Granted Patent B2

US 12,272,118 · App. 17/880,721 · Granted Apr 8, 2025

Inventor: Timothy Peter Wiles (North Yorkshire, GB)

Assignee: Peratech IP Ltd

G06V10/764G06V10/751G06V10/7715G06V10/82

Abstract

Claims (27)

1. A method of classifying pressure inputs in a sensing array, said sensing array comprising a plurality of sensing elements responsive to pressure inputs, said method comprising the steps of:

identifying a plurality of pressure inputs in said sensing array;

converting said plurality of pressure inputs into an output image;

comparing said output image with a data set comprising a plurality of images comprising undesirable pressure inputs by means of an artificial neural network, each said image including an associated mask; and

applying one said associated mask consistent with said output image in response to said comparing step to remove undesirable pressure inputs; wherein

said step of converting said plurality of pressure inputs into an output image comprises mapping each said pressure input to a pixel of a pixel array, said pixel array corresponding to said sensing array.

2. The method of claim 1 , further comprising the step of: producing said images in a greyscale format.

3. The method of claim 2 , wherein pressure inputs are defined as high or low and further wherein high-pressure inputs are designated as white and low-pressure inputs are designated as black.

4. The method of claim 1 , further comprising the step of: creating said data set comprising a plurality of images comprising undesirable pressure inputs.

5. The method of claim 4 , wherein said step of creating said data set comprises augmenting said data set by adjusting each said image.

6. The method of claim 5 , wherein said step of adjusting each said image comprises at least one of the following: rotating said image; flipping said image; scaling said image.

7. The method of claim 4 , further comprising the step of: training said artificial neural network through said data set to identify undesirable pressure inputs.

8. The method of claim 1 , further comprising the step of: extending said associated mask by tailoring a kernel size of said artificial neural network.

9. The method of claim 1 , further comprising the step of: creating said associated mask by identifying undesirable pressure inputs from said plurality of images in said data set.

10. The method of claim 1 , wherein said step of identifying a plurality of pressure inputs is conducted on a frame-by-frame basis.

11. The method of claim 1 , wherein said artificial neural network is a convolutional neural network.

12. An apparatus for classifying pressure inputs, comprising:

a sensing array comprising a plurality of sensing elements responsive to pressure inputs, a pixel array corresponding to said sensing array, and a processor, said processor being configured to:

identify a plurality of pressure inputs in said sensing array;

convert said plurality of pressure inputs into an output image by mapping each said pressure input to a pixel of said pixel array;

compare said output image with a data set comprising a plurality of images comprising undesirable pressure inputs by means of an artificial neural network, each said image including an associated mask; and

apply one said associated mask consistent with said output image in response to comparing said output image to remove undesirable pressure inputs.

13. The apparatus of claim 12 , wherein said sensing array is configured to provide two-dimensional positional data and an extent property in response to an applied pressure.

14. The apparatus of claim 12 , wherein said associated mask has a bit depth of one bit.

15. The apparatus of claim 12 , wherein said images are produced in a greyscale format.

16. A touch screen comprising the apparatus of claim 12 .

17. An electronic device comprising the touch screen of claim 16 .

Assignments (4)

Priority Claims (1)

Continuity (2)

References Cited (17)