Image analysis – Applications
Reexamination Certificate
1999-07-08
2003-12-09
Ahmed, Samir (Department: 2623)
Image analysis
Applications
C382S249000, C713S176000
Reexamination Certificate
active
06661904
ABSTRACT:
TECHNICAL FIELD
The present invention is directed to the field of electronic transmission of data.
BACKGROUND OF THE INVENTION
From the earliest days of commerce, personal presence has had an important role in the conduct of business. Face-to-face dealings have lent an air of credibility to transactions. A handshake has proven an invaluable instrument for establishing trust. Later, formalized documents became acceptable proxies for such face-to-face transactions. The personal signature of a recognized party to the transaction became an acceptable substitute for face-to-face presence. As the information age progressed, faxed signatures often became recognized as the equivalent to an original physical signature.
As technology continues to evolve, there is need to provide equivalent means to the handshake and personal signature for Internet and other electronic transactions. The present invention teaches a method and apparatus for sending personal data from a first computer to a second computer across a network and establishing the equivalent to personal presence.
One embodiment of the present invention relates to the field of fractal transformation. Fractal transformation may be used to compress audio, video, graphical, and other types of data. Throughout this document, fractal graphical image processing will be used to illustrate the techniques taught. Similar art may be practiced on audio, video, and other types of data.
In fractal image processing, an image is divided into a number of library regions. The library regions are identified, or indexed, by their location in the original image. In some prior art, library regions are called ranges. The image is then divided into a number of similarly shaped target regions which, taken together, tile the entire image. In some prior art, target regions are called domains. In one variant, each target region is sequentially compared to each of the indexed library regions to find the one that most closely corresponds to it. In this comparison, the library region is allowed to be rotated, scaled, and offset relative to the target region. Scaling refers to changing the amount of gain or contrast within the library region to make it more closely match the target region. Offsetting refers to changing the overall lightness or darkness.
The most closely matched library region for each target region including such rotation, scaling, and offsetting is noted and stored in memory. The values for rotating, scaling, and offsetting each substitution are called transformation coefficients. After the transformations for all the target regions have been made, the resultant series of geometric relationships between library and target regions and their associated transformation coefficients are saved as a fractal transformation of the base image. The fractal transformation generally has the property of requiring significantly less computer memory to store than the base image. The actual data set is termed a fractal encodation.
Decoding a fractally transformed image is performed as an iterated series of back-substitutions of library region locations from a starting image for target region locations in an intermediate image using the transformation coefficients to adjust the offset, contrast, and rotation of each substituted target region. Following the first iteration of back-substitutions, the entire resultant intermediate image itself is substituted for the starting image. Following subsequent iterations, the resultant intermediate image is similarly substituted for the intermediate image resulting from the previous iteration. In one embodiment, the process is halted once any two successive intermediate images are sufficiently similar to one another. At this point, the decoded image is said to have converged.
One rather amazing property of a fractally encoded image is that it doesn't matter what the starting image for decoding looks like. The iterated geometric relationships and transformation coefficients themselves contain all the information necessary to reconstruct the original base image. The particular starting image chosen simply affects the number of iterations necessary for the decoded image to converge.
Another embodiment of the present invention relates to discrete wavelet transformation. Like fractal transformation, discrete wavelet transformation may be used to compress image, audio, video, and other data. Also like fractal transformation, the discrete wavelet transforms produce as one component a set of coefficients, the values of which are used to derive the image reproduction. One subcategory of discrete wavelet transformation is wavelet scalar transformation. Discrete wavelet transformation is the type of compression used in JPEG2000.
The invention taught herein is equally applicable to other types of data compression technologies and, in particular, image compression technologies that produce coefficients for deriving data reproduction and image reproduction.
SUMMARY OF THE INVENTION
One aspect of the present invention relates to methods of providing a digital signature for remote transactions. Another aspect of the present invention relates to methods of verifying the identity of a source of data. Embodiments of the present invention teach methods and apparatus for establishing the functional equivalent to personal presence, the handshake, and the personal signature in forms appropriate for use across electronic media. Another aspect of the present invention teaches methods for automatically transmitting information relevant to a particular transaction. In particular, the present invention teaches technologies appropriate for use in local area network and Internet transactions.
The present invention makes use of digital graphical bitmaps to establish a visual representation of a sender's identity and authority. Corresponding data streams are used to transmit the sender's identity, authority, and data associated with the sender across a network.
A specifically selected or generated digital graphical bitmap used to establish a visual representation is termed a base image. A base image which is further processed to make it unique is called a personal logo. The data stream representing a personal logo is called a unique graphic personal identifier or UGPI. A UGPI to which personal data has been added is called a data conveyance object or DCO. In general, the base image and personal logo appear virtually identical to one another and the base image may be displayed in place of the personal logo without violating the scope or intent of the present invention. In some applications, it is not necessary or desirable for a base image to be made unique from all other instances of its use. In these applications, the personal logo and the base image may, in fact, be identical. In some embodiments of the present invention, the personal logo is related to the UGPI by a compression algorithm or other transformation, the UGPI being the compacted or transformed data stream representing the personal logo. In other embodiments where neither compaction nor transformation is used, the UGPI may simply be the bitmap of the personal logo in its electronic, non-displayed form. One method for making the base image unique is the integration of personal data. In this case, the UGPI itself may act as the DCO. It should be noted that other processes that occur during data transmission may further compress or otherwise alter the DCO temporarily.
The means for transmitting data taught by the present invention is the embedding of data into a UGPI. It is desirable for the embedding of data to alter the nominal appearance of the personal logo minimally or not at all. Several methodologies may be used for embedding such data including direct substitution or appending of data bits into bitmap pixel data, discrete cosine transformation, discrete wavelet quantization, and fractal transformation.
Another aspect of the present invention relates to methods for controlling the use of data. Along with personal data, restrictions as to the use of said personal data m
Ashby Robert Jason
Cushing John Aikin
Cushing Judith Bayard
Fisher Yuval
Fulwood S. Leigh
Ahmed Samir
Bhatnagar Anand
Davison James L
Personalogo
LandOfFree
Method and system for automated electronic conveyance of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for automated electronic conveyance of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for automated electronic conveyance of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3110057