Arabic character recognition using fourier descriptors and character contour encoding

Indian script character recognition: a survey

“…The feature-based approaches can be of two types, namely spatial domain and transform domain approaches [25][26][27][28][29][30][31][32]. Spatial domain approaches derive features directly from the pixel representation of the pattern.…”

Shape Understanding System – Knowledge Implementation and Learning

“…In literature the term 'shape' often refers to the geometry of an object's physical surface [16]. Existing methods of shape analysis are mostly concerned with shape recognition [84], [85] [86], [87], [88,89], [90], [91], [92], [93] [94]. Visual systems applying shape as their knowledge are called model-based object recognition systems and have been used extensively by vision researchers [70], [95], [96], [97].…”

Hierarchical On-line Arabic Handwriting Recognition

“…Some of these approaches extract features from the boundaries or the skeleton of the letters to define Fourier descriptors [11,16,4] or rely on the Fourier spectrum of the characters [4,17]. Other approaches extract features from the character strokes, which are fed to a Bayes classifier to recognize the input character [18].…”