-
公开(公告)号：CA2307264A1
公开(公告)日：2000-12-09
申请号：CA2307264
申请日：2000-05-01
Applicant: IBM
Inventor： RAVI TUMKUR V RAO , KULSHRESHTHA ASHUTOSH , GHOSAL SUQATA , AGGARWAL GAURAV , DUBEY PRADEEP K
Abstract: A methodology of highly interactive intra-object relevance feedback is used to retrieve images from a database. The query is adjusted using the information fed-back by the user about the relevance of previously extracted part(s) from the object itself, such that the adjusted query is a better approximation to the user's perception. Since a single query image is utilized in the system according to the invention, high-performance learning techniques can be employed for this intra-object learning of user's perception. The refined query is subsequently used for intra-object relevance feedback where images are retrieved from the database based on parameters learnt by intra-image feedback mechanism, and the user provides feedback by ranking the retrieved objects in order of their relevance to him or her. In the system according to the invention, intraobject learning of user's perception is expedited by utilizing the learnt parameters in the intra-object relevance feedback. Furthermore, the methodology of the invention allows for building refined queries based on part(s) of the query object rather than the entire object itself, thereby reducing the number of irrelevant objects, retrieved from the database.
-
公开(公告)号：GB2362240B
公开(公告)日：2004-04-28
申请号：GB0022615
申请日：2000-09-15
Applicant: IBM
Inventor： AGGARWAL GAURAV , DUBEY PRADEEP , KULSHRESHTHA ASHUTOSH , MARTENS MARCO , TRESSER CHARLES PHILIPPE LOUIS , WU CHAI WAH
Abstract: A method is presented for marking high-quality digital images with a robust and invisible watermark. It requires the mark to survive and remain detectable and authenticatable through all image manipulations that in themselves do not damage the image beyond useability. These manipulations include JPEG "lossy" compression and, in the extreme, the printing and rescanning of the image. The watermark also has the property that it can detect if the essential contents of the image has changed. The first phase of the method comprises extracting a digest or number N from the image so that N only (or mostly) depends on the essential information content, such that the same number N can be obtained from a scan of a high quality print of the image, from the compressed form of the image, or in general, from the image after minor modifications (introduced inadvertently by processing, noise etc.). The second phase comprises the marking. This can be done in form of an invisible robust watermark, or in form of some visible signature or watermark.
-
公开(公告)号：CA2307264C
公开(公告)日：2007-12-04
申请号：CA2307264
申请日：2000-05-01
Applicant: IBM
Inventor： AGGARWAL GAURAV , DUBEY PRADEEP K , RAVI TUMKUR V RAO , KULSHRESHTHA ASHUTOSH , GHOSAL SUQATA
Abstract: A methodology of highly interactive intra-object relevance feedback is used to retrieve images from a database. The query is adjusted using the information fed-back by the user about the relevance of previously extracted part(s) from the object itself, such that the adjusted query is a better approximation to the user's perception. Since a single query image is utiliz ed in the system according to the invention, high-performance learning techniques can be employed for this intra-object learning of user's perception. The refined query is subsequently used for intra-objec t relevance feedback where images are retrieved from the database based on parameters learnt by intra-image feedback mechanism, and the user provides feedback by ranking the retrieved objects i n order of their relevance to him or her. In the system according to the invention, intra- object learning of user's perception is expedited by utilizing the learnt parameters in the intra-obje ct relevance feedback. Furthermore, the methodology of the invention allows for building refined queries based on part(s) of the query object rather than the entire object itself, thereby reducing t he number of irrelevant objects, retrieved from the database.
-
公开(公告)号：GB2362240A
公开(公告)日：2001-11-14
申请号：GB0022615
申请日：2000-09-15
Applicant: IBM
Inventor： AGGARWAL GAURAV , DUBEY PRADEEP , KULSHRESHTHA ASHUTOSH , MARTENS MARCO , TRESSER CHARLES PHILIPPE LOUIS , WU CHAI WAH
Abstract: A watermarking method for an image wherein a number N is embedded into an image as a invisible or visible robust watermark. The number N is generated from the essential components of the image such that the number N will be the same when generated from analysis of the image after processing (such as compression or printing) or noise introduces minor discrepancies or modifications which do not effect the images authenticity. The number N can be generated by performing a principal component analysis on a sample set of images with minor discrepancies or by using Wold's decomposition. Alternatively the image can be divided into n x n blocks and then transform coefficients are generated for each block, these are then quantized, the result is encrypted using a private key and this is then hashed using an algorithm to generate the number N. The authenticity of an image can be verified by generating the number N again when verifying and then comparing this to the number N embedded in the image, if the two do not match then it is considered that the image has been altered deliberately.
-
-
-
Search Results
4
records
(took 0.015 seconds)
