公开(公告)号：US20140111701A1

公开(公告)日：2014-04-24

申请号：US14054438

申请日：2013-10-15

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Regunathan Radhakrishnan

IPC: G10L19/018 ,  H04N5/60 ,  H04N21/439

CPC classification number: G10L19/018 ,  H04N5/602 ,  H04N21/233 ,  H04N21/4394 ,  H04N21/8358

Abstract: An audio data spread spectrum embedding and detection method is presented. For each audio frame, a noise sequence is chosen according to the data to be embedded. Then, a spectrum of a chosen noise sequence is shaped by a spectrum of a current audio frame and subtracted from a current frame's spectrum. During detection, a detector is used on a watermarked audio frame to first whiten the watermarked audio frame. Detection scores are then computed against two competing Adaboost learning models. A detected bit is chosen according to the model with a maximum detection score.

Abstract translation: 提出了一种音频数据扩频嵌入和检测方法。 对于每个音频帧，根据要嵌入的数据选择噪声序列。 然后，所选择的噪声序列的频谱由当前音频帧的频谱整形并从当前帧的频谱中减去。 在检测期间，在加水印的音频帧上使用检测器来首先使加水印的音频帧变白。 然后针对两个竞争的Adaboost学习模型计算检测分数。 根据具有最大检测分数的型号选择检测到的位。

公开(公告)号：US09742554B2

公开(公告)日：2017-08-22

申请号：US14765563

申请日：2014-01-28

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Regunathan Radhakrishnan

IPC: H04L7/04 ,  G10L19/018

CPC classification number: H04L7/042 ,  G10L19/018

Abstract: Systems and methods for detecting a synchronization code word embedded in a plurality of frames of a signal are described. In one example embodiment, the synchronization code word contains “s” bits, embedded one bit per frame in “s” frames of an input signal. The method of detecting this synchronization code word includes: initiating a first segmentation procedure wherein “n” segments are defined in each signal frame of the input signal. A first correlation threshold value, which is based on the synchronization code word, is used to identify in the “n” segments, a first segment having the highest likelihood of containing at least a portion of the synchronization code word. The first segment is used to initiate a recursive detection procedure incorporating one or more additional segmentation procedures and one or more additional correlation threshold values, to detect the synchronization code word in a sub-divided portion of the first segment.

公开(公告)号：US09564139B2

公开(公告)日：2017-02-07

申请号：US14985047

申请日：2015-12-30

Applicant: DOLBY LABORATORIES LICENSING CORPORATION

Inventor： Regunathan Radhakrishnan ,  Michael Smithers ,  David S. McGrath

IPC: G06F17/00 ,  G10L19/018

CPC classification number: G10L19/018

Abstract: A spread spectrum data hiding for audio signals is described. A set of pseudo-random noise sequences is added to an audio signal according to a data to be embedded. A masking curve is used to shape the added noise. A transient detection step can be used to control whether a shaped noise sequence is to be added or not. Embedded information is detected by first performing a whitening step and then performing a phase-only correlation with a same set of pseudo-random noise sequences. A detection method that is based on correlation of multiplexed noise sequences with a noise sequence embedded in the audio is also described.

Abstract translation: 描述用于音频信号的扩频数据隐藏。 根据要嵌入的数据，将一组伪随机噪声序列添加到音频信号。 使用掩蔽曲线来调整添加的噪声。 可以使用瞬态检测步骤来控制是否添加成形噪声序列。 通过首先执行白化步骤然后对相同的伪随机噪声序列组执行仅相位相关性来检测嵌入信息。 还描述了基于多路复用噪声序列与嵌入在音频中的噪声序列的相关性的检测方法。

公开(公告)号：US10349125B2

公开(公告)日：2019-07-09

申请号：US15466685

申请日：2017-03-22

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Michael Grant ,  Regunathan Radhakrishnan ,  Jeff Edwin Chandler ,  Stewart Murrie ,  Michael G. Babbitt

IPC: H04N21/44 ,  G06F16/40 ,  G06F16/43 ,  H04N21/81 ,  G06F16/783 ,  H04N21/234 ,  H04N21/439 ,  H04N21/442 ,  G06F16/9535

Abstract: A media fingerprint archive system generates and archives media fingerprints from second media content portions such as commercials. A downstream media measurement system can extract/derive query fingerprints from an incoming signal, and query the media fingerprint archive system whether any of the query fingerprints matches any archived fingerprints. If so, the media measurement system can perform media measurements on a specific secondary media content portion from which the matched query fingerprint is derived. If not, the media measurement system can analyze media characteristics of a media content portion to determine whether the media content portion is a secondary media content portion and perform media measurement if needed to. The media measurement system may send fingerprints from an identified secondary media content portion to the media fingerprint archive system for storage.

公开(公告)号：US10296534B2

公开(公告)日：2019-05-21

申请号：US14730185

申请日：2015-06-03

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Tianshi Gao ,  Regunathan Radhakrishnan ,  Wenyu Jiang ,  Claus Bauer

IPC: G06F17/30 ,  G06F16/51 ,  G06F16/903 ,  G06K9/00 ,  G06K9/52 ,  G06K9/62

Abstract: Attributes are identified in media content. A classification value of the media content is computed based on the identified attributes. Thereafter, a fingerprint derived from the media content is stored or searched for based on the classification value of the media content.

公开(公告)号：US20140336800A1

公开(公告)日：2014-11-13

申请号：US14117576

申请日：2012-05-15

Applicant: Dolby Laboratories Licensing Corporation

Inventor： Regunathan Radhakrishnan ,  Sevinc Bayram ,  Jeffrey Riedmiller

IPC: G10L19/018 ,  H04S5/00

CPC classification number: G10L19/018 ,  G10L19/008 ,  G10L25/06 ,  H04S3/02 ,  H04S5/005 ,  H04S2400/01

Abstract: A media signal is accessed, which has been generated with one or more first processing operations. The media signal includes one or more sets of artifacts, which respectively result from the one or more processing operations. One or more features are extracted from the accessed media signal. The extracted features each respectively correspond to the one or more artifact sets. Based on the extracted features, a conditional probability score and/or a heuristically based score is computed, which relates to the one or more first processing operations.

Abstract translation: 访问已经通过一个或多个第一处理操作生成的媒体信号。 媒体信号包括分别由一个或多个处理操作产生的一组或多组伪影。 从所访问的媒体信号中提取一个或多个特征。 所提取的特征分别对应于一个或多个伪像集。 基于提取的特征，计算与一个或多个第一处理操作有关的条件概率得分和/或启发式分数。

公开(公告)号：US20140330556A1

公开(公告)日：2014-11-06

申请号：US14360257

申请日：2012-12-10

Applicant: DOLBY LABORATORIES LICENSING CORPORATION ,  DOLBY INTERNATIONAL AB

Inventor： Barbara Resch ,  Regunathan Radhakrishnan ,  Arijit Biswas ,  Jonas Engdegard

IPC: G10L19/00

CPC classification number: G10L19/00 ,  G10H1/0008

Abstract: Low complexity detection of a time-wise position of a representative segment in media data is described. A subset of offset values is located in a set of offset values in media data using a first type of one or more types of features, which are extractable from (e.g., derivable from components of) the media data. The subset of offset values comprise values that are selected from the set of offset values based on one or more selection criteria. A set of candidate seed time points is identified based on the subset of offset values using a second type of the one or more types of features.

Abstract translation: 描述媒体数据中代表性段的时间位置的低复杂度检测。 偏移值的子集位于媒体数据的一组偏移值中，使用可从媒体数据的（例如，可从组件导出）的一个或多个类型的特征的第一类型。 偏移值的子集包括从基于一个或多个选择标准的偏移值集合中选择的值。 使用第二类型的一种或多种类型的特征，基于偏移值的子集来识别一组候选种子时间点。

公开(公告)号：US20140088978A1

公开(公告)日：2014-03-27

申请号：US14116113

申请日：2012-04-30

Applicant: DOLBY INTERNATIONAL AB ,  DOLBY LABORATORIES LICENSING CORPORATION

Inventor： Harald H. Mundt ,  Arijit Biswas ,  Regunathan Radhakrishnan

IPC: G10L19/00

CPC classification number: G10L19/00 ,  G10L19/008 ,  G10L21/02 ,  G10L21/038 ,  G10L25/03

Abstract: The present document relates to audio forensics, notably the blind detection of traces of parametric audio encoding/decoding. In particular, the present document relates to the detection of parametric frequency extension audio coding, such as spectral band replication (SBR) or spectral extension (SPX), from uncompressed waveforms such as PCM (pulse code modulation) encoded waveforms. A method for detecting frequency extension coding history in a time domain audio signal is described. The method may comprise transforming the time domain audio signal into a frequency domain, thereby generating a plurality of subband signals in a corresponding plurality of subbands comprising low and high frequency subbands; determining a degree of relationship between subband signals in the low frequency subbands and subband signals in the high frequency subbands; wherein the degree of relationship is determined based on the plurality of subband signals; and determining frequency extension coding history if the degree of relationship is greater than a relationship threshold.

Abstract translation: 本文件涉及音频取证，特别是盲目检测参数音频编码/解码的痕迹。 特别地，本文件涉及从诸如PCM（脉冲编码调制）编码波形的未压缩波形检测参数频率扩展音频编码，例如频谱带复制（SBR）或频谱扩展（SPX）。 描述了用于检测时域音频信号中的频率扩展编码历史的方法。 该方法可以包括将时域音频信号变换成频域，由此在包括低频和高频子带的相应多个子带中产生多个子带信号; 确定低频子带中的子带信号与高频子带中的子带信号之间的关系程度; 其中所述关系度基于所述多个子带信号来确定; 以及如果所述关系度大于关系阈值，则确定频率扩展编码历史。