RESOLVING MERGED TOUCH CONTACTS

• US 20120113017A1
• Filed: 11/08/2010
• Published: 05/10/2012
• Est. Priority Date: 11/08/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Benko, Hrvoje, Wilson, Andrew
• Granted Patent: US 9,122,341 B2
• Time in Patent Office: Days
• US Class Current: 345/173
• CPC Class Codes: G06F 3/04186 Touch location disambiguation
• CPC Class Codes: G06F 3/044 by capacitive means

Abstract

A method for resolving merged contacts detected by a multi-touch sensor includes resolving a first touch contact to a first centroid(N) for a frame(N) and resolving a second touch contact, distinct from the first touch contact, to a second centroid(N) for the frame(N). Responsive to the first touch contact and the second touch contact merging into a merged touch contact in a frame(N+1), the merged touch contact is resolved to a first centroid(N+1) and a second centroid(N+1).

Claims

1. A method of tracking touch contacts, the method comprising:

resolving a first touch contact to a first centroid(N) for a frame(N);

resolving a second touch contact, distinct from the first touch contact, to a second centroid(N) for the frame(N);

responsive to the first touch contact and the second touch contact merging into a merged touch contact in a frame(N+1), resolving the merged touch contact to a first centroid(N+1) and a second centroid(N+1).

2. The method of claim 1, wherein detecting the merged contact includes identifying a continuous contact in frame(N+1) having a combined statistic approximately equal to a sum of an individual statistic of the first touch contact and an individual statistic of the second touch contact.

3. The method of claim 1, wherein resolving the merged touch contact includes assigning a first weight(N+1) and a second weight(N+1) to each particular pixel of the merged contact in frame(N+1).

4. The method of claim 3, further comprising:

calculating the first weight(N+1) of each pixel by multiplying an intensity of the merged touch contact observed at that pixel in frame(N+1) by a first probabilistic likelihood that particular pixel belongs to the first touch contact; and

calculating the second weight(N+1) of each pixel by multiplying an intensity of the merged touch contact observed at that pixel in frame(N+1) by a second probabilistic likelihood that particular pixel belongs to the second touch contact.

5. The method of claim 4, wherein the intensity of the merged touch contact observed at each pixel is acquired from raw touch sensor measurements.

6. The method of claim 4, wherein the first probabilistic likelihood and the second probabilistic likelihood include a Gaussian distribution.

7. The method of claim 4, further comprising:

normalizing the first weight(N+1) and the second weight(N+1) of each pixel such that a sum of the first weight(N+1) and the second weight(N+1) for that pixel equals 1;

normalizing each pixel for the probabilistic likelihood of the first touch contact such that a sum of first weights(N+1) of all pixels of the merged touch contact equals 1; and

normalizing each pixel for the probabilistic likelihood of the second touch contact such that a sum of second weights(N+1) of all pixels of the merged touch contact equals 1.

8. The method of claim 4, further comprising resolving the merged contact to a first centroid(N+1) of normalized first weights(N+1) and a second centroid(N+1) of normalized second weights(N+1).

9. The method of claim 8, further comprising associating a first mean and covariance matrix with the first centroid(N+1) and a second mean and covariance matrix with the second centroid(N+1).

10. The method of claim 8, further comprising calculating a first new probabilistic likelihood from the first centroid(N+1) and a second new probabilistic likelihood from the second centroid(N+1) for weighting pixels in a subsequent frame.

11. The method of claim 1, further comprising detecting the first touch contact, the second touch contact, and the merged touch contact with a multi-touch sensor.

12. A method of tracking touch contacts, the method comprising:

resolving a first touch contact to a first centroid(N) for a frame(N);

resolving a second touch contact, distinct from the first touch contact, to a second centroid(N) for the frame(N);

responsive to the first touch contact and the second touch contact merging into a merged touch contact in a frame(N+1), for each particular pixel of the merged touch contact;

assigning a first weight(N+1) to that particular pixel with respect to the first touch contact, the first weight(N+1) being a product of an intensity of the merged touch contact observed at that particular pixel in frame(N+1) and a first probabilistic likelihood that particular pixel belongs to the first touch contact based on the first centroid(N) for frame(N);

assigning a second weight(N+1) to that particular pixel with respect to the second touch contact, the second weight(N+1) being a product of an intensity of the merged touch contact observed at that particular pixel in frame(N+1) and a second probabilistic likelihood that particular pixel belongs to the second touch contact based on the second centroid(N) for frame(N);

normalizing the first weight(N+1) and the second weight(N+1) such that a sum of first weight(N+1) and second weight(N+1) equals 1;

normalizing each pixel for the probabilistic likelihood of the first touch contact such that a sum of first weights(N+1) of all pixels of the merged touch contact equals 1;

normalizing each pixel for the probabilistic likelihood of the second touch contact such that a sum of second weights(N+1) of all pixels of the merged touch contact equals 1;

resolving the merged touch contact to a first centroid(N+1) of normalized first weights(N+1) and a second centroid(N+1) of normalized second weights(N+1).

13. The method of claim 12, wherein detecting the merged contact includes identifying a continuous contact in frame(N+1) having a combined statistic approximately equal to a sum of an individual statistic of the first touch contact and an individual statistic of the second touch contact.

14. The method of claim 12, wherein the intensity of the merged touch contact observed at each pixel is acquired from raw touch sensor measurements.

15. The method of claim 12, wherein the first probabilistic likelihood and the second probabilistic likelihood include Gaussian distributions.

16. The method of claim 12, further comprising associating a first mean and covariance matrix with the first centroid(N+1) and a second mean and covariance matrix with the second centroid(N+1).

17. The method of claim 12, further comprising calculating a new first probabilistic likelihood from the first centroid(N+1) and a new second probabilistic likelihood from the second centroid(N+1) for weighting pixels in a subsequent frame.

18. The method of claim 17, wherein the first new probabilistic likelihood and second new probabilistic likelihood include a Gaussian distribution.

19. The method of claim 12, further comprising detecting the first touch contact, the second touch contact, and the merged touch contact with a multi-touch sensor.

20. A data-holding subsystem holding instructions executable by a logic subsystem to:

resolve a first touch contact to a first centroid(N) for a frame(N);

responsive to the first touch contact and a second touch contact merging into a merged touch contact in a frame(N+1), resolve the merged touch contact to a first centroid(N+1) and a second centroid(N+1).