Decorrelated channel sampling for digital imaging
View Patent ↗In association with an imaging device which generates a portion of an image from a plurality of channels within a first row by sampling each channel during a sampling time corresponding to the channel, circuitry offsets sampling times of at least first and second channels within the first row, thereby reducing noise correlation between the first and second channels in the first row. Pixel sampling times may be defined by start times of the channels within a row, end times or both. Offsetting may be accomplished using a predetermined set of sampling time values or by randomizing sampling time values.
1. Apparatus comprising:
in association with a device which generates a portion of an image from a plurality of channels within a first row by sampling each channel during a sampling time corresponding to the channel, circuitry which offsets sampling times of at least first and second channels within the first row, wherein sampling times are selected using at least one of randomization and pseudo-randomization, thereby reducing noise correlation between the first and second channels in the first row, wherein the circuitry comprises a charge amplifier having a first switched RC filter and a second switched RC filter for correlated double sampling.
2. The apparatus of claim 1 wherein at least one of the sampling times is different from all other channels.
3. The apparatus of claim 1 wherein the sampling time of each channel within the first row is different from all other channels within the first row.
4. The apparatus of claim 1 wherein there are a plurality of rows, and wherein the sampling time of the first channel within the first row is different from the sampling time of the first channel within a second row.
5. The apparatus of claim 1 wherein there are a plurality of rows, and wherein the sampling time of each channel within the first row is different from all channels within a second row.
6. The apparatus of claim 1 wherein the sampling time is defined by a start time.
7. The apparatus of claim 1 wherein the sampling time is defined by an end time.
8. The apparatus of claim 1 wherein the sampling time is a total pixel data tracking time.
9. The apparatus of claim 8 wherein total pixel data tracking time varies between channels in the first row.
10. The apparatus of claim 8 wherein total pixel data tracking time varies within the first channel between the first row and the second row.
11. The apparatus of claim 1 wherein the circuitry defines a pixel sample start window having a plurality of start time slots randomly distributed among channels within the first row.
12. The apparatus of claim 11 wherein the circuitry defines a pixel sample end window having a plurality of end time slots.
13. The apparatus of claim 12 wherein the start and end time slots drive switches of the first and second switched RC filters to respectively control a start of a pixel sample interval and an end of a pixel sample interval.
14. The apparatus of claim 13 wherein each switch of the first and second switched RC filters has an input with a dedicated, programmable delay line.
15. The apparatus of claim 14 wherein the programmable delay lines are programmed by random selection between a pre-identified number of time slots of a pixel sample window.
16. The apparatus of claim 15 wherein the programmable delay lines are re-programmed between charge amplifier read-outs such that sampling time varies between adjacent rows in a common channel.
17. The apparatus of claim 11 characterized by a window size of 20 microseconds apportioned into 10-50 slots.
18. A method comprising:
generating a portion of an image from a plurality of channels within a first row by sampling each channel during a sampling time corresponding to the channel, including offsetting sampling times of at least first and second channels within the first row, including selecting sampling times using at least one of randomization and pseudo-randomization, thereby reducing noise correlation between the first and second channels in the first row, and including using a charge amplifier having a first switched RC filter and a second switched RC filter for correlated doubling sampling.
19. The method of claim 18 including selecting at least one of the sampling times to be different from all other channels.
20. The method of claim 18 including setting the sampling time of each channel within the first row to a value different from that of all other channels within the first TOW.
21. The method of claim 18 , wherein there are a plurality of rows, and including setting the sampling time of the first channel within the first row to a value different from that of the sampling time of the first channel within a second row.
22. The method of claim 18 , wherein there are a plurality of rows, and including setting the sampling time of each channel within the first row to a value different from those of all channels within a second row.
23. The method of claim 18 including setting a sampling start time corresponding to the sampling time.
24. The method of claim 18 including setting a sampling end time corresponding to the sampling time.
25. The method of claim 18 including setting a total pixel data tracking time corresponding to the sampling time.
26. The method of claim 25 including setting different total pixel data tracking times for channels in the first row.
27. The method of claim 25 including setting different total pixel data tracking times within the first channel between the first row and the second row.
28. The method of claim 18 including defining a pixel sample start window having a plurality of start time slots randomly distributed among channels within the first row.
29. The method of claim 28 including defining a pixel sample end window having a plurality of end time slots.
30. The method of claim 29 , further comprising using the start and end time slots to drive switches of the first and second switched RC filters which respectively control a start of a pixel sample interval and an end of a pixel sample interval.
31. The method of claim 30 including controlling the switches of the first and second switched RC filters with an input with a dedicated, programmable delay line.
32. The method of claim 31 including programming the programmable delay lines by random selection between a pre-identified number of time slots of a pixel sample window.
33. The method of claim 32 including reprogramming the programmable delay lines between charge amplifier read-outs such that sampling time varies between adjacent rows in a common channel.
34. The method of claim 28 including setting a window size of 20 microseconds apportioned into 10-50 slots.