Dynamic range compression method

• US 7,336,309 B2
• Filed: 07/03/2001
• Issued: 02/26/2008
• Est. Priority Date: 07/05/2000
• Status: Expired due to Fees

• Current Assignee: Vision-Sciences Incorporated (2941881 Canada, Inc.)
• Original Assignee: Vision-Sciences Incorporated (2941881 Canada, Inc.)
• Inventors: Stark, Moshe
• Primary Examiner(s): Ye; Lin
• Assistant Examiner(s): Giles; Nicholas G
• Application Number:
  US10/332,151
• Publication Number:
  US 20040036797A1
• Time in Patent Office: 2,429 Days
• Field of Search: 348/297, 348/294, 348/308, 348/310
• US Class Current: 348/297
• CPC Class Codes: H04N 23/741 by increasing the dynamic range of the image compared to the dynamic range of the electronic image sensors
• CPC Class Codes: H04N 25/533 by using differing integration times for different sensor regions
• CPC Class Codes: H04N 25/57 Control of the dynamic range
• CPC Class Codes: H04N 25/571 involving a non-linear response
• CPC Class Codes: H04N 25/74 Circuitry for scanning or addressing the pixel array

Abstract

A method for compressing the dynamic range of an image sensor (202) including a multiplicity of pixels. The method includes the steps of exposing each of the pixels to light and producing an associated photocurrent per pixel, representative of the light exposure. Then, on a per-pixel basis, controlling exposure time of each of the pixels on the basis of a monotonically rising convex function of the associated photocurrent of each of the pixel.

Claims

1. A method for compressing the dynamic range of an image sensor comprising a multiplicity of pixels, the method comprising the steps of:

exposing each of said pixels to light and producing an associated photocurrent per pixel, representative of said light exposure;

on a per-pixel basis, controlling exposure time of each of said pixels on the basis of a monotonically rising convex function of said associated photocurrent of each said pixel,

wherein;

controlling exposure time includes;

calculating a photocurrent based on an accumulated voltage; and

determining an exposure time, based on said monotonically rising convex function of said calculated photocurrent; and

said monotonically rising convex function fulfills the following conditions;

for I_ph1>

I_ph2, f(I_ph1)>

f(I_ph2) and f′

(I_ph1)<

f′

(I_ph2),

 

where

f represents the monotonically rising convex function;

f′

represents the first derivative of the function f; and

I_ph1 and I_ph2 are arguments to function f and its first derivative f′

.

2. The method according to claim 1, further comprising the step of storing said determined exposure time of each said pixel as a q-bit value in memory.

3. The method according to claim 2, wherein said step of controlling comprises the steps of:

accessing said stored exposure times from a memory; and

programming said pixels according to said accessed exposure times.

4. The method according to claim 1, further comprising calculating a first derivative of said monotonically rising convex function based upon a desired sensitivity of said exposure time of each said pixel.

5. The method according to claim 1, further comprising:

accumulating charge generally linearly representative of said photocurrent; and

calculating said photocurrent as a product of said accumulated charge.

6. The method according to claim 1, wherein said step of controlling is performed by circuitry outside said multiplicity of pixels.

7. An exposure controller for use in an image sensor comprising a multiplicity of pixels, the controller comprising:

a memory for storing calculated exposure time values;

a per-pixel parameter table containing local parameters of said pixels; and

a processor for combining based on a set of convergence criteria, said stored exposure time values and said parameters in order to determine, on a per-pixel basis, an exposure time of said pixels, wherein;

the processor is configured to;

calculate a photocurrent from an accumulated voltage; and

determine an exposure time, based on a monotonically rising convex function of said calculated photocurrent; and

wherein said monotonically rising convex function fulfills the following conditions;

for I_ph1>

I_ph2, f(I_ph1)>

f(I_ph2) and f′

(I_ph1)<

f′

(I_ph2),

f represents the monotonically rising convex function;

f′

represents the first derivative of the function f; and

I_ph1 and I_ph2 are arguments to function f and its first derivative f′

.

8. The exposure controller of claim 7, wherein said set of convergence criteria is based on a set of algorithms capable of deriving an appropriate exposure time for every pixel based on a monotonically rising convex function.

9. The exposure controller of claim 8, wherein said set of algorithms is capable of deriving a fine tuned exposure time of said pixels.

10. The exposure controller of claim 7, wherein said exposure controller is a chip-set external to said image sensor.

11. The exposure controller of claim 7, wherein said exposure controller is at least partially integrated on the same substrate as said image sensor.

12. The exposure controller of claim 7, wherein said memory and said processor are disposed outside said plurality of pixels.

13. A method for determining an exposure time of each of a multiplicity of pixels belonging to an image sensor, the method comprising:

initially setting an exposure time of each pixel to a minimum value t_E^MIN;

performing a first charge integration of each pixel over said exposure time to obtain a first charge voltage V^c, and;

if said first charge voltage V^c is less than a cutoff voltage V_CO, adjusting the exposure time of said each pixel to a maximum value t_E^MAX;

wherein t_E<sup>MAX>

t</sup>_E^MIN;

if said first charge voltage V^c is between said cutoff voltage V_CO and a saturation voltage V_Sat, adjusting the exposure time of said each pixel based on a monotonically rising convex function of a photocurrent of that pixel; and

if said first charge voltage V^c is greater than the saturation voltage V_Sat, leaving the exposure time of said each pixel at t_E^MIN; and

performing a second charge integration of each pixel whose exposure time was set to t_E^MAX to obtain a second charge voltage, and;

if said second charge voltage is greater than said cutoff voltage V_CO, adjusting the exposure time of said each pixel whose exposure time was set to t_E^MAX, based on a monotonically rising convex function of a photocurrent of that pixel;

wherein said monotonically rising convex function fulfills the following conditions;

for I_ph1>

I_ph2, f(I_ph1)>

f(I_ph2) and f′

(I_ph1)<

f′

(I_ph2), where;

f represents the monotonically rising convex function;

f represents the first derivative of the function f; and

I_ph1 and I_ph2 are arguments to function f and its first derivative f′

.

14. The method according to claim 13, further comprising:

iteratively adjusting the exposure time of each pixel until a predetermined convergence criteria is met.

15. A method for compressing the dynamic range of an image sensor comprising a multiplicity of pixels arranged in a pixel array, the method comprising the steps of:

for each pixel;

exposing the pixel to light and producing an associated photocurrent representative of said light exposure;

determining, with a processor, an exposure time of the pixel on the basis of a monotonically rising convex function of said associated photocurrent;

storing said determined exposure time as a q-bit value in memory located outside the pixel array;

accessing the stored exposure time from said memory; and

programming the pixel to accumulate charge for the accessed exposure time;

wherein said monotonically rising convex function fulfills the following conditions;

for I_ph1>

I_ph2, f(I_ph1)>

f(I_ph2) and f′

(I_ph1)<

f′

(I_ph2), where;

f represents the monotonically rising convex function;

f′

represents the first derivative of the function f; and

I_ph1 and I_ph2 are arguments to function f and its first derivative f′

.

16. A pixel electronic control shutter system comprising:

an image sensor comprising a pixel array having multiplicity of pixels, the image sensor producing at least one analog output voltage in response to light received by a pixel;

an analog-to-digital converter receiving the at least one analog output voltage and outputting a corresponding digitized voltage; and

an exposure controller receiving said digitized voltage from the analog-to-digital converter, the exposure controller comprising;

a memory for storing calculated exposure time values;

a per-pixel parameter table containing local parameters of said pixels; and

a processor for combining based on a set of convergence criteria, said stored exposure time values and said parameters in order to determine, on a per-pixel basis, an exposure time of said pixels, wherein;

the processor is configured to;

calculate a photocurrent from said digitized voltage; and

determine an exposure time, based on a monotonically rising function of said calculated photocurrent; and

wherein said monotonically rising convex function fulfills the following conditions;

for I_ph1>

I_ph2, f(I_ph1)>

f(I_ph2) and f′

(I_ph1)<

f′

(I_ph2), where;

f represents the monotonically rising convex function;

f′

represents the first derivative of the function f; and

I_ph1 and I_ph2 are arguments to function f and its first derivative f′

.

17. The pixel electronic control shutter system according to claim 16, wherein:

said memory and said processor are disposed outside said plurality of pixels; and

said determined exposure time is stored as a q-bit value in said memory.