Method for automatic detection of targets within a digital image

• US 6,434,261 B1
• Filed: 02/23/1998
• Issued: 08/13/2002
• Est. Priority Date: 02/23/1998
• Status: Expired due to Term

• Current Assignee: III Holdings 1, LLC
• Original Assignee: Board of Regents of the University of Texas System (University of Texas System)
• Inventors: Zhang, Xiao-Ping, Desai, Mita D.
• Primary Examiner(s): Au, Amelia M.
• Assistant Examiner(s): Dastouri, Mehrdad
• Application Number:
  US09/028,250
• Time in Patent Office: 1,632 Days
• Field of Search: 382/128, 382/132, 382/169, 382/170, 382/171, 382/172
• US Class Current: 382/132
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/20008 Globally adaptive
• CPC Class Codes: G06T 2207/20064 Wavelet transform [DWT]
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 2207/30096 Tumor; Lesion
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/143 involving probabilistic approaches, e.g. Markov random field [MRF] modelling

Abstract

The present invention comprises a method for the detection and segmentation of bright images within a digital image using wavelets. One example of such a bright image is microcalcifications within a mammogram. Multiresolution analysis may be used to detect and segment the possible microcalcification areas by combining Bayes classifiers. By analyzing the time-frequency characteristics of clustered microcalcifications, the inventors first choose the optimized wavelet for the detection of microcalcifications. A wavelet packet analysis is then used to detect different size microcalcifications. An adaptive method of choosing the threshold for detection by using a one-dimensional wavelet transform to analyze the PDF of the images at different scales is used. Then, a scheme to detect different size microcalcifications in different scale wavelet packet transformed images is developed. The processing results show that the inventors'"'"' methods are quite effective on a broad sample of mammograms in detecting microcalcifications even at low contrast. These segmentation and detection methods will prove useful in other applications, such as other medical imaging infrared applications, forward-looking infrared radar (FLIR), and other technologies.

Claims

1. A method of determining an adaptive threshold for a wavelet analysis of a digital image, comprising the steps of:

decomposing the digital image by wavelet packet transform to obtain a plurality of transformed images of different scale channels;

obtaining a histogram of said transformed images;

decomposing the histogram by wavelet transform to obtain a plurality of histogram images of different scale channels; and

selecting said adaptive threshold from said plurality of histogram images.

2. The method of claim 1, wherein said digital image comprises a mammogram.

3. The method of claim 1, wherein said digital image comprises a forward-looking infrared image.

12. The method of claim 1, wherein the threshold is determined as a local minimum point of a scale of the histogram of the transformed image.

13. The method of claim 1, wherein the threshold is determined as the largest local minimum point of the scales of the histogram of the transformed image.

14. The method of claim 1, wherein the threshold is determined as a weighted average of the local minima of the scales of the histogram.

4. A method of determining an adaptive threshold for a wavelet analysis of a digital image, comprising the steps of:

decomposing the digital image by wavelet packet transform to obtain a plurality of transformed images of different scale channels; and

selecting a minimum point of a probability distribution function of said plurality of transformed images by performing a wavelet analysis of the probability distribution function.

5. The method of claim 4, wherein said digital image comprises a mammogram.

6. The method of claim 4, wherein said digital image comprises a forward-looking infrared image.

7. A method for automatically detecting an abnormal portion of a digital image, comprising the steps of:

obtaining the digital image;

selecting a mother wavelet optimized for said digital image;

decomposing the digital image by wavelet packet transform into a plurality of channels;

determining an adaptive threshold from all of said plurality of channels by finding a local minimum point for each of one or more scales of a probability distribution function of the image transform;

comparing said adaptive threshold to each of said plurality of channels;

eliminating each of said plurality of channels less than said minimum threshold to obtain an adjusted singularity map;

restoring said adjusted singularity map to restored image; and

locating regions in said digital image corresponding to said restore image.

8. The method of claim 7, wherein said digital image comprises a mammogram.

9. The method of claim 7, wherein said digital image comprises a forward-looking infrared image.

15. The method of claim 7, wherein the threshold is determined as a local minimum point of a scale of the PDF of the transformed image.

16. The method of claim 7, wherein the threshold is determined as the largest local minimum point of the scales of the PDF of the transformed image.

17. The method of claim 7, wherein the threshold is determined as a weighted average of the local minima of the scales of the PDF.

10. A method of obtaining a microcalcifications map of a digitalized mammogram image, comprising the steps of:

inputting said digitalized mammogram image into a processing system;

preprocessing said image;

obtaining wavelet packet transforms of the image to obtain transformed images;

performing wavelet-based analysis of a probability distribution function of the transformed images at a plurality of channels;

selecting thresholds and obtaining segmented images for each said plurality of channels;

mapping the segmented images of each said channel to the original digitalized mammogram image to obtain a plurality of mapped images; and

combining all of the mapped images to obtain the microcalcifications map.

11. A method of obtaining a microcalcifications map of a digitalized mammogram image, comprising the steps of:

inputting said digitalized mammogram image into a processing system;

preprocessing said image;

obtaining wavelet packet transforms of the image to obtain transformed images;

performing wavelet-based analysis of probability distribution function of the transformed images at a plurality of channels;

selecting thresholds and obtaining segmented images for each said plurality of channels;

mapping the segmented images of each said channel to the original digitalized mammogram image to obtain a plurality of mapped images; and

combining all of the mapped images to obtain the microcalcifications map;

wherein said selecting and obtaining step further comprises the step of;

selecting a one dimensional wavelet analysis for each said plurality of channels and obtaining a wavelet transformed probability distribution function; and

using a first local minimum larger than a middle point of the transformed probability distribution function as the threshold.

18. A method of determining an adaptive threshold for a wavelet analysis of a digital image, comprising the steps of:

decomposing the digital image by wavelet transformation to obtain a plurality of transformed images of different scale channels; and

determining the largest local minimum point of scales of a wavelet transformed probability distribution function of the transformed images.

19. The method of claim 18, wherein the probability distribution function is determined as the histogram of the image transforms.

20. A method of determining an adaptive threshold for a wavelet analysis of a digital image, comprising the steps of:

decomposing the digital image by wavelet transformation to obtain a plurality of transformed images of different scale channels; and

determining a weighted average of a plurality of local minimum points of scales of a wavelet transformed probability distribution function of the transformed images.

21. The method of claim 20, wherein the probability distribution function is determined as the histogram of the image transforms.