Method for automatically identifying targets in sonar images

• US 5,181,254 A
• Filed: 12/14/1990
• Issued: 01/19/1993
• Est. Priority Date: 12/14/1990
• Status: Expired due to Term

• Current Assignee: Northrop Grumman Systems Corporation (Northrop Grumman Corporation)
• Original Assignee: Westinghouse Electric Company LLC (Brookfield Asset Management Incorporated)
• Inventors: Schweizer, Philipp F., Petlevich, Walter J. Jr.
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US07/628,116
• Time in Patent Office: 767 Days
• Field of Search: 382/54, 382/22, 382/41, 382/1, 382/43, 367/131, 367/7, 209/590, 358/112
• US Class Current: 382/100
• CPC Class Codes: G01S 7/539 using analysis of echo signal for target characterisation; Target signature; Target cross-section
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components

Abstract

To detect and classify features in a sonar image comprised of a matrix of pixels each pixel having a known greyness level a set of windows is defined such that each window is comprised of a set of adjacent pixels such that every pixel of the image is included in at least one window. Properties of the greyness level for the pixels in each window are defined to enable selection of those windows having specific features of greyness level in excess of a predetermined threshold. Matched filter correlations are performed on the pixels in the selected windows to identify highlights and shadows. The identifying highlights and shadows are classified as targets, anomalies or background.

Claims

1. A method of detecting and classifying features in a sonar image comprised of a matrix of pixels, each pixel having a known greyness level comprising the steps of:

a) defining a set of windows, each window comprised of a set of adjacent pixels such that each pixel of the image is included in at least one window;

b) thereafter for each window performing the steps of;

i) filtering each window;

ii) performing a Fourier transform of each window;

iii) scaling each window;

iv) classifying each window which has been processed under steps (i) thru (iii) as a highlight, shadow or background according to the greyness levels of the pixels in each window;

c) selecting those windows which have been classified as highlight windows and shadow windows;

d) recording a location for each selected window relative to the image; and

e) classifying a set of windows as one of a highlight cluster, a shadow cluster, a highlight ridge, a shadow trough, an anomaly and background.

2. The method of claim 1 wherein each window is classified as highlight, shadow or background by applying a neural network algorithm to the window.

3. The method of claim 1 wherein a shadowhighlight algorithm is applied in step b(iv) by performing the steps of:

a) creating a histogram for the image;

b) defining sections of the histogram, each section having a mean and standard deviation;

c) computing a threshold value for each section based on the mean and standard deviation;

d) creating a binary image using the threshold value for each section of each convolved image;

e) combining all binary images to produce a final binary image corresponding to all detections

f) determining a number of all detections and a location for each detection by thresholding the final binary image; and

g) reporting the location of each detection.

4. The method of claim 1 wherein a statistical detection algorithm is applied, the statistical algorithm performing steps which include:

a) defining an image background window as reference window; and

b) comparing each scanned overlapping window in the image with this background reference window.

5. The method of claim 1 wherein each window is classified as a highlight cluster, a shadow cluster, a highlight ridge, a shadow trough, an anomaly or background by applying a neural network algorithm to the window.