A Search Engine for Handwritten Documents

The design and functionality of a versatile search engine on handwritten documents is described. Documents are indexed using global image features, e.g., stroke width, slant, word gaps, as well local features that describe shapes of characters and words. Image indexing is done automatically using page analysis, page segmentation, line separation, word segmentation and recognition of characters and words. Several types of searches are permitted: (i) Word / Phrase Spotting; (ii) Text to Image Search; (iii) Plain Text Search; (iv) Word Recognition from Lexicon. The words in the document are characterized by various features and it forms the basis for the different searching techniques. The system was implemented in Microsoft Visual C++. The paper reports on the functional capabilities of the various search techniques and their performance.