Methods and systems for identifying a thin object

• US 8,254,676 B2
• Filed: 12/31/2007
• Issued: 08/28/2012
• Est. Priority Date: 12/31/2007
• Status: Active Grant

• Current Assignee: Morpho Detection (Safran SA)
• Original Assignee: Morpho Detection (Safran SA)
• Inventors: Ioannou, Dimitrios, Gable, Todd
• Primary Examiner(s): CHAWAN, SHEELA C
• Application Number:
  US11/967,464
• Publication Number:
  US 20090169104A1
• Time in Patent Office: 1,702 Days
• Field of Search: 382/181, 382/170, 382/168, 382/173, 382/190, 382/191, 382/232, 382/254, 382/276, 382/305, 382/224, 382/308
• US Class Current: 382/170
• CPC Class Codes: G06V 10/507 Summing image-intensity values; Histogram projection analysis
• CPC Class Codes: G06V 2201/05 Recognition of patterns representing particular kinds of hidden objects, e.g. weapons, explosives, drugs

Abstract

A method for identifying an object within a container is provided. The method includes acquiring image data representing an image, applying a morphological operator to the acquired image data to generate morphed image data, calculating a histogram based on the morphed image data, and classifying the image using the calculated histogram. A classification of the image may be displayed and/or stored in a computer-readable memory.

Claims

1. A method for identifying an object within a container, said method comprising:

acquiring image data representing an image;

applying a morphological operator to the acquired image data to generate morphed image data in which each image element of the image data is assigned an integer value representing contact with neighboring image elements;

calculating, by a processor, a histogram based on the integer values of each image element in the morphed image data;

classifying the image using the calculated histogram; and

outputting a classification of the image.

2. A method in accordance with claim 1, further comprising segmenting the acquired image data into a plurality of image segments.

3. A method in accordance with claim 2, further comprising labeling each image segment of the plurality of image segments.

4. A method in accordance with claim 1, wherein applying a morphological operator further comprises performing a binary operation on the acquired image data to assign each image element a binary value.

5. A method in accordance with claim 4, wherein applying a morphological operator further comprises assigning the integer value to each image element of the acquired image data based on a binary value of an image element of interest and binary values of image elements surrounding the image element of interest.

6. A method in accordance with claim 1, wherein applying a morphological operator further comprises applying an averaging filter to the acquired image data.

7. A method in accordance with claim 1, wherein classifying the image using the calculated histogram further comprises comparing the calculated histogram to a plurality of pre-stored image signatures.

8. A method in accordance with claim 1, wherein classifying the image further comprises classifying the image as one of a bulk object, a thin object, and a random aggregation of image elements.

9. A system for identifying an object within a container, said system comprising:

a data collection system; and

a detection classification system coupled to said data collection system, said detection classification system configured to;

acquire image data representing an image;

apply a morphological operator to the acquired image data to generate morphed image data in which each image element of the image data is assigned an integer value representing contact with neighboring image elements;

calculate a histogram based on the integer values of each image element in the morphed image data;

classify the image using the calculated histogram; and

output a classification of the image.

10. A system in accordance with claim 9, wherein said detection classification system is further configured to segment the acquired image data into a plurality of image segments.

11. A system in accordance with claim 10, wherein said detection classification system is further configured to label each image segment of the plurality of image segments.

12. A system in accordance with claim 9, wherein said detection classification system is further configured to perform a binary operation on the acquired image data to assign each image element a binary value.

13. A system in accordance with claim 12, wherein said detection classification system is further configured to assign the integer value to each image element of the acquired image data based on a binary value of an image element of interest and binary values of image elements surrounding the image element of interest.

14. A system in accordance with claim 9, wherein said detection classification system is further configured to apply an averaging filter to the acquired image data.

15. A system in accordance with claim 9, wherein said detection classification system is further configured to compare the calculated histogram to a plurality of pre-stored image signatures.

16. A system in accordance with claim 9, wherein said detection classification system is further configured to classify the image as one of a bulk object, a thin object, and a random aggregation of image elements.

17. A computer program embodied on a non-transitory computer-readable medium, said computer program comprising a code segment that configures a processor to:

receive image data representing an image;

apply a morphological operator to the acquired image data to generate morphed image data in which each image element of the image data is assigned an integer value representing contact with neighboring image elements;

calculate a histogram based on the integer values of each image element in the morphed image data;

classify the image using the calculated histogram; and

output a classification of the image.

18. A computer program embodied on a non-transitory computer-readable medium in accordance with claim 17, wherein the code segment further configures the processor to:

perform a binary operation on the acquired image data;

apply an averaging filter to the acquired image data; and

assign the integer value to each image element of the acquired data.