Image analysis – Applications
Reexamination Certificate
2000-12-13
2003-10-14
Patel, Jayanti K. (Department: 2625)
Image analysis
Applications
C382S100000, C380S054000
Reexamination Certificate
active
06633654
ABSTRACT:
TECHNICAL FIELD
The invention relates to multimedia signal processing, and in particular relates to perceptual modeling of media signals, such as images, video and audio.
BACKGROUND AND SUMMARY
Perceptual modeling is often used in media signal processing applications to assess the extent to which changes to a media signal are noticeable to the human eye or ear. A perceptual model analyzes a media signal's ability to hide or mask changes. Such models are used in lossy signal compression and digital watermarking to minimize the perceptibility of these processes to a viewer or listener of the processed signal.
Lossy signal compression typically quantizes signal samples to reduce the memory or bandwidth required to store and transmit image and audio signals. Media signal codecs, like those defined in MPEG standards, use perceptual models to identify parts of media signal that can be more heavily compressed while staying within a desired quality.
Digital watermarking is a process for modifying media content to embed a machine-readable code into the data content. The data may be modified such that the embedded code is imperceptible or nearly imperceptible to the user, yet may be detected through an automated detection process. Most commonly, digital watermarking is applied to media such as images, audio signals, and video signals. However, it may also be applied to other types of data, including documents (e.g., through line, word or character shifting), software, multi-dimensional graphics models, and surface textures of objects.
Digital watermarking systems have two primary components: an embedding component that embeds the watermark in the media content, and a reading component that detects and reads the embedded watermark. The embedding component embeds a watermark signal by altering data samples of the media content. The reading component analyzes content to detect whether a watermark is present. In applications where the watermark encodes information, the reader extracts this information from the detected watermark.
In digital watermarking, one aim is to insert the maximum possible watermark signal without significantly affecting signal quality. Perceptual models may be used to determine how to embed the watermark in a host media signal such that signal masks the watermark. In image watermarking, a watermark embedder can take advantage of the masking effect of the eye to increase the signal strength of a watermark in busy or high contrast image areas. However if this is done for all high frequency areas, a visually objectionable watermark or ‘ringing’ may become visible on connected directional edges.
In audio watermarking, busy or high contrast segments of an audio signal tend to have a greater masking effect. However, embedding a watermark in portions of an audio signal that represent pure tones may make the watermark more audible.
The invention provides methods for perceptual analysis of media signals. While particularly adapted to image signals, the invention applies to other types of media signals as well. One aspect of the invention is a method for perceptually analyzing a media signal to reduce perceptible artifacts around directional edges. The method analyzes the media signal to compute a measure of directional edges. Based at least in part on the measure of directional edges, the method computes control data used to control changes to the media signal in a manner that controls perceptibility of the changes around directional edges.
For digital watermark applications, this method may be used to reduce perceptible artifacts around connected edges. The method may also be used to reduce artifacts around directional edges in lossy signal compression schemes.
Another aspect of the invention is a method for perceptual analysis of a media signal based on local contrast. This method analyzes the media signal to compute measures of local contrast at samples within the media signal. Based at least in part on the measures of local contrast, it computes a measure of visual sensitivity to changes of the media signal at the samples. To compute visual sensitivity to the local contrast, it applies a human visual model that relates local contrast to visual sensitivity.
In one implementation, the human visual model performs a non-linear mapping function that is tuned to the eye's sensitivity to local contrast. In a plot of visual sensitivity versus contrast, visual sensitivity initially increases with contrast and then decreases. The mapping function exploits this attribute.
As in the case of the perceptual analysis based on directional edges, the perceptual analysis based on local contrast may be applied to a variety of media signal processing applications. Some examples include digital watermarking and lossy signal compression.
The perceptual analyses based on directional edges and local contrast may be used independently, in combination with each other, or in combination with other perceptual models.
Further features will become apparent with reference to the following detailed description and accompanying drawings.
REFERENCES:
patent: 5568570 (1996-10-01), Rabbani
patent: 5636292 (1997-06-01), Rhoads
patent: 5848155 (1998-12-01), Cox
patent: 5859920 (1999-01-01), Daly et al.
patent: 5862260 (1999-01-01), Rhoads
patent: 5905819 (1999-05-01), Daly
patent: 5915027 (1999-06-01), Cox et al.
patent: 6061793 (2000-05-01), Tewfik et al.
patent: 6181802 (2001-01-01), Todd
patent: 6185312 (2001-02-01), Nakamura et al.
patent: 6222932 (2001-04-01), Rao et al.
patent: 6226387 (2001-05-01), Tewfik et al.
patent: 6233347 (2001-05-01), Chen et al.
U.S. patent application Ser. No. 60/286,239, Kim, filed Apr. 24, 2001.
U.S. patent application Ser. No. 09/302,663, Gustafson et al., filed Apr. 30, 1999.
U.S. patent application Ser. No. 09/503,881, Rhoads et al., filed Feb. 14, 2000.
U.S. patent application Ser. No. 09/596,658, Hannigan et al., filed Jun. 19, 2000.
U.S. patent application Ser. No. 09/940,872, Stach, filed Aug. 27, 2001.
U.S. patent application Ser. No. 09/945,244, Sharma et al., filed Aug. 31, 2001.
U.S. patent application Ser. No. 09/967,913, Kim et al., filed Sep. 28, 2001.
Autrusseau et al., “A Psychovisual Approach for Watermarking”, Proceedings of SPIE vol. 4314, 2001, pp. 495-504.
Bartens, “Contrast Sensitivity of the Human Eye and It's Effects on Image Quality”, pp. 147-151, SPIE Press, 1999.
Reviews of John Canny's “A Computational Approach to Edge Detection”,IEEE Trans. Pattern Analyses and Machine Intelligence, pp. 679-698, 1986.
Cox, et al., “A Review of Watermarking and the Importance of Perceptual Modeling”, copy to appear in the Proc. Of Electronic Imaging '97, Feb. 1997, 8 pages.
Delaigle, et al., “Watermarking Algorithm Based on a Human Visual Model”,Signal Processing, Oct. 1998, 23 pages.
De Rosa, et al., “Watermark Capacity Measure Incorporating a Model of the Human Visual System”, Proceedings of SPIE vol. 4314, 2001, pp. 483-494.
Abstract of Echizen, et al., “Maintaining Image Quality in Digital Watermarking Based on Contour Preserving”, as published in Transactions of the Information Processing Society of Japan, vol. 41, No. 6, pp. 1828-1839.
Hernandez, et al., “Statistical Analysis of Watermarking Schemes for Copyright Protection of Images”, Proceedings of the IEEE, vol. 87, No. 7, Jul. 1999, pp. 1142-1166.
Kovesi, “Lecture 6, Classical Feature Detection”, The University of Western Australia, 2000, 12 pages.
Lim, “Two-Dimensional Signal and Image Processing”, PTR Prentice Hall, New Jersey, 1990, pp. 476-495.
Podilchuk et al., “Image-Adaptive Watermarking Using Visual Models”, 1998, 37 pages.
Pratt,Digital Image Processing, John Wiley & Sons, New York, 1978, pp. 478-492.
Robert et al., “Masking Models and Watermark unDetection”, Proceeding of SPIE, vol. 4314, 2001, pp. 455-467.
Sharma et al., “Practical Challenges For Digital Watermarking Applications”, May, 2001, 10 pages.
Yong Ju Jung, et al., “Novel Watermark Embedding Technique Based on Human Visual system”, Proceeding of SPIE vol. 4314, 2001, pp. 475-482.
Bradley Brett Alan
Hannigan Brett T.
Reed Alastair M.
Digimarc Corporation
Digimarc Corporation
Meyer Joel R.
Patel Jayanti K.
Tabatabai Abolfazl
LandOfFree
Perceptual modeling of media signals based on local contrast... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Perceptual modeling of media signals based on local contrast..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Perceptual modeling of media signals based on local contrast... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3131465