Image sensing system with histogram modification

1. A method of processing an image, comprising:

obtaining values indicative of the image;

placing said values in bins representing values of a histogram, said bin each representing a value for gray scale of the image part;

setting a pixel threshold value, below which values of pixels will be combined;

setting a number of pixel value, indicating a maximum number of pixels which can be combined together; and

testing each pixel to determine if it is below said pixel threshold and below said pixel number threshold, and combining said pixel with another pixel only if so.

2. A method for modifying a histogram comprising:

determining gray scale values of each pixel of an image, said gray scale values contained in said histogram;

normalizing said gray scale values;

constructing a plurality of gray scale bins;

assigning each normalized gray scale value to one of said plurality of gray scale bins; and

modifying said histogram of normalized gray scale values according to a transformation function.

3. The method according to claim 2 , wherein said transformation function transforms each normalized gray scale value to a transformed gray scale value.

4. The method according to claim 3 , wherein said transformation function is a probability density function of the transformed gray scale values.

5. The method according to claim 2 , wherein said transformation function is single valued, said single values monotonically increase in the normalized interval, wherein said normalized interval is between 0 and 1.

6. The method according to claim 2 , further comprising compressing dip areas of said histogram.

7. The method according to claim 6 , wherein compressing said dip areas further comprises:

initializing a first threshold value;

initializing a second threshold value, wherein said second threshold value limits a total number of pixel gray scale values that can be combined;

setting a gray shift left value to zero;

setting a value of a common last gray scale value to a minimum gray scale value;

comparing each normalized gray scale value of each bin to said first and said second threshold;

adjusting said gray scale values based on a result of said comparison;

determining a gray scale maximum;

subtracting said gray scale minimum from said gray scale maximum;

adaptively adjusting said first threshold; and

limiting said adaptively adjusted first threshold to a number of pixels divided by 2 n+1 .

8. The method according to claim 7 , wherein said first threshold value is a number of pixels divided by a number of gray scale levels.

9. The method according to claim 7 , wherein said first threshold value is a number of pixels divided by 2 n+2 .

10. The method according to claim 7 , wherein said second threshold value is a number of pixels divided by a number of gray scale levels divided by 2.

11. The method according to claim 7 , wherein said first threshold value is a number of pixels divided by 2.

12. The method according to claim 7 , wherein said minimum gray scale value is set to zero.

13. A image processor comprising:

an input for receiving pixel image values; and

a processing section for:

determining gray scale values of each pixel of an image, said gray scale values contained in a histogram;

normalizing said gray scale values;

constructing a plurality of gray scale bins;

assigning each normalized gray scale value to one of said plurality of gray scale bins; and

modifying said histogram of normalized gray scale values according to a transformation function.

14. The image processor according to claim 13 , said processing section further performing the following operation;

compressing dip areas of said histogram.

15. The processor according to claim 14 , wherein said image processor performs the operation of compressing dip areas of said histogram by further performing the following operations:

initializing a first threshold value;

initializing a second threshold value, wherein said second threshold value limits a total number of pixel gray scale values that can be combined;

setting a gray shift left value to zero;

setting a value of a common last gray scale value to a minimum gray scale value;

comparing each normalized gray scale value of each bin to said first and said second threshold;

adjusting said gray scale values based on a result of said comparison;

determining a gray scale maximum;

subtracting said gray scale minimum from said gray scale maximum;

adaptively adjusting said first threshold; and

limiting said adaptively adjusted first threshold to a number of pixels divided by 2 n+1 .

16. The image processor according to claim 13 , wherein said processor is a digital signal processor.