System and methods thereof for generation of searchable structures respective of multimedia data content

• US 8,868,619 B2
• Filed: 09/04/2012
• Issued: 10/21/2014
• Est. Priority Date: 10/26/2005
• Status: Active Grant

• Current Assignee: Cortica Ltd.
• Original Assignee: Cortica Ltd.
• Inventors: Raichelgauz, Igal, Ordinaev, Karina, Zeevi, Yehoshua Y.
• Primary Examiner(s): GIRMA, ANTENEH B
• Application Number:
  US13/602,858
• Publication Number:
  US 20120331011A1
• Time in Patent Office: 777 Days
• Field of Search: None
• US Class Current: 707/803
• CPC Class Codes: G06F 16/14 Details of searching files based on file metadata
• CPC Class Codes: G06F 16/152 using file content signatures, e.g. hash values
• CPC Class Codes: G06F 16/41 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99948 Application of database or data structure, e.g. distributed, multimedia, or image

Abstract

A method for creating a multimedia data search engine platform to allow fast search of multimedia content data elements (MMDEs). The method comprises collecting MMDEs from at least an external source storing MMDEs; generating a plurality of signatures for each of the collected MMDEs; generating signature reduced clusters (SRCs) for the collected MMDEs by clustering the plurality of signatures generated for each of the collected MMDEs; and generating concept structures from the generated SRCs, wherein the concept structures generated for different SRCs are utilized to compare between different MMDEs, thereby searching for an input MMDE that matches the collected MMDEs.

Claims

1. A method for generating concept structures from signature reduced clusters (SRCs), comprising:

retrieving from a storage at least one SRC including a cluster of signatures respective of a plurality of multimedia data elements (MMDEs);

generating metadata for the cluster of signatures included in the at least one retrieved SRC;

associating the generated metadata with the at least one retrieved SRC;

reducing a number of signatures in the at least one retrieved SRC to include only signatures that uniquely identify the at least one retrieved SRC by temporarily removing a signature from the at least one retrieved SRC, checking if MMDEs associated with the at least one retrieved SRC can be recognized without the removed signature, and permanently removing the signature from the at least one retrieved SRC, if the MMDEs associated with the SRC can be recognized; and

matching the at least one retrieved SRC with a plurality of previously generated SRCs stored in the storage, wherein each of the plurality of previously generated SRCs that matches the at least one retrieved SRC forms a concept structure, wherein the concept structure includes at least metadata associated with the SRC matching the at least one retrieved SRC.

2. The method of claim 1, wherein associating the generated metadata with the at least one retrieved SRC further comprising:

associating the at least one retrieved SRC with a piece of metadata which is a combination of pieces of metadata with each of the signatures included in the at least one retrieved SRC.

3. The method of claim 2, wherein the combination of the pieces of metadata does not include repetition of metadata pieces.

4. The method of claim 2, wherein a match between the at least one retrieved SRC and a SRC from the plurality of previously generated SRCs is determined based on a first matching threshold, wherein the first matching threshold is defined by a percentage of overlap between the metadata associated with the at least one retrieved SRC and metadata associated with the SRC from the plurality of previously generated SRCs.

5. The method of claim 1, wherein a match between the at least one retrieved SRC and a SRC from the plurality of previously generated SRCs is determined based on a second matching threshold, wherein the second matching threshold is defined by a number of matches between signatures in the at least one retrieved SRC and signatures in the SRC from the plurality of previously generated SRCs.

6. The method of claim 1, wherein concept structures generated for different SRCs are utilized to compare between different MMDEs.

7. The method of claim 1, further comprising:

storing in a storage unit at least one of the generated concept structures.

8. The method of claim 1, wherein the MMDEs include at least one of:

an image, graphics, a video stream, a video clip, an audio stream, an audio clip, a video frame, a photograph, and images of signals.

9. The method of claim 8, wherein the images of signals include images of at least one of:

medical signals, geophysical signals, subsonic signals, supersonic signals, electromagnetic signals, and infrared signals.

10. The method of claim 1, wherein the signatures are generated by a signature generator, wherein the signature generator comprises a plurality of computational cores, each computational core having properties statistically independent of each other of the computational cores, each computational core generates, responsive to a received MMDE at least a signature comprising a first signature element and a second signature element, the first signature element being a robust signature.

11. A non-transitory computer readable medium having stored thereon instructions for causing a processor to perform a method of claim 1.

12. A method for creating a multimedia data search engine platform to allow fast search of multimedia content data elements (MMDEs), comprising:

collecting MMDEs from at least an external source storing MMDEs;

generating a plurality of signatures for each of the collected MMDEs;

generating signature reduced clusters (SRCs) for the collected MMDEs by clustering the plurality of signatures generated for each of the collected MMDEs, wherein clustering the plurality of signatures generated further includes;

generating a clustering score for each signature of an MMDE of the collected MMDEs versus all other MMDEs of the collected MMDEs, determining a size of a diagonal matrix having a size corresponding to a number of the collected MMDEs, placing the clustering score in a diagonal matrix in storage, one clustering score for each pair of MMDEs, creating a new cluster element for each two cluster elements in the diagonal matrix having a clustering score that exceeds a threshold, and repeating the process at the newly created cluster element level each time using the new cluster elements generated as the plurality of cluster elements for the subsequent sequence, until a single cluster is reached or until it is determined that a single cluster cannot be reached; and

generating concept structures from the generated SRCs, wherein the concept structures generated for different SRCs are utilized to compare between different MMDEs, thereby searching for an input MMDE that matches the collected MMDEs.

13. The method of claim 12, further comprising:

storing the generated concept structures, the collected MMDEs, the generated signatures, and the generated SRCs in a storage unit.

14. The method of claim 12, wherein the generating the concept structures from the SRCs, further comprising:

for each SRC performing;

generating metadata for the cluster of signatures included in the SRC;

associating the generated metadata with the SRC;

matching the SRC with a plurality of stored SRCs, wherein each of the plurality of stored SRCs that matches the SRC forms a concept structure, wherein the concept structure includes at least metadata associated with each SRC that is found to be a match.

15. The method of claim 14, wherein generating the plurality of signatures for each of the collected MMDEs further comprising:

generating a plurality of items from each of the collected MMDEs;

determining which of the plurality of the items are of interest for signature generation; and

generating at least a signature responsive to at least one item of interest.

16. The method of claim 14, wherein the signatures are generated by a signature generator, wherein the signature generator comprises a plurality of computational cores, each computational core having properties statistically independent of each other of the computational cores, each computational core generates, responsive to a received MMDE at least a signature comprising a first signature element and a second signature element, the first signature element being a robust signature.

17. A non-transitory computer readable medium having stored thereon instructions for causing a processor to perform a method of claim 12.