GRAMMATIC EXAMINATION OF VISUAL STRUCTURES OF THE DOCUMENT

摘要

1. A system that facilitates recognition, comprising:! a receiving component that receives input associated with the visual structure of the document; and! a grammatical component that uses, at least in part, a model of the grammatical hierarchical structure of the visual structure of the document to simplify the grammatical analysis of the input data. ! 2. The system according to claim 1, in which the grammatical component further comprises a component of the selection of the document structure, which highlights the layout structure associated with the document, which uses local and / or global features. ! 3. The system according to claim 2, in which the component of highlighting the structure of the document uses machine learning, evaluating the analysis based on the image representations, enhancing learning during analysis and / or signs of quick recognition to simplify highlighting the structure of the document layout. ! 4. The system according to claim 1, in which the grammatical component further comprises:! an analysis component that uses at least one classifier to simplify determining the optimal analysis result from global search results. ! 5. The system of claim 4, wherein the analysis component uses a classifier to simplify the determination of the grammatical value function. ! 6. The system of claim 5, wherein the classifier comprises a classifier trained using a conventional machine learning technique. ! 7. The system of claim 6, wherein the machine learning technique comprises, at least in part, a perceptron-based technique. ! 8. The system according to claim 1, in which the grammatical component uses the process of grammatical analysis on �