Handwritten tifinagh character recognition using simple geometric shapes and graphs

Abstract

<p>In this paper, a graph based handwritten Tifinagh character recognition system is presented. In preprocessing Zhang Suen algorithm is enhanced. In features extraction, a novel key point extraction algorithm is presented. Images are then represented by adjacency matrices defining graphs where nodes represent feature points extracted by a novel algorithm. These graphs are classified using a graph matching method. Experimental results are obtained using two databases to test the effectiveness. The system shows good results in terms of recognition rate.</p>

Recognition of Tifinagh Characters Using Optimized Convolutional Neural Network

“…The Fig. 6 shows that the non-respect of the standards of writing of the grapheme of a character causes an ambiguity to classify the characters either for the machine (the algorithm) or for the human (taking example 2 of the same figure the real class is yakw , but the system predicted yagw (same problem for 1, 3,5,6). For sample 4 in Fig.…”

“…Ouadid et al [6] presented a work based on the extraction of structural features of a character and the spectral method to find the coherence between two graphs (characters) by calculating the eigenvalues and the eigenvectors of the graphic product of two graphs.…”

“…• The AHARRANE team's approach [8] is sensitive to noise and distortions since it is based on the statistical method to extract the characteristics of a letter, as well as the systems of [7] and [8] use only 31 characters instead of 33 Amazigh characters (they excluded yakw and yagw ). • The system [7] is tested only on black and white images • The Ouadid system [6] does not exceed 85% in accuracy for in the AMCHD database and it requires considerable time for preprocessing and recognition. • The Benaddy team's system [10] is tested only on RGB images; their approach needs 162.61 s to recognize 7,722 images.…”

Corpus-based technique for improving Arabic OCR system

“…For several decades, newspaper articles, books, and researches have been digitized to make resources available to researchers and readers. The digitization process is done by optical character recognition system is branch of pattern recognition and artificial intelligence that convert the image of the text into an machine-readable text, which makes these texts usable to be processed by other tools or task such as indexing, machine translation, and search engine [3]. Optical character recognition is difficult task for many reasons such as low scanning and printing quality.…”

Annealing strategy for an enhance rule pruning technique in ACO-Based rule classification

“…This model can then be utilized to classify the class values of other instances. Examples of classification tasks include intrusion detection, medical diagnosis, handwritten digit recognition, spam email detection, and bankruptcy determination [5][6][7][8][9]. In recent years, the field of swarm intelligence (SI) has been derived by observing the swarming behavior of some animals, such as ant colonies, flocking birds, and fish schools [10,11].…”