A segmentation-free approach to text recognition with application to Arabic text

Abstract

Abstract. In recognizing cursive scripts, a major undertaking is segmenting cursive words into characters and isolating merged characters. The segmentation is usually the pivotal stage in the system to which a sizable portion of processing is devoted and a considerable share of recognition errors is attributed. The most notable feature of Arabic writing is its cursiveness. Compared to other features, the cursiveness of Arabic words poses the most difficult problem for recognition algorithms. In this work, we describe the design and implementation of an Arabic word recognition system. To recognize a word, the system does not segment it into characters in advance; rather, it recognizes the input word by detecting a set of "shape primitives" on the word. It then matches the regions of the word (represented by the detected primitives) with a set of symbol models. A spatial arrangement of symbol models that are matched to regions of the word, then, becomes the description of the recognized word. Since the number of potential arrangements of all symbol models is combinatorially large, the system imposes a set of constraints that pertain to word structure and spatial consistency. The system searches the space made up of the arrangements that satisfy the constraints, and tries to maximize the a posteriori probability of the arrangement of symbol models. We measure the accuracy of the system not only on words but on isolated characters as well. For isolated characters, it has a recognition rate of 99.7% for synthetically degraded symbols and 94.1% for scanned symbols. For isolated words the system has a recognition rate of 99.4% for noise-free words, 95.6% for synthetically degraded words, and 73% for scanned words.

A Recognition-Based Approach to Segmenting Arabic Handwritten Text

“…Adjacent letters that form shapes similar to those of classes 15,16,17,18,19,20,21 and 22 may wrongly be combined although they are correctly classified before grouping. This scenario is very clear in Figure 9, in the word ‫,"ﺳﯿﺪي"‬ where the adjacent letters ‫"ﯿ"‬ and part of the letter ‫"ﺳ"‬ were recognized and classified as letter ‫,"ﺳ"‬ and the remaining part of ‫"ﺳ"‬ was classified as letter ‫.…”

“…Khorsheed and Clocksin proposed in 1999 another holistic system where features were extracted from a word's skeleton for recognition without prior segmentation [16].…”

Offline Arabic handwriting recognition: a survey

“…In 1996, Ymin and Aoki presented a two-step segmentation system which used vertical projection onto a horizontal line followed by feature extraction and measurements of character width [27]. Al-Badr and Haralick presented a holistic recognition system based on shape primitives that were detected with mathematical morphology operations (1996,1998) [28], [29]. Alherbish et al presented a parallel OCR algorithm which achieved a speed-up of 5.34 in 1997 [30].…”

Collaborative Combination of Neuron-Linguistic Classifiers for Large Arabic Word Vocabulary Recognition

“…Free segmentation approaches are based on modeling words in the form of feature vectors, extracted from overlapping windows 24,25,36,37,48 or in the form of spatial distribution of symbol models. 1 To recognize a word, the system proposed by Ref. 1 does not commit itself to a segmentation of the word, rather it simulates trying di®erent segmentation points and then chooses the best set of segmentation points that provides the best recognition.…”