| 摘要 |
The present invention provides a computer-implemented method for indexing numerical information embedded in one or more electronic files. The method comprises determining whether an electronic file comprises one or more images containing embedded numerical data, including the steps of inputting the one or more images into a classification system comprising a plurality of interconnected classifiers; and classifying the one or more images using the classification system to output data classifying each image. The output data classifies each image as one of: containing embedded numerical data or not containing embedded numerical data. The method further comprises analysing the file to output data classifying it as one of: containing tabulated numerical data or not containing tabulated numerical data. If the outputted data indicates that the file comprises one or more images with embedded numerical data and/or contains tabulated numerical data, and the method further comprises extracting text and/or other data associated with the numerical data and indexing this text and/or other data in a database. |
