Novell, Inc.

Edwards E Reed

Stephen R Carter

Radia Joy Perlman

Del Jensen

Tammy G Carter

Methods and systems are provided for identifying and characterizing boundaries in digital images and other digital data sets. At least two points within a digital image are selected. Clusters of sample points around the points are chosen. A pixel value histogram is generated for each sample point cluster. These histograms define frequency functions. The distance between the frequency functions, called a &ldquo;density distance&rdquo; is calculated and analyzed. The nature of the density distance is then used to determine if the two areas are relatively uniform, or if they indicate a presence of a boundary. To further refine the procedure, the method may be applied iteratively by modifying cluster variances, sizes, and distributions, and then reanalyzing the frequency functions generated.

Boundary identification and characterization through density differencing

The present invention is an improved certificate revocation process that improves the efficiency of an authentication exchange in a public key distributed network system. Specifically, the present invention includes a novel revocation service (RS) that, in response to a unique request from a server node, selects certain revoked certificates from a current CRL to include in its reply so as to consume minimal system bandwidth. The unique request includes a number of parameters for consideration by the RS in generating its reply, including a maximum CRL size and/or a timestamp. The maximum CRL size indicates the largest number of revoked certificate serial numbers that the server node can process and thus receive in the revocation service reply, whereas the timestamp indicates the latest certificate revocation date of the certificates included in the CRL presently retained by the server node. Significantly, the RS generates an optimal CRL for its reply that contains all, part, or none of the current CRL revoked certificate serial numbers. Determination of the optimal CRL entails consideration of any number and combination of optimization factors, including the number of revoked certificates stored in the CRL storage facility and the time remaining before the current CRL is to be updated by a certificate authority (CA), the expiration date of the certificates, as well as the maximum CRL size and/or timestamp parameters provided to the RS in the server node request. The server node may control whether it will receive an optimal CRL and if so, what portion of the current CRL it will include by manipulating the parameters it provides to the RS. This enables each server node to request the CRL based upon its own specific security needs while optimizing the certificate revocation process. Further, the RS and/or server node may discard certificate serial numbers as their expiration dates come to pass.

Growing community of inventors

Lindon, UT, United States of America

Edwards E Reed