Method and apparatus for performing an adaptive extended dynamic range algorithm

• US 6,546,124 B1
• Filed: 10/25/1999
• Issued: 04/08/2003
• Est. Priority Date: 07/02/1999
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Langan, David Allen, Hopple, Michael Robert
• Primary Examiner(s): Johns, Andrew W.
• Assistant Examiner(s): NAKHJAVAN, SHERVIN K
• Application Number:
  US09/426,251
• Time in Patent Office: 1,261 Days
• Field of Search: 382/128, 382/132, 382/168, 382/169, 382/209, 382/232, 382/237, 382/257, 382/260, 382/263, 382/274, 382/282, 382/308, 348/384.1, 348/387.1, 250/227.2, 250/227.21, 378/4, 378/5, 378/16, 378/21, 378/23, 378/91, 378/98.7, 378/98.11, 378/98.12, 378/162, 708/230, 708/300, 708/303, 708/313, 708/315, 708/490, 708/505, 708/670, 708/819, 326/22, 326/88
• US Class Current: 382/132
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement

Abstract

A method and apparatus are provided for performing an extended dynamic range (EDR) algorithm. The apparatus comprises an EDR processing component which performs the EDR algorithm of the invention in order to compress an image input to the EDR processing component down to a desired gray scale range. The EDR processing component automatically adapts to the dynamic range of an input image so that the input image is compressed down to a desired gray scale range while preserving high frequency contrast information contained in the input image. In addition to automatically adjusting to the dynamic range of the image, the EDR processing component compresses the input image in such a manner that the difference between the input image intensity and the local mean intensity value is taken into account, which prevents artifacts associated with enhanced negative contrast from appearing in the compressed image.

Claims

1. An extended dynamic range processing component for processing an input image to generate an output image which has been compressed down to a desired gray scale intensity range, the EDR processing component performing an EDR algorithm, the EDR processing component comprising:

logic configured to process the input image in accordance with the EDR algorithm to generate the compressed output image, the logic determining a local mean intensity value of the input image based on an input image intensity value and a plurality of image intensity values surrounding the input image intensity value, the logic utilizing the local mean intensity value to automatically determine the dynamic range of the input image, the logic utilizing the determination of the dynamic range and the local mean intensity value to determine a percentage of the local mean intensity value to be subtracted from the input image intensity value, the logic subtracting the determined percentage of the local mean intensity value from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range.

2. The EDR processing component of claim 1, wherein after the logic subtracts the determined percentage of the local mean intensity value from the input image intensity value, the logic determines whether the compressed input image intensity value has been compressed down to the desired gray scale intensity range, and, if it is determined that the compressed input image intensity value has not been fully compressed down to the desired gray scale intensity range, the logic further compresses the compressed input image intensity value down to the desired gray scale range.

3. The EDR processing component of claim 2, wherein the logic comprises:

a BOXCAR component, the BOXCAR component receiving the input image intensity value and the plurality of image intensity values surrounding the input image intensity value and calculating the local mean intensity value based on the input image intensity value and the plurality of surrounding input image intensity values;

a MAX component, the MAX component receiving the local mean intensity value calculated by the BOXCAR component, the MAX component utilizing the local mean intensity value calculated by the BOXCAR component to determine the dynamic range of the input image;

a BOOST component, the BOOST component receiving the local mean intensity value from the BOXCAR component and receiving an indication of the dynamic range of the input image from the MAX component, the BOOST component utilizing the local mean intensity value and the indication of the dynamic range to determine the percentage of the local mean intensity value to be subtracted from the input image intensity value;

an ADD component, the ADD component receiving the percentage of the local mean intensity value from the BOOST component and receiving the input image intensity value, the ADD component subtracting the percentage of the local mean intensity value from the input image intensity value to obtain the compressed input image intensity value; and

a GAMMA component, the GAMMA component receiving the compressed input image intensity value from the ADD component, wherein the GAMMA component determines whether the compressed input image intensity value is within the desired gray scale intensity range, the GAMMA component further compressing the compressed input image intensity value down to the desired gray scale range if the GAMMA component determines that the compressed input image intensity value received from the ADD component is not within the desired gray scale intensity range.

4. The EDR processing component of claim 3, wherein the BOOST component comprises at least one lookup table, the BOOST component lookup table being indexed by the local mean intensity value received from the BOXCAR component and by the indication of the dynamic range received from the MAX component, the BOOST component lookup table outputting a value from the indexed location in the lookup table, wherein the value output from the BOOST component lookup table corresponds to the percentage of the local mean intensity value received by the ADD component and subtracted from the input image intensity value by the ADD component.

5. The EDR processing component of claim 4, wherein the BOOST component comprises a plurality of lookup tables, and wherein the indication of the dynamic range received by the BOOST component from the MAX component selects one of the plurality of lookup tables, and wherein the local mean intensity value received by the BOOST component from the BOXCAR component indexes a particular location in the selected lookup table.

6. The EDR processing component of claim 5, wherein the GAMMA component receives the indication of the dynamic range from the MAX component, the GAMMA component comprising at least one lookup table, the GAMMA component lookup table being indexed by the compressed input image intensity value received from the ADD component and by the indication of the dynamic range received from the MAX component, the GAMMA component lookup table outputting an output image intensity value which is within the desired dynamic range, wherein the values stored in the GAMMA component lookup table are pre-selected to perform a roll off function, wherein when the indication of the dynamic range indicates that the dynamic range is relatively large, the output image intensity value output from the GAMMA component lookup table will correspond to a relatively gradual roll off of the compressed input image intensity value received from the ADD component, and wherein when the indication of the dynamic range indicates that the dynamic range is relatively small, the output image intensity value output from the GAMMA component lookup table will correspond to a relatively less gradual roll off of the compressed input image intensity value received from the ADD component.

7. The EDR processing component of claim 6, wherein the EDR algorithm performed by the EDR processing component is expressed algorithmically as:

8. The EDR processing component of claim 7, wherein the logic comprises a computer, the computer comprising a processor configured to perform the EDR algorithm, the EDR algorithm corresponding to functionalities of the BOXCAR, MAX, BOOST, ADD and GAMMA components, wherein the EDR algorithm is performed in software being executed by the processor.

9. The EDR processing component of claim 7, wherein the logic is comprised as a combination of integrated circuits configured to perform the EDR algorithm wherein the combination of integrated circuits corresponds to hardware configured to perform the EDR algorithm.

10. An extended dynamic range processing component for processing an input image to generate an output image which has been compressed down to a desired gray scale intensity range, the EDR processing component performing an EDR algorithm, the EDR processing component comprising:

logic configured to process the input image in accordance with the EDR algorithm to generate the compressed output image, the logic determining a local mean intensity value of the input image based on an input image intensity value and a plurality of image intensity values surrounding the input image intensity value, the logic calculating a difference between the local mean intensity value and the input image intensity value, the logic utilizing the calculated difference to determine a percentage of the local mean intensity value to be subtracted from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range, the logic subtracting the determined percentage of the local mean intensity value from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range.

11. The EDR processing component of claim 10, wherein after the logic subtracts the determined percentage of the local mean intensity value from the input image intensity value, the logic determines whether the compressed input image intensity value has been compressed down to the desired gray scale intensity range, and, if it is determined that the compressed input image intensity value has not been fully compressed down to the desired gray scale intensity range, the logic further compresses the compressed input image intensity value down to the desired gray scale range.

12. The EDR processing component of claim 11, wherein the logic comprises;

a BOXCAR component, the BOXCAR component receiving the input image intensity value and the plurality of image intensity.values surrounding the input image intensity value and calculating the local mean intensity value based on the input image intensity value and the plurality of surrounding input image intensity values;

a FUNCTION component, the FUNCTION component receiving the 10 local mean intensity value calculated by the BOXCAR component, the FUNCTION component receiving the input image intensity value, the FUNCTION component calculating the difference between the local mean intensity value calculated by the BOXCAR component and the input image intensity value, the FUNCTION component processing the difference in accordance with a predetermined function to obtain an Adjusted_Mean value;

a BOOST component, the BOOST component receiving the Adjusted_Mean value from the FUNCTION component, the BOOST component utilizing the Adjusted_Mean value to obtain a percentage of the local mean intensity value to be subtracted from the input image intensity value;

an ADD component, the ADD component receiving the percentage of the local mean intensity value from the BOOST component and receiving the input image intensity value, the ADD component subtracting the percentage of the local mean intensity value from the input image intensity value to obtain the compressed input image intensity value; and

a GAMMA component, the GAMMA component receiving the compressed input image intensity value from the ADD component, wherein the GAMMA component determines whether the compressed input image intensity value is within the desired gray scale intensity range, the GAMMA component further compressing the compressed input image intensity value down to the desired gray scale range if the GAMMA component determines that the compressed input image intensity value received from the ADD component is not within the desired gray scale intensity range.

13. The EDR processing component of claim 12, wherein the BOOST component comprises at least one lookup table, the BOOST component lookup table being indexed by the Adjusted_Mean value received from the FUNCTION component, the BOOST component lookup table outputting a value from the indexed location in the lookup table, wherein the value output from the BOOST component lookup table corresponds to the percentage of the local mean intensity value received by the ADD component and subtracted from the input image intensity value by the ADD component.

14. The EDR processing component of claim 13, wherein the GAMMA component comprises at least one lookup table, the GAMMA component lookup table being indexed by the compressed input image intensity value received from the ADD component, the GAMMA component lookup table outputting an output image intensity value which is within the desired dynamic range, wherein the values stored in the GAMMA component lookup table are pre-selected to perform a roll off function which gradually compresses the compressed input image intensity values down to the desired gray scale intensity range.

15. The EDR processing component of claim 14, wherein the EDR algorithm performed by the EDR processing component is expressed algorithmically as:

16. The EDR processing component of claim 15, wherein the logic comprises a computer, the computer comprising a processor configured to perform the EDR algorithm, the EDR algorithm corresponding to functionalities of the BOXCAR, BOOST, ADD and GAMMA components, wherein the EDR algorithm is performed in software being executed by the processor.

17. The EDR processing component of claim 15, wherein the logic is comprised as a combination of integrated circuits configured to perform the EDR algorithm wherein the combination of integrated circuits corresponds to hardware configured to perform the EDR algorithm.

18. An extended dynamic range processing component for processing an input image to generate an output image which has been compressed down to a desired gray scale intensity range, the EDR processing component performing an EDR algorithm, the EDR processing component comprising:

logic configured to process the input image in accordance with the EDR algorithm to generate the compressed output image, the logic determining a local mean intensity value of the input image based on an input image intensity value and a plurality of image intensity values surrounding the input image intensity value, the logic calculating a difference between the local mean intensity value and the input image intensity value, the logic utilizing the local mean intensity value to automatically determine the dynamic range of the input image, the logic utilizing the determination of the dynamic range and the calculated difference to determine a percentage of the local mean intensity value to be subtracted from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range, the logic subtracting the determined percentage of the local mean intensity value from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range.

19. The EDR processing component of claim 18, wherein after the logic subtracts the determined percentage of the local mean intensity value from the input image intensity value, the logic determines whether the compressed input image intensity value has been compressed down to the desired gray scale intensity range, and, if it is determined that the compressed input image intensity value has not been fully compressed down to the desired gray scale intensity range, the logic further compresses the compressed input image intensity value down to the desired gray scale range.

20. The EDR processing component of claim 19, wherein the logic comprises:

a BOXCAR component, the BOXCAR component receiving the input image intensity value and the plurality of image intensity values surrounding the input image intensity value and calculating the local mean intensity value based on the input image intensity value and the plurality of surrounding input image intensity values;

a FUNCTION component, the FUNCTION component receiving the local mean intensity value calculated by the BOXCAR component, the FUNCTION component receiving the input image intensity value, the FUNCTION component calculating the difference between the local mean intensity value calculated by the BOXCAR component and the input image intensity value, the FUNCTION component processing the difference in accordance with a predetermined function to obtain an Adjusted_Mean value;

a MAX component, the MAX component receiving the local mean intensity value calculated by the BOXCAR component, the MAX component utilizing the local mean intensity value calculated by the BOXCAR component to determine the dynamic range of the input image;

a BOOST component, the BOOST component receiving the Adjusted_Mean value from the FUNCTION component, the BOOST component receiving an indication of the dynamic range from the BOXCAR component, the BOOST component utilizing the Adjusted_Mean value and the indication of the dynamic range to obtain a percentage of the local mean intensity value to be subtracted from the input image intensity value;

an ADD component, the ADD component receiving the percentage of the local mean intensity value from the BOOST component and receiving the input image intensity value, the ADD component subtracting the percentage of the local mean intensity value from the input image intensity value to obtain the compressed input image intensity value; and

a GAMMA component, the GAMMA component receiving the compressed input image intensity value from the ADD component, wherein the GAMMA component determines whether the compressed input image intensity value is within the desired gray scale intensity range, the GAMMA component further compressing the compressed input image intensity value down to the desired gray scale range if the GAMMA component determines that the compressed input image intensity value received from the ADD component is not within the desired gray scale intensity range.

21. The EDR processing component of claim 20, wherein the BOOST component comprises at least one lookup table, the BOOST component lookup table being indexed by the Adjusted_Mean value received from the FUNCTION component and by the indication of the dynamic range received from the MAX component, the BOOST component lookup table outputting a value from the indexed location in the lookup table, wherein the value output from the BOOST component lookup table corresponds to the percentage of the local mean intensity value received by the ADD component and subtracted from the input image intensity value by the ADD component.

22. The EDR processing component of claim 21, wherein the GAMMA component comprises at least one lookup table, the GAMMA component receiving the indication of the dynamic range from the MAX component, the GAMMA component lookup table being indexed by the compressed input image intensity value received from the ADD component and by the indication of the dynamic range, the GAMMA component lookup table outputting an output image intensity value which is within the desired dynamic range, wherein the values stored in the GAMMA component lookup table are pre-selected to perform a roll off function, wherein when the indication of the dynamic range indicates that the dynamic range is relatively large, the output image intensity value output from the GAMMA component lookup table will correspond to a relatively gradual roll off of the compressed input image intensity value received from the ADD component, and wherein when the indication of the dynamic range indicates that the dynamic range is relatively small, the output image intensity value output from the GAMMA component lookup table will correspond to a relatively less gradual roll off of the compressed input image intensity value received from the ADD component.

23. The EDR processing component of claim 22, wherein the EDR algorithm performed by the EDR processing component is expressed algorithmically as:

24. The EDR processing component of claim 23, wherein the logic comprises a computer, the computer comprising a processor configured to perform the EDR algorithm, the EDR algorithm corresponding to functionalities of the BOXCAR, MAX, FUNCTION, BOOST, ADD and GAMMA components, wherein the EDR algorithm is performed in software being executed by the processor.

25. The EDR processing component of claim 23, wherein the logic is comprised as a combination of integrated circuits configured to perform the EDR algorithm wherein the combination of integrated circuits corresponds to hardware configured to perform the EDR algorithm.

26. A method for performing an extended dynamic range (EDR) algorithm, the algorithm processing an input image to generate an output image which has been compressed down to a desired gray scale intensity range, the method comprising the steps of:

determining a local mean intensity value of the input image based on an input image intensity value and a plurality of image intensity values surrounding the input image intensity value;

utilizing the local mean intensity value to automatically determine the dynamic range of the input image;

utilizing the determination of the dynamic range and the local mean intensity value to determine a percentage of the local mean intensity value to be subtracted from the input image intensity value; and

subtracting the determined percentage of the local mean intensity value from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range.

27. A method for performing an extended dynamic range (EDR) algorithm, the algorithm processing an input image to generate an output image which has been compressed down to a desired gray scale intensity range, the method comprising the steps of:

determining a local mean intensity value of the input image based on an input image intensity value and a plurality of image intensity values surrounding the input image intensity value;

calculating a difference between the local mean intensity value and the input image intensity value;

utilizing the calculated difference to determine a percentage of the local mean intensity value to be subtracted from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range; and

subtracting the determined percentage of the local mean intensity value from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range.

28. A method for performing an extended dynamic range (EDR) algorithm, the algorithm processing an input image to generate an output image which has been compressed down to a desired gray scale intensity range, the method comprising the steps of:

determining a local mean intensity value of the input image based on an input image intensity value and a plurality of image intensity values surrounding the input image intensity value;

calculating a difference between the local mean intensity value and the input image intensity value;

utilizing the local mean intensity value to automatically determine the dynamic range of the input image;

utilizing the determination of the dynamic range and the calculated difference to determine a percentage of the local mean intensity value to be subtracted from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range; and

subtracting the determined percentage of the local mean intensity value from the input image intensity value to compress the input image intensity value down to the desired gray scale intensity range.