test object.

Claims

1. A method for determining shape and orientation of a touch on a touch interface, the method comprising:

receiving input data from a touch surface corresponding to the touch from the touch interface, the touch surface comprising an axis and a plurality of nodes, the input data for the touch comprising a plurality of values corresponding to a difference of mutual capacitance at each node, and the touch comprising a shape and an orientation with respect to the axis of the touch surface;

determining a normalized peak amplitude of the input data for the touch;

comparing the normalized peak amplitude of the input data for the touch with different angles of orientation of each of a plurality of two-dimensional (2D) Gaussian curves, each 2D Gaussian curve respectively corresponding to input data for a training-set input touch of a plurality of training-set input touches;

determining a minimum error between the input data for the touch and each of differently oriented plurality of 2D Gaussian curves; and

associating attributes of the training-set input touch corresponding to the minimum error with an input touch, the attributes of the training-set input touch corresponding to the minimum error comprising an input-touch difference amplitude, a length of a major axis of the input touch and a length of a minor axis of the input touch.

2. The method of claim 1, wherein comparing the normalized amplitude of the input data for the touch with different angles of orientation of each of a plurality of two-dimensional (2D) Gaussian curves comprises:

mapping a plurality of touch data for each of a plurality of training-set input touches onto a densely spaced 2D grid for a plurality of orientations with respect to the axis of the touch surface, the spaced 2D grid comprising a plurality of equally spaced touch points between electrodes of the touch surface; and

comparing the normalized amplitude of the input data for the touch with an amplitude each training-set input touch.

3. The method of claim 2, wherein mapping a plurality of touch data for each of a plurality of training-set input touches onto a densely spaced 2D grid for a plurality of orientations further comprises:

extracting the attributes for each training-set input touch.

4. The method of claim 1, further comprising:

obtaining training-set input touch data for each of a plurality of training-set input touches and orientations, the training-set input touch data comprising touch data and indirect touch data;

mapping difference mutual capacitance values for each of the plurality of training-set input touches onto a capacitance variation curve over a grid location corresponding to each training-set input touch;

identifying a peak value of the difference mutual capacitance values for each training-set input touch of the plurality of training-set input touches;

normalizing the capacitance variation curve for each training-set input touch of the plurality of training-set input touches based on the peak value for respective training-set input touch;

approximating the training-set input data for each training-set input touch of the plurality of training-set input touches by a two-dimensional Gaussian curve; and

determining a standard deviation of a major axis and a standard deviation of a minor axis for each training-set input touch of the plurality of training-set input touches.

5. The method of claim 4, further comprising:

initializing the capacitance variance curve for the touch surface based on a size of the touch surface.

6. The method of claim 5, further comprising:

storing the standard deviation of a major axis and the standard deviation of a minor axis for each training-set input touch of the plurality of training-set input touch.

7. The method of claim 5, wherein the touch surface is part of a touch-screen display.

8. The method of claim 7, wherein the touch-screen display is part of a smart phone or a tablet.

9. A system, comprising:

a touch-screen display comprising a touch surface having an axis and a plurality of nodes; and

at least one processor coupled to the touch-screen display, the at least one processor being configured to;

receive input data from the touch surface corresponding to a touch, the input data for the touch comprising a plurality of values corresponding to a difference of mutual capacitance at each node of the touch surface, and the touch comprising a shape and an orientation with respect to the axis of the touch surface;

determine a normalized peak amplitude of the input data for the touch;

compare the normalized peak amplitude of the input data for the touch with different angles of orientation of each of a plurality of two-dimensional (2D) Gaussian curves, each 2D Gaussian curve respectively corresponding to input data for a training-set input touch of a plurality of training-set input touches;

determine a minimum error between the input data for the touch and each of differently oriented plurality of 2D Gaussian curves;

associate attributes of the training-set input touch corresponding to the minimum error with an input touch, the attributes of the training-set input touch corresponding to the determined minimum error comprising an input-touch difference amplitude, a length of a major axis of the input touch and a length of a minor axis of the input touch; and

determine a touch vector for the input touch.

10. The system of claim 9, wherein the at least one processor is further configured to:

map a plurality of touch data for each of a plurality of training-set input touches onto a densely spaced 2D grid for a plurality of orientations with respect to the axis of the touch surface, the densely spaced 2D grid comprising a plurality of equally spaced touch points between electrodes of the touch surface; and

compare the normalized amplitude of the input data with an amplitude each training-set input touch.

11. The system of claim 10, wherein the at least one processor is further configured to:

extract the attributes for each training-set input touch.

12. The system of claim 9, wherein the at least one processor is further configured to:

obtain training-set input touch data for each of a plurality of training-set input touches and orientations, the training set input touch data comprising touch data and indirect touch data;

map difference mutual capacitance values for each of the plurality of training-set input touches onto a capacitance variation curve over a grid location corresponding to each training-set input touch;

identify a peak value of the difference mutual capacitance values for each training-set input touch of the plurality of training-set input touches;

normalize the capacitance variation curve for each training-set input touch of the plurality of training-set input touches based on the peak value for respective training-set input touch;

approximate the training-set input data for each training-set input touch of the plurality of training-set input touches by a two-dimensional Gaussian curve; and

determine a standard deviation of a major axis and a standard deviation of a minor axis for each training-set input touch of the plurality of training-set input touches.

13. The system of claim 12, wherein the at least one processor is further configured to:

initialize the capacitance variance curve for the touch surface based on a size of the touch surface.

14. The system of claim 12, wherein the at least one processor is further configured to:

store the standard deviation of a major axis and the standard deviation of a minor axis for each training-set input touch of the plurality of training-set input touch.

15. The system of claim 9, wherein the touch-screen display is part of a smart phone or a tablet.

16. A method for determining shape and orientation of a touch on a touch interface, the method comprising:

obtaining training-set input touch data for each of a plurality of training-set input touches and orientations, a touch surface comprising an axis and a plurality of nodes, and the training set input touch data comprising touch data and indirect touch data received from the touch interface;

mapping difference mutual capacitance values for each of the plurality of training-set input touches onto a capacitance variation curve over a grid location corresponding to each training-set input touch;

identifying a peak value of the difference mutual capacitance values for each training-set input touch of the plurality of training-set input touches;

normalizing the capacitance variation curve for each training-set input touch of the plurality of training-set input touches based on the peak value for respective training-set input touch;

approximating the training-set input data for each training-set input touch of the plurality of training-set input touches by a two-dimensional Gaussian curve;

determining a standard deviation of a major axis and a standard deviation of a minor axis for each training-set input touch of the plurality of training-set input touches;

storing the standard deviation of a major axis and the standard deviation of a minor axis for each training-set input touch of the plurality of training-set input touch.

17. The method of claim 16, further comprising:

receiving input data from the touch surface corresponding to a first touch from the touch surface, the first touch comprising a shape and an orientation with respect to the axis of the touch surface;

determining a normalized peak amplitude of the input data for the first touch;

comparing the normalized peak amplitude of the input data for the first touch with different angles of orientation of each of a plurality of two-dimensional (2D) Gaussian curves, each 2D Gaussian curve respectively corresponding to input data for a training-set input touch of the plurality of training-set input touches;

determining a minimum error between the input data of the first touch and each of differently oriented plurality of 2D Gaussian curves;

associating attributes of the training-set input touch corresponding to the minimum error with the input touch, the attributes of the training-set input touch corresponding to the minimum error comprising an input-touch difference amplitude, a length of a major axis of the first touch and a length of a minor axis of the first touch; and

determining a touch vector for the first touch.

18. The method of claim 17, wherein comparing the normalized amplitude of the input data of the first touch with different angles of orientation of each of a plurality of two-dimensional (2D) Gaussian curves comprises:

mapping a plurality of touch data for each of the plurality of training-set input touches onto a densely spaced 2D grid for a plurality of orientations with respect to the axis of the touch surface, the densely spaced 2D grid comprising a plurality of equally spaced touch points between electrodes of the touch surface; and

comparing the normalized amplitude of the input data for the first touch with an amplitude each training-set input touch.

19. The method of claim 16, wherein the touch surface is part of a touch-screen display.

20. The method of claim 19, wherein the touch-screen display is part of a smart phone or a tablet.