Abstract: Machine translation algorithms for translating between a first language and a second language are often trained using parallel fragments, comprising a first language corpus and a second language corpus comprising an element-for-element translation of the first language corpus. Such training may involve large training sets that may be extracted from large bodies of similar sources, such as databases of news articles written in the first and second languages describing similar events; however, extracted fragments may be comparatively “noisy,” with extra elements inserted in each corpus. Extraction techniques may be devised that can differentiate between “bilingual” elements represented in both corpora and “monolingual” elements represented in only one corpus, and for extracting cleaner parallel fragments of bilingual elements.

Abstract: Machine translation algorithms for translating between a first language and a second language are often trained using parallel fragments, comprising a first language corpus and a second language corpus comprising an element-for-element translation of the first language corpus. Such training may involve large training sets that may be extracted from large bodies of similar sources, such as databases of news articles written in the first and second languages describing similar events; however, extracted fragments may be comparatively “noisy,” with extra elements inserted in each corpus. Extraction techniques may be devised that can differentiate between “bilingual” elements represented in both corpora and “monolingual” elements represented in only one corpus, and for extracting cleaner parallel fragments of bilingual elements.

Abstract: A method of matching a query fingerprint to a plurality of file fingerprints. The method comprises the steps of determining a plurality of partial features of each of the file fingerprints. For each partial feature, derive a list of all file fingerprints which have said partial feature as one of their partial features. Determine a plurality of query partial features of the query fingerprint, and derive a ranked list of the file fingerprints based on identifying the individual query partial features in the partial features of the respective file fingerprints. Then perform one-to-one matching of the query fingerprint with selected ones of the ranked list of the file fingerprints.

Abstract: In an automatic fingerprint authentication or identification system, the fingerprint image acquisition is severely effected by the limitations of the acquisition process. The two modes of input, viz. scanning inked fingerprints from paper records or directly from a finger using live-scan fingerprint scanners suffer from the following noise sources in the input in addition to standard noise in the camera. Non-uniform ink application, uneven pressure while rolling on the paper or pressing on the scanner surface and external dirt like oil and climatic variations in the moisture content of skin are some of the main causes for the ridges and valleys not to be imaged clearly. This invention deals with a method of learning a set of partitioned least-sqaures filters that can be derived from a given set of images and ground truth pairs as an offline process. The learned filters are convolved with input fingerprint images to obtain the enhanced image.

Abstract: Fingerprint verification matching a claimant fingerprint with one of a number of recorded enrollee fingerprints uses various adaptations of model alignment techniques involving affine transformation to provide an algorithm providing relatively fast and accurate verification suitable for realtime applications. These techniques involve sampling minutiae into subsets based on geometrical proximity to a particular index minutiae for each subset, eliminating minutiae which are near the fingerprint boundary, ordering the minutiae of each subset in a predetermined manner, classifying each subset in bins which discriminate on the properties of the subsets, checking transformations proposed as matches between fingerprints for consistency with each other, and checking the topological as well as the geometrical correspondence of potentially matching fingerprints. In a preferred form, a score indicative of the correlation between fingerprints is used to ultimately decide whether the fingerprints match.

Abstract: In an automatic fingerprint authentication or identification system, the fingerprint image acquisition is severely effected by the limitations of the acquisition process. The two modes of input, viz. scanning inked fingerprints from paper records or directly from a finger using live-scan fingerprint scanners suffer from the following noise sources in the input in addition to standard noise in the camera. Non-uniform ink application, uneven pressure while rolling on the paper or pressing on the scanner surface and external dirt like oil and climatic variations in the moisture content of skin are some of the main causes for the ridges and valleys not to be imaged clearly. This invention deals with a method of learning a set of partitioned least-sqaures filters that can be derived from a given set of images and ground truth pairs as an offline process. The learned filters are convolved with input fingerprint images to obtain the enhanced image.