Digit recognition for Arabic/Jawi and Roman using features from triangle geometry

Abstract

A statistical approach to automatic speech recognition using the atomic speech units constructed from overlapping articulatory features Abstract. A novel method is proposed to recognize the Arab/Jawi and Roman digits. This new method is based on features from the triangle geometry, normalized into nine features. The features are used for zoning which results in five and 25 zones. The algorithm is validated by using three standard datasets which are publicly available and used by researchers in this field. The first dataset is HODA that contains 60,000 images for training and 20,000 images for testing. The second dataset is IFHCDB. This dataset has 52,380 isolated characters and 17,740 digits. Only the 17,740 images of digits are used for this research. For the roman digit, MNIST are chosen. MNIST dataset has 60,000 images for training and 10,000 images for testing. Supervised (SML) and Unsupervised Machine Learning (UML) are used to test the nine features. The SML used are Neural Network (NN) and Support Vector Machine (SVM). Whereas the UML uses Euclidean Distance Method with data mining algorithms; namely Mean Average Precision (eMAP) and Frequency Based (eFB). Results for SML testing for HODA dataset are 98.07% accuracy for SVM, and 96.73% for NN. For IFHCDB and MNIST the accuracy are 91.75% and 93.095% respectively. For the UML tests, HODA dataset is 93.91%, IFHCDB 85.94% and MNIST 86.61%. The train and test images are selected using both random and the original dataset's distribution. The results show that the accuracy of proposed algorithm is over 90% for each SML trained datasets where the highest result is the one that uses 25 zones features.

State of the Art in Digital Paleography

“…The study of Roman handwriting has been in existence for more than four decades [23]. There were no available standard datasets that could be reached by the research community at that time.…”

“…Then, the dataset was named MNIST [24]. The MNIST dataset has been widely used among researchers in handwriting recognition [23], [25].…”

Skew correction for mushaf Al-Quran: a review

“…However, skew correction is performed on the document after scanning the document and before it appears on the computer screen. Arabic/Jawi based skew correction processes have been proposed by [2]. However, the step to detect the skew of pages or lines for the Mushaf Al-Quran cannot be the same as Arabic/Jawi documents.…”

Skew Detection and Correction of Mushaf Al-Quran Script using Hough Transform

“…Furthermore, accurate skew detection and correction helps other processes of OCR to be more successful. In [2] a novel method was proposed to recognize Arab / Jawi and roman digit by OCR. In [3] skew in documents can be classified into three class namely global skew, multiple skew and no-uniform text line skew.…”

“…However, several methods are used for skew correction like (direct method, indirect method contour oriented projection based and others), rotation of the image is done through Affine Transformation (2).…”