Intensity Estimation Using Binary Sensor Array With Spatially Varying Thresholds

• US 20110149274A1
• Filed: 12/23/2009
• Published: 06/23/2011
• Est. Priority Date: 12/23/2009
• Status: Active Grant

• Current Assignee: Nokia Technologies Oy (Nokia Corporation)
• Original Assignee: Nokia Corporation
• Inventors: Maki-Marttunen, Tuomo, Rissa, Tero P., Koskinen, Samu T., Viikinkoski, Matti
• Granted Patent: US 8,420,996 B2
• Time in Patent Office: Days
• US Class Current: 356/222
• CPC Class Codes: G01J 1/18 using comparison with a reference electric value
• CPC Class Codes: G01J 1/46 using a capacitor
• CPC Class Codes: H04N 25/77 Pixel circuitry, e.g. memories, A/D converters, pixel amplifiers, shared circuits or shared components

Abstract

An apparatus includes an array containing N sub-diffraction limit light sensors each having an associated light absorption activation threshold for switching from a reset state to an activated state, where the light absorption activation values lie within a range of values. The apparatus further includes a processor connected with a memory including computer program code, where the memory and computer program code are configured to, with the processor, cause the apparatus at least to perform estimating an intensity of light that illuminates the array based on electrical outputs of the array.

Claims

1. An apparatus, comprising:

an array containing N sub-diffraction limit light sensors each having an associated light absorption activation threshold for switching from a reset state to an activated state, where the light absorption activation values lie within a range of values; and

a processor connected with a memory including computer program code, where the memory and computer program code are configured to, with the processor, cause the apparatus at least to perform estimating an intensity of light that illuminates the array based on electrical outputs of the array.

2. The apparatus as in claim 1, where estimating uses training data to determine the light absorption activation threshold of each sensor and, based on the electrical outputs of the array, to estimate the intensity using the determined light absorption activation thresholds of the sensors and their locations.

3. The apparatus of claim 2, where estimating uses a Bayesian method.

4. The apparatus of claim 2, where estimating uses a feedforward neural network, where the training data is used to train the neural network to estimate intensity from the electrical outputs of the array.

5. The apparatus of claim 1, where estimating uses the electrical output of the array to determine a number of activated sensors and a probability mass function to determine a distribution of the light absorption activation thresholds of the sensors.

6. The apparatus of claim 5, where estimating the intensity uses a maximum likelihood estimate.

7. The apparatus of claim 1, where N is in a range of about 10⁶ to about 10⁹ or greater.

8. The apparatus of claim 1, where the light absorption activation values lie within a range of values of about 1 to about 4.

9. The apparatus of claim 1, embodied as a camera in a user device.

18. The method of claim 1, executed in a camera module embodied in a user device.

10. A method, comprising:

providing an array containing N sub-diffraction limit light sensors each having an associated light absorption activation threshold for switching from a reset state to an activated state, where the light absorption activation values lie within a range of values; and

estimating an intensity of light that illuminates the array based on electrical outputs of the array.

11. The method as in claim 10, where estimating uses training data to determine the light absorption activation threshold of each sensor and, based on the electrical outputs of the array, to estimate the intensity using the determined light absorption activation thresholds of the sensors and their locations.

12. The method of claim 11, where estimating uses a Bayesian method.

13. The method of claim 11, where estimating uses a feedforward neural network, where the training data is used to train the neural network to estimate intensity from the electrical outputs of the array.

14. The method of claim 10, where estimating uses the electrical output of the array to determine a number of activated sensors and a probability mass function to determine a distribution of the light absorption activation thresholds of the sensors.

15. The method of claim 14, where estimating the intensity uses a maximum likelihood estimate.

16. The method of claim 10, where N is in a range of about 10⁶ to about 10⁹ or greater.

17. The method of claim 10, where the light absorption activation values lie within a range of values of about 1 to about 4.