SYSTEM AND RELATED TECHNIQUES FOR DETECTING AND CLASSIFYING FEATURES WITHIN DATA

• US 20100287161A1
• Filed: 04/04/2008
• Published: 11/11/2010
• Est. Priority Date: 04/05/2007
• Status: Active Grant

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Naqvi, Waseem
• Granted Patent: US 8,566,314 B2
• Time in Patent Office: Days
• US Class Current: 707/740
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69

Abstract

An information processing system and method for gathering and interpreting information includes capturing information from at least one of a plurality of information streams/sensors wherein the information includes video, audio, seismic, radio frequency (RF), and/or text then applying a standardized tag to an event at a predetermined time or over a predetermined period of time and storing the standardized tag in a repository which can be interrogated rapidly for situation/scene understanding. The information processing system and method include providing a plurality of segmentation algorithms, determining the type of information to be processed and selecting one or more of the segmentation algorithms to process the information based upon the type of information to be processed.

Claims

1. A data processing system comprising:

a segmentation system adapted to receive data, said segmentation system including;

(a) a storage device having a plurality of different feature extraction techniques stored therein;

(b) a feature extraction processor in communication with said storage device, said feature extraction processor adapted to execute at least each of the plurality of different feature extraction techniques stored in said storage device and adapted to extract one or more features from the data provided to said segmentation system; and

(c) a feature extraction technique selector which selects one or more of the plurality of different feature extraction techniques based upon the type of data provided to said segmentation system;

a feature classifier adapted to receive one or more features from said segmentation system and to apply a feature tag to each of the features;

a repository adapted to receive feature tags from said feature classifier;

a search engine coupled to said repository; and

a user interface coupled to said search engine.

2. The system of claim 1 wherein:

said segmentation system is provided as an image segmentation system;

said storage device is provided as a memory;

said plurality of different feature extraction techniques stored I said memory are provided as image feature extraction techniques;

the segmentation processor is provided as an image segmentation processor adapted to execute a plurality of different image feature extraction techniques;

said object classifier is adapted to receive one or more image segments from said image segmentation processor; and

said object repository is provided as a database.

3. The system of claim 1 wherein each feature is a subset of a data segment and wherein said segmentation processor is adapted to combine features to make higher order segments of a picture or audio stream.

4. A system comprising:

an image segmentation system including;

(a) a storage device having a plurality of different feature extraction techniques stored therein;

(b) an image segmentation processor in communication with said storage device, said image segmentation processor adapted to execute at least each of the plurality of different image segmentation algorithms stored in said storage device;

an object classifier adapted to receive one or more image segments from said image segmentation processor and to apply a tag to each object in each of the image segments;

an object repository adapted to receive object tags from said object classifier;

a search engine coupled to said object repository; and

a user interface coupled to said search engine.

5. The system of claim 4 further comprising a tag repository and wherein the tag applied to each object by said object classifier is stored in said tag repository.

6. The system of claim 4 wherein said tag comprises a source ID field, a media type field, a date field, a time field, an object type field and a characteristics field.

7. A method comprising:

capturing information from at least one of a plurality of information streams/sensors wherein the information comprises video, audio, seismic, radio frequency (RF), and/or text.

applying a standardized tag to an event at a predetermined time or over a predetermined period of time;

storing the standardized tag in a repository which can be interrogated rapidly for situation/scene understanding.

8. The method of claim 7 further comprising applying an event tag to the selected portion of the data.

9. The method of claim 8 wherein applying an event tag to the selected portion of the data comprises one of:

(a) manually applying an event tag to the selected portion of the data;

or

(b) automatically applying an event tag to the selected portion of the data.

10. An information processing method comprising:

providing a plurality of segmentation algorithms;

determining the type of information to be processed; and

selecting one or more of the segmentation algorithms to process the information based upon the type of information to be processed.

11. The method of claim 10 further comprising applying an event tag to the selected portion of the data.

12. The method of claim 11 wherein applying an event tag to the selected portion of the data comprises one of:

(a) manually applying an event tag to the selected portion of the data;

or

(b) automatically applying an event tag to the selected portion of the data.

13. A method for processing data comprising:

selecting a portion of the data

identifying one or more specific features within the selected portion of the data;

applying a feature tag to each of the specific features which have been identified; and

storing each of the feature tags in a repository.

14. The method of claim 13 further comprising applying an event tag to the selected portion of the data.

15. The method of claim 14 wherein applying an event tag to the selected portion of the data comprises one of:

(a) manually applying an event tag to the selected portion of the data;

or

(b) automatically applying an event tag to the selected portion of the data.