Method and apparatus for image contrast quality evaluation

• US 5,642,433 A
• Filed: 07/31/1995
• Issued: 06/24/1997
• Est. Priority Date: 07/31/1995
• Status: Expired due to Term

• Current Assignee: Tripath Imaging Incorporated (Becton, Dickinson & Co)
• Original Assignee: NeoPath, Inc. (Becton, Dickinson & Co)
• Inventors: Lee, Shih-Jong J., Wilhelm, Paul S.
• Primary Examiner(s): JOHNS, ANDREW W
• Application Number:
  US08/509,185
• Time in Patent Office: 694 Days
• Field of Search: 382/128, 382/133, 382/168, 382/199, 382/256, 382/257, 382/274, 382/286, 382/308
• US Class Current: 382/199
• CPC Class Codes: G01N 15/1433 using image recognition
• CPC Class Codes: G02B 21/365 Control or image processing arrangements for digital or video microscopes G02B21/361, G02B21/362 take precedence
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/20036 Morphological image processing
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: G06V 20/693 Acquisition

Abstract

A method for single image based image contrast quality evaluation is used in a computer controlled image system including an automated microscope for acquiring images. An image is acquired. An image contrast score is computed for the image. An image reliability score is computed for the image. The image contrast score and image reliability score are accumulated and combined to produce a contrast quality score. The contrast quality score is compared to a predetermined limit so as to produce a quality of processing result.

Claims

1. A method for measuring the contrast quality of an image comprising the steps of:

(a) acquiring an image;

(b) operating on the image to compute a first edge intensity value; and

(c) operating on the image to compute an edge of edge intensity value; and

(d) computing an image contrast score for the image as the ratio of the edge of edge intensity value to the first edge intensity value.

2. The method of claim 1 wherein the step of operating on the image to compute a first edge intensity value further comprises the steps of:

(a) detecting an edge gradient of the single image to produce an edge gradient image; and

(b) accumulating values of the edge gradient image.

3. The method of claim 2 further comprising the steps of:

(a) image conditioning the edge gradient image so as to produce a conditioned edge gradient image; and

(b) processing the conditioned edge gradient image so as to produce an edge of edge image.

4. The method of claim 3 wherein the step of image conditioning the edge gradient image so as to produce a conditioned edge gradient image limits the edge gradient image value to a predetermined minimum limit.

5. The method of claim 3 wherein the step of operating on the image to compute an edge of edge intensity value further comprises computing an accumulated value of the edge of edge image.

6. The method of claim 2 wherein the step of detecting an edge gradient of the single image to produce an edge gradient image comprises a morphological operation.

7. The method of claim 6 wherein the morphological operation comprises applying a dilation residue operator to the image.

8. The method of claim 6 wherein the morphological operation comprises applying an erosion residue operator to the image.

9. The method of claim 6 wherein the morphological operation comprises the step of morphologically combining detected edges.

10. The method of claim 1 further comprising the step of computing an image reliability score for the image derived from the first edge intensity value.

11. The method of claim 10 further comprising the step of comparing the image reliability score to a predetermined limit so as to produce a quality of processing result.

12. The method of claim 10 wherein the image reliability score is low, it is an indication that the image contrast data may be derived from an image with insufficient material for reliable contrast evaluation.

13. The method of claim 10 further comprising the step of combining the image contrast score and image reliability score.

14. A method for contrast quality evaluation of a biological specimen comprising the steps of:

(a) acquiring at least one image for a specimen;

(b) computing at least one image contrast score for the at least one image;

(c) computing at least one image reliability score for the at least one image;

(d) accumulating and combining the at least one image contrast score and the at least one image reliability score to produce a quality of processing measure; and

(e) comparing the quality of processing measure to a predetermined limit to produce a quality of processing result.

15. The method of claim 14 wherein the step of accumulating and combining the image contrast score and image reliability score further comprises the steps of:

(a) accumulating the image contrast score and image reliability score in a histogram; and

(b) computing a summary statistic of the histogram so as to produce a quality of processing measure.

16. A method for single image based image contrast quality evaluation, comprising the steps of:

(a) acquiring an image;

(b) detecting an edge gradient of the image to produce an edge gradient image;

(c) processing the edge gradient image so as to produce an edge of edges image;

(d) processing the edge gradient image so as to produce an accumulated edge intensity value;

(e) processing the edge of edges image so as to produce an accumulated edges of edges intensity value;

(f) computing an image contrast score for the image wherein the image contrast score comprises a ratio of the accumulated edges of edges intensity value to the accumulated edge intensity value; and

(g) computing an image reliability score for the image wherein the image reliability score comprises the accumulated edge intensity value; and

(h) combining the image contrast score and the image reliability score so as to produce a quality of processing measure.