Abstract: A method of detecting the encoding utilized in an electronic document includes testing the text strings to determine whether the electronic document contains only text strings having legal numeric codes. A statistical analysis of the text strings is then conducted to provide a mapping of legally coded candidates. The legally coded candidates are ranked and combined with an expected ranking of legally coded candidates to provide a most probable character mapping.

Abstract: A neural network based improving the performance of an omni-font classifier by using recognized characters for additional training is presented. The invention applies the outputs of the hidden layer nodes of the neural net as the feature vector. Characters that are recognized with high confidence are used to dynamically train a secondary classifier. After the secondary classifier is trained, it is combined with the original main classifier. The invention can re-adjust the partition or boundary of feature space, based on on-line learning, by utilizing the secondary classifier data to form an alternative partition location. The new partition can be referred to when a character conflict exists during character recognition.

Abstract: A parallax method of wavelength labeling is based on optical triangulation. A complementary color pattern projected onto the surface is characterized by a continuous variation of the power ratio of two wavelength bands, and the profile can be measured at all points in the field of view. Shifts of the wavelength bands on separate detector arrays correspond to profile deviations. A signal processor calculates a normalized signal that is independent of surface reflectivity and roughness variations; the phase shift of this signal yields depth data from which the surface profile can be mapped.