Method of downsampling documents
AssigneeLucent Technologies Inc.
InventorsOscar Ernesto Agazzi, Kenneth Ward Church, William Arthur Gale
Life of this patent
Estimating the term…
Abstract
A method downsamples characters in a document by font substitution when the characters in the document are identified with sufficient reliability. Otherwise, the characters are downsampled by decimation. The method also allows downsampling by applying characters to a two-dimensional nonlinear filter and decimating.
Claims17
A method of downsampling a document or a portion thereof comprising a set of one or more components, wherein each component in said set of one or more components is in a text representation and has an associated reliability measure, said reliability measure indicating the probability that the associated text representation correctly identifies said each component, the method comprising the steps of:
- downsampling said each component by a first method of downsampling if said reliability measure is above a threshold;
- and
- otherwise downsampling said each component by a second method of downsampling if said reliability measure is below said threshold.
A method comprising the steps of:
- receiving a bit map representation of a component;
- generating a text representation of said component;
- generating a reliability measure indicating the accuracy of said text representation;
- downsampling said component by font substitution if said reliability measure is above a threshold;
- and
- downsampling said component by decimation if said reliability measure is below a threshold.
A method comprising the steps of:
- receiving a bit map representation of a component;
- generating a text representation of said component;
- generating a reliability measure indicating the accuracy of said text representation;
- downsampling said component by font substitution if said reliability measure is above a threshold;
- and
- downsampling said component by filtering and decimating if said reliability measure is below a threshold.
A method of downsampling a document or a portion thereof comprising a set of one or more components, wherein each component in said set of one or more components is in a text representation and has an associated reliability measure, said reliability measure indicating the probability that the associated text representation correctly identifies said each component, the method comprising the steps of:
- downsampling said each component in said document by font substitution if the reliability associated with said each component measure is above a threshold;
- and
- downsampling said each component by decimation if said reliability measure associated with said each component is below said threshold.
An apparatus for downsampling a component in a document, wherein said component is in a text representation and wherein said component has an associated reliability measure, said reliability measure indicating the probability that the associated text representation correctly identifies said component, the method comprising the steps of:
- means for downsampling said component in said document by font substitution if the reliability associated with said component measure is above a threshold;
- and
- means for downsampling said component by decimation if said reliability measure associated with said component is below said threshold.
Dependencies are inferred from the claim text.