Pattern Matching Model for Recognition of Stone Inscription Characters

Abstract

As there are countless significant works done for handwritten character recognition, very meager effort has been reported for inscription characters especially for Tamil stone inscriptions. The real challenge faced in handling stone inscription is dataset collection and foreground and background discrimination. Till present days, the archeological department follows traditional way of capturing, preserving and deciphering stone inscriptions which is manual, more time consuming and need expert assistance. Hence digitized recognition is essential and efficient pattern matching algorithm is needed to be developed to deal with variations in shape and size of complex structured characters present in Tamil stone inscriptions. In this paper, an automated character recognition by pattern matching approach is developed, where character features were extracted by using pattern matching algorithm that helps achieving good recognition rate. Recognition of ancient Tamil stone inscriptions characters and finding their corresponding contemporary Tamil character is done by Image-based Character Pattern Identification (ICPI) system. Modified Speeded Up Robust Feature with Bag of Grapheme (MSURF-BoG) algorithm is implemented to detect the strongest key points from the input character with different orientations. These key point features were created for training the image as a model called Bag of Grapheme (BoG) with code word creation. Hence unsupervised key point features were extracted and pattern matching is performed. 11th century Tamil stone inscriptions were taken as samples which has 7 vowels and 17 consonants, totally 24 characters were used. Here samples with different orientation from each 24 character were used for training the system. The proposed system is evaluated by recognition accuracy which is reported for character wise at the maximum of 96%.

Manhole Cover Classification Based on Super-Resolution Reconstruction of Unmanned Aerial Vehicle Aerial Imagery

“…Traditional text recognition is achieved through optical character recognition (OCR) technology [28,29], whereas the text obtained from drone aerial photography loses a large number of textual features, and the text is not in the form of a standardized font. It is not feasible through traditional text recognition methods, and this can be ignored by using image classification networks for text recognition.…”

Self-Adaptive Hybridized Lion Optimization Algorithm With Transfer Learning for Ancient Tamil Character Recognition in Stone Inscriptions

“…Offline isolated Tamil character recognition [36 ] Otsu thresholding method, ANN OCR based model for both ancient and modern tamil character recognition [37] Convolutional Neural Network based model Trained in offline mode with Tamil characters [38] VGG16 and CNN 16 layers of Feed forward deep learning and Transfer learning are used to perform classification [39] Soft computing techniques and boggle algorithm Segmented image is decomposed using hybrid feature extraction technique along with Chi-square test and final characters are converted into word form [40] Artificial Neural Network Associate with Opposition Based Grey Wolf Optimization Algorithm ANN recognizes ancient south Indian characters and their weights are updated using Opposition Based Grey Wolf Optimization Algorithm [41] Image-based Character Pattern Identification (ICPI) & Modified Speeded Up Robust Feature with Bag of Grapheme (MSURF-BoG) algorithm Detect the strongest important spots from the input character with varied orientations algorithm.…”

Comparative Study: Enhancing Legibility of Ancient Indian Script Images from Diverse Stone Background Structures using 34 Different Pre-Processing Methods

“…The final step integrates morphological processing to enhance characters, particularly in cases where thresholding or other preprocessing steps may have eroded or added pixels to characters. There are many languages in India and so the diversity in scripts throughout, much research has been documented about preprocessing of ancient inscriptions; however holistic approaches to understanding these recent technologies and evaluation of all such techniques are very few [9]. For instance; Buzykanov 2012 [10] proposed the methodology for improving pixel density of text images via low pass signal Gaussian Laplacian filtering algorithm; however evaluation of such technique and comparing this method with other efficient methods are not been documented since then.…”