Image-dependent automatic area of interest enhancement

• US 5,581,370 A
• Filed: 06/05/1995
• Issued: 12/03/1996
• Est. Priority Date: 06/05/1995
• Status: Expired due to Term

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Fuss, William A., Eschbach, Reiner
• Primary Examiner(s): Coles, Sr., Edward L.
• Assistant Examiner(s): NGUYEN, ANH-VINH THI
• Application Number:
  US08/461,913
• Time in Patent Office: 547 Days
• Field of Search: 358/447, 358/445, 358/451, 358/455, 358/500, 358/518, 358/520, 358/522, 382/254, 382/260, 382/261, 382/274
• US Class Current: 358/447
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/90 Dynamic range modification of images or parts thereof

Abstract

A method of improving the contrast in a natural scene image. A relevant histogram of the image is derived for from a selected subset of local histograms representing regions of the image. The signal describing the histogram is operated on with a filter having the characteristic of weakening strong peaks and valleys in the function, but not effecting flat portions of the signal. The filtered histogram signal is used for controlling the TRC mapping in a device at which the image is to be printed. To assure optimum selection of local histograms, regions including the black point and white point of an image are determined and added to the subset of local histograms representing regions of the image.

Claims

1. A method of improving the contrast in a natural scene image recorded as a set of electronic signals, including the steps of:

assuring that at least one component of the original electronic signal forming the natural scene image is defined with a signal describing intensity of the image;

from the intensity component of the electronic signals, deriving a set of local histograms signals, each describing the population of signals at possible intensity levels within the image for a region of the image;

selecting a subset of the local histogram signals as representing a maximum dynamic range of the natural scene image;

determining locations, by region, of the black point and white point within the electronic signals of the image;

assuring that the histograms representing regions in which the black point and white point occur are members of the subset of local histogram signals, and if not, adding them thereto;

combining the subset of local histograms with a selected weighting function to generate a relevant histogram;

operating on the relevant histogram signal with a filter that has the characteristics of reducing strong peaks and valleys in the signal;

using the filtered histogram signal, deriving a tonal mapping of input signals to output driver signals;

for each electronic signal forming the natural scene image, mapping the electronic signal to an output driver signal, using the tonal mapping derived.

2. The method as defined in claim 1, wherein each original electronic signal is defined in terms of red-green-blue color space.

3. The method as defined in claim 2, wherein each original electronic signal is defined in terms of red-green-blue color space is converted to luminance-chrominance space, and the signal describing intensity of the image is the luminance signal.

4. The method as defined in claim 1, wherein the step of deriving a tonal mapping of input signals to output driver signals includes the step of using a standard histogram equalization algorithm on the filtered histogram.

5. The method as defined in claim 1, including the initial step of sampling the electronic signals at a resolution less than the resolution of the electronic signals.

6. The method as defined in claim 1, wherein the step of operating on the histogram signal with a filter that has the characteristics of reducing strong peaks and valleys in the function, can be characterized by the function

7. A method of improving the contrast in a natural scene image recorded as a set of electronic signals, including the steps of:

receiving a set of color image-describing electronic signals from a source of natural scene images;

converting the received color image describing signals into a signal representing overall intensity of the image;

from the intensity signal, generating a global histogram signal, describing the population of intensity signals at possible intensity levels;

comparing the global histogram signal to a reference-flat signal and deriving a global variance from the comparison, representing the flatness of the histogram signal;

from the intensity signals, dividing the image into a plurality of discrete areas;

for each discrete area of the image, generating a local histogram signal, describing the population of intensity signals at possible intensity levels therewithin;

comparing each local histogram to a reference-flat signal, and deriving a local variance from the comparison, representing the flatness of the local histogram;

comparing each local variance to the global variance;

if any local variances are less than the global variance, summing a subset of local histogram signals having a variance less than the global variance value and obtaining a relevant histogram signal;

determining locations, by region, of the black point and white point within the electronic signals of the image;

assuring that the histograms representing regions in which the black point and white point occur are members of the subset of local histogram signals, and if not, adding them thereto;

combining the subset of local histograms with a selected weighting function to generate a relevant histogram;

filtering the relevant histogram signal or the global histogram signal if no local variance was less than the global variance;

using the filtered histogram signal, deriving a tonal mapping of input signals to printer driver signals;

for each electronic signal forming the natural scene image mapping the electronic signal to an output driver signal, using the tonal mapping derived.

8. The method as defined in claim 7, wherein the step of operating on the histogram signal with a filter that has the characteristics of reducing strong peaks and valleys in the function, can be characterized by the function

9. The method as defined in claim 7, including the step of sampling the red, green and blue electronic image signals at a resolution less than the resolution of the electronic image signals prior to converting the received red, green and blue image describing signals into signals representing overall intensity of the image.

10. The method as defined in claim 7, wherein the global histogram is used as the relevant histogram if not at least a preset number of local histograms have a variance value smaller than the global variance value.

11. The method as defined in claim 7, wherein the global variance is adjusted using a multiplicative adjustor, giving a new global variance to be used in the comparison with the local variances.

12. The method as defined in claim 7, wherein the multiplicative adjustor, is varied as a function of the global variance.

13. A method of improving contrast in a natural scene image represented by electronic signals including the steps:

a) developing a histogram representing critical areas of the image, where the critical areas include

i) areas having represented therein a significant portion of the dynamic range of the image, and

ii) areas in which the black point of the image and the white point of the image occur;

b) filtering the histogram with a filter to reduce strong peaks and valleys in the image;

c) developing a tone reproduction curve for the image mapping input signals to output signals based on the filtered histogram;

d) reproducing the image using the tone reproduction curve developed.