Identifying an area of interest using histogram data arranged in predetermined sequence

• US 5,588,071 A
• Filed: 10/26/1994
• Issued: 12/24/1996
• Est. Priority Date: 10/26/1994
• Status: Expired due to Fees

• Current Assignee: CareStream Health Incorporated
• Original Assignee: Minnesota Mining and Manufacturing Co (3M Company)
• Inventors: Schultz, John C.
• Primary Examiner(s): JOHNS, ANDREW W
• Application Number:
  US08/329,493
• Time in Patent Office: 790 Days
• Field of Search: 382/132, 382/168, 382/169, 382/171, 382/282, 382/291, 358/453, 358/538, 364/413.13
• US Class Current: 382/168
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/20132 Image cropping
• CPC Class Codes: G06T 2207/30061 Lung
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/90 Dynamic range modification of images or parts thereof
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

A method for automatic analysis and enhancement of digitized images to identify the area of interest (AOI) from background areas divides a digitized image into a number of regions, and calculates histograms of gray scale or density for each region. An image of the regions in histogram space is created according to a selected sequence of the regions, chosen to preserve spatial proximity. This image is then processed by peak detection to identify features grouped spatially and by similarity of histograms. From this the AOI is identified, and the contrast enhancement is applied to the original image, according to the characteristics of the output device. The method is particularly adapted for enhancing medical X-ray images, for output to viewing devices.

Claims

1. A method of analyzing a data set of a digitized image to identify an area of interest, the method comprising the steps of:

dividing the digitized image into a plurality of regions;

calculating histogram data for frequencies of image values within each of the plurality of regions;

organizing the histogram data for each of the regions according to a predetermined sequence of the plurality of regions to create a data set for an image in histogram space of the plurality of regions; and

analyzing said data set for the image in histogram space of the plurality of regions to identify localities of maxima and minima in grouped image value characteristics which correspond to the area of interest within the digitized image.

2. The method of claim 1, wherein said predetermined sequence comprises a spiral sequence of the plurality of regions around the digitized image.

3. The method of claim 1, wherein said predetermined sequence comprises a Peano sequence of the plurality of regions around the digitized image.

4. The method of claim 1, wherein said predetermined sequence comprises a raster sequence of the plurality of regions around the digitized image.

5. The method of claim 1, wherein said predetermined sequence comprises a sweep sequence of the plurality of regions around the digitized image.

6. The method of claim 1, wherein said digitized image represents an image obtained by radiography.

7. A method of analyzing a data set of a digitized image to identify an area of interest, the method comprising the steps of:

dividing the digitized image into an array of regions;

calculating histogram data for frequencies of density values within each of the regions;

establishing an ordered sequence of the regions;

organizing the histogram data for each of the regions according to the ordered sequence to create a data set for an image in histogram space of the regions; and

analyzing said histogram space image data set to identify localities of maxima and minima in grouped density values which correspond to the area of interest within the digitized image.

8. The method of claim 7, wherein the step of establishing said ordered sequence comprises establishing a spiral sequence of the regions around the digitized image.

9. The method of claim 7, wherein the step of establishing said ordered sequence comprises establishing a Peano sequence of the regions around the digitized image.

10. The method of claim 7, wherein the step of establishing said ordered sequence comprises establishing a raster sequence of the regions around the digitized image.

11. The method of claim 7, wherein the step of establishing said ordered sequence comprises establishing a sweep sequence of the regions around the digitized image.

12. The method of claim 7, wherein said digitized image represents an image obtained by radiography.

13. A method of enhancing image contrast of an area of interest in a digitized image represented by a set of pixel data, comprising:

dividing the digitized image into an array of regions;

calculating histogram data for frequencies of density values for pixels within each of the regions;

establishing an ordered sequence of the regions;

ordering the histogram data according to said ordered sequence of the regions to create a set of data representing an image of the regions in histogram space;

analyzing said set of data representing the image of the regions in histogram space to identify localities of peaks in density values which correspond to an area of interest within the digitized image; and

applying a contrast enhancement function to create a new set of pixel data, the contrast enhancement function changing density values of the pixel data in the area of interest to create the new set of pixel data.

14. The method of claim 13, wherein the step of establishing said ordered sequence comprises establishing a spiral sequence of the regions around the digitized image.

15. The method of claim 13, wherein the step of establishing said ordered sequence comprises establishing a Peano sequence of the regions around the digitized image.

16. The method claim 13 wherein the digitized image is a radiography image, and wherein the identified localities of the peaks correspond to physiological structures in the radiography image.

17. The method of claim 13 wherein the contrast enhancement function adjusts contrast of the area of interest to match characteristics of an output viewing device.

18. A method of identifying an area of interest in a digitized image, the method comprising the steps of:

dividing the digitized image into a plurality of regions;

calculating histogram data for frequencies of image values within each of the plurality of regions;

storing the histogram data for each of the regions according to a predetermined spatial sequence of the plurality of regions to create a data set for an image in histogram space of the plurality of regions, said predetermined spatial sequence substantially separating histogram data for the plurality of regions at an edge portion of said digitized image from histogram data for the plurality of regions at an interior portion of said digitized image; and

identifying localities of maxima and minima of said image values in said data set to determine the area of interest within the digitized image.

19. The method of claim 18, wherein the predetermined spatial sequence comprises a spiral sequence of the plurality of regions around the digitized image.

20. The method of claim 18, wherein the ordered sequence comprises establishing a Peano sequence of the plurality of regions around the digitized image.

21. The method of claim 18, wherein the predetermined spatial sequence comprises a raster sequence of the plurality of regions around the digitized image.

22. The method of claim 18, wherein the predetermined spatial sequence comprises a sweep sequence of the plurality of regions around the digitized image.

23. The method of claim 18, wherein said digitized image represents an image obtained by radiography.