A View-Based T¿eplitz-Matrix-Supported System for Word Recognition without Segmentation

Abstract

In this paper, new modifications and experiments for word recognition and classification are presented. The algorithm is based on recognizing the whole words without separating them into letters. The whole word is treated and analyzed as an image. The method is based on the modification of a novel view-based word recognition algorithm -an approach that was successfully used by the authors' in previous works.This method shows how to recognize words without segmentation. The top and bottom views of the word are analyzed in order to create the feature vector. Then the feature vector is processed by the aid of Töeplitz matrices. The obtained series of Töeplitz matrix minimal eigenvalues are used for classification. The results are promising.

Handwritten Script and Word Recognition – A View-Based Approach

“…Here we present essential results of our experiments with words and characters. All of presented results are taken from our previous papers [2][3][4][5][6]. Here we select the most interesting and significant ones.…”

“…Further experiments were performed on unsegmented words. Two separate databases were used -a set of 75 English names of animals [4][5] and set of 150 most common English words [6]. This list was composed basing on appearance frequency in real-world documents [14].…”

“…In next works [4][5][6] this method was used for recognizing unsegmented words. This approach bypasses the stage of segmentation -the whole word is treated as a single image, and processed in that way.…”