Content-recognition facilitator

• US 7,152,163 B2
• Filed: 11/04/2004
• Issued: 12/19/2006
• Est. Priority Date: 04/24/2001
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Venkatesan, Ramarathnam, Mihcak, M. Kivanc
• Primary Examiner(s): Barron, Jr., Gilberto
• Assistant Examiner(s): Dinh, Minh
• Application Number:
  US10/980,918
• Publication Number:
  US 20050066177A1
• Time in Patent Office: 775 Days
• Field of Search: None
• US Class Current: 713/180
• CPC Class Codes: G06F 16/433 using audio data
• CPC Class Codes: G06F 16/634 Query by example, e.g. query by humming
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: G10H 1/0058 Transmission between separate instruments or between individual components of a musical system G10H1/0083 takes precedence

Abstract

Described herein is a technology for facilitating the recognition of the content of digital signals. This abstract itself is not intended to limit the scope of this patent. The scope of the present invention is pointed out in the appending claims.

Claims

1. A computer-implemented hashing method, comprising:

transforming an original digital signal into a digital signal transform;

first computing a hash value representative of the original digital signal such that perceptually distinct digital signals result in hash values that are approximately independent of one another and perceptually similar digital signals result in proximal first hash values, wherein the computing of the first hash value comprises;

obtaining the digital signal transform of the original digital signal;

quantizing the digital signal transform;

geometric-region-growing the digital signal transform;

based upon the geometrical-region-grown digital signal transform, generating the first hash value of the original digital signal;

second computing a second hash value in representative of the original digital signal such that perceptually distinct digital signals result in second hash values that are approximately independent of one another and perceptually same digital signals result in identical second hash values, wherein the computing of the second hash value comprises;

obtaining the digital signal transform of the original digital signal;

randomly dividing the digital signal transform into multiple chunks, each chunk containing signal data, wherein the dividing is carried out recursively to form hierarchical levels of overlapping chunks;

averaging, for each of the chunks, the signal data to produce corresponding chunk averages;

generating, based in part on the chunk averages, an exponential distribution having multiple distinct quantization levels;

randomly rounding each of the chunk averages to one of the quantization levels to produce rounded values;

hashing a composite of the rounded values;

storing the first and second hash values in relationship with the original digital signal;

watermarking the original digital signal using, in part, the hash value to produce a watermarked original digital signal;

subsequently distributing the watermarked original digital signal over a network;

collecting a subject digital signal from a remote site on the network;

in accordance with the first computing act, computing a hash value of the subject digital signal collected from the remote site;

comparing the hash value of the collected subject digital signal with the stored hash value of the original digital signal;

in accordance with the second computing act, computing a second hash value of the subject digital signal collected from the remote site;

comparing the second hash value of the collected subject digital signal with the stored second hash value of the original digital signal; and

identifying the collected subject digital signal as a pirated version of the original digital signal if the second hash values match.

2. A computer-implemented method as recited in claim 1, wherein the original and subject digital signals are selected from a group consisting of digital audio signals, digital image signals, digital video signals, and digital multimedia signals.

3. A computer-implemented method as recited in claim 1, wherein the original and subject digital signals are digital audio signals.

4. One or more computer storage media having computer-executable instructions stored thereon that, when executed by a computer, perform acts comprising:

transforming an original digital signal into a digital signal transform;

first computing a hash value representative of the original digital signal such that perceptually distinct digital signals produce hash values that are approximately independent of one another and perceptually similar digital signals result in proximal first hash values, wherein the computing of the first hash value comprises;

obtaining the digital signal transform of the original digital signal;

quantizing the digital signal transform;

geometric-region-growing the digital signal transform;

based upon the geometrical-region-grown digital signal transform, generating the first hash value of the original digital signal;

second computing a second hash value representative of the original digital signal such that perceptually distinct digital signals result in second hash values that are approximately independent of one another and perceptually same digital signals result in identical second hash values, wherein the computing of the second hash value comprises;

obtaining the digital signal transform of the original digital signal;

randomly dividing the digital signal transform into multiple chunks, each chunk containing signal data, wherein the dividing is carried out recursively to form hierarchical levels of overlapping chunks;

averaging, for each of the chunks, the signal data to produce corresponding chunk averages;

generating, based in part on the chunk averages, an exponential distribution having multiple distinct quantization levels;

randomly rounding each of the chunk averages to one of the quantization levels to produce rounded values;

hashing a composite of the rounded values;

storing the first and second hash values and an indication of an association with the original digital signal;

watermarking the original digital signal using, in part, the second hash value to produce a watermarked original digital signal;

subsequently distributing the watermarked original digital signal over a network;

collecting a subject digital signal from a remote site on the network;

in accordance with the first computing act, third computing a first hash value of the subject digital signal collected from the remote site;

comparing the first hash value of the collected subject digital signal with the stored first hash value of the original digital signal;

in accordance with the second computing act, computing a second hash value of the subject digital signal collected from the remote site;

comparing the second hash value of the collected subject digital signal with the stored second hash value of the original digital signal;

identifying the subject digital signal as a pirated version of the original digital signal if the stored hash value of the original digital signal and second hash value of the subject digital signal match.

5. One or more computer storage media as recited in claim 4, wherein the original and subject digital signals are selected from a group consisting of digital audio signals, digital image signals, digital video signals, and digital multimedia signals.

6. A computer-implemented hashing method, comprising:

transforming an original digital signal into a digital signal transform;

first computing a first hash value representative of the original digital signal such that perceptually distinct digital signals result in first hash values that are approximately independent of one another and perceptually similar digital signals result in proximal first hash values, wherein the computing of the first hash value comprises;

obtaining the digital signal transform of the original digital signal;

quantizing the digital signal transform;

geometric-region-growing the digital signal transform;

based upon the geometrical-region-grown digital signal transform, generating the first hash value of the original digital signal;

second computing a second hash value in representative of the original digital signal such that computations of perceptually distinct digital signals produce second hash values that are approximately independent of one another and computations of perceptually same digital signals produce identical second hash values, wherein the computing of the second hash value comprises;

obtaining the digital signal transform of the original digital signal;

randomly dividing the digital signal transform into multiple chunks, each chunk containing signal data, wherein the dividing is carried out recursively to form hierarchical levels of overlapping chunks;

averaging, for each of the chunks, the signal data to produce corresponding chunk averages;

generating, based in part on the chunk averages, an exponential distribution having multiple distinct quantization levels;

randomly rounding each of the chunk averages to one of the quantization levels to produce rounded values;

hashing a composite of the rounded values;

storing the first and second hash values in relationship with the original digital signal;

watermarking the original digital signal using, in part, the second hash value to produce a watermarked original digital signal.

7. A computer-implemented method as recited in claim 6 further comprising subsequently distributing the watermarked original digital signal over a network.

8. A computer-implemented method as recited in claim 6, wherein the original and subject are selected from a group consisting of digital audio signals, digital image signals, digital video signals, and digital multimedia signals.