Devanagari

We report our recent work on the recognition of scene text captured by mobile cameras, which we have named Kannada Pado. The text region is currently manually cropped using a user-friendly interface, which permits repeated croppings from the captured image in a hierarchical fashion. The scene text segment is then binarized using the algorithm, midline analysis, and propagation for segmentation. The segmented binary text image is recognized using Lipi Gnani Kannada OCR. The recognized text can be transcribed in Roman, Devanagari, and other principal Indian scripts. Such tools will be of immense use in metropolitan cities such as Bengaluru for business visitors and tourists to be able to read important textual information using their mobile itself. The entire implementation is of low computational complexity and hence, runs fully on the mobile itself, without any backend computation. Currently, text recognition accuracy is the bottleneck, which, when improved, will make the app immediately usable by people. Then, it will be made available to the public from Google Playstore.

The Shiva Sutra was revealed to and written down by Vasugupta (ca 875--925 CE). The Sutra is considered mystical and of divine origin. For Kashmir Shaivism, it is one of the most important key sources. It outlines the teachings of Shaiva non-dualism, where the focus is on attaining the Ultimate Reality in which everything is created and dissolved. This ultimate state is called Param Shiva and is beyond description. For attaining this state of Shiva for those who remember to reside in their own inherent-self-nature, which is of the nature of Shiva, no effort or no way is needed. For everyone else there are three ways for the attainment of Param Shiva described in the Shiva Sutra. There is no strict order given for meditating on the Sutra. It depends on one's stage of evolution. Translation from the original Sanskrit with notes.

Abstract and Book Reviews
Abstract:
This publication details how (copying) and when well before 3200 years ago) the Western Alphabet character sequence (abecedary) was modeled after a pre-Sanskrit character grid (abugida).

By forcing the Western-Late Roman Alphabet character sequence (the ABC or abecedary) into a tabular format by starting out to line up the vowels in one column followed by fitting the consonants sequentially into three adjacent columns, a somewhat orderly and surprising result was obtained.

By comparing the resultant character grid with an early simpler phase of the Sanskrit Devanāgarī character grid, it became possible to discover clear links between Western alphabets and an early form of the Sanskrit Devanāgarī abugida (alphasyllabary).

This dependent linkage between the Western Late-Roman and the Devanāgarī abugida is evidenced by their similar if not almost identical ordering of characters in grids of comparable phoneme classifications.

In this study, the origin of the dissimilarities - or, in fact, irregularities - has been traced back to two errors made by an ancient Near Eastern linguist after his study-visit to ancient India.

As the Western Late-Roman Alphabet in this study's tabular arrangement appeared to be irregular in certain places, especially when compared to the highly organized and well ordered Devanāgarī abugida grid, this study demonstrates how, why and approximately when (before 1200 BCE) those irregularities came about.

When the two sequential formats are superficially compared to each other - before the tracing of errors and before the tabular reconstruction - a percentage of similarity of only 20 % (4 out of 20 characters) was calculated.
However, after the error identifications (described in detail) and while taking into account the varying but close pronunciations of a number of comparable characters in side-by-side alphabet and abugida grids, and while focusing on the placement of nearly all Western Late-Roman characters in their appropriate vowel, labial, guttural and dental columns (illustrated in detail) an 85.00 % match between them was arrived at.

This study also demonstrates how the Ugarit abecedary from Syria (1400 -1200 BCE) is evidence that even before it was inscribed in cuneiform, that a 'West-of-India' style irregular alphabetic sequence was already in use, which, as demonstrated in this study, was based on a very early form of a Pre-Sanskrit Brāhmī abugida."

Book Reviews:

“This Alphabet paper is simply brilliant stuff, serendipity at its finest. The tabulation similarity of ABCD [the western abecedary] with Panini’s abugida [Sanskrit alphabet] is totally new and totally brilliant [and] on par with a description of Fleming’s discovery of Penicillin.”
~ Dr. Shiv Sastry MD

“A study by my friend, original thinker along the lines of Vartak, Einstein, Newton and Kepler. So expect flight of fancy, joy of speculation and wonderful research.
One of a few brilliant Jewels.
~ Nilesh Oak (Author of “When Did The Mahabharata War Happen? - The Mystery of Arundhati” and "The Historic Rama")

“Wim Borsboom takes the reader on a fascinating brief trip through the history of real magic; reading and writing.”
~ S. M. Sulllivan (Author of “The Indus Script Dictionary”)

“Wim Borsboom has hit upon a brilliant idea related to the origins of the western alphabet. He shows that in its basic form, it is similar to the scientifically organized Sanskrit alphabet. This research is of importance not only to the student of alphabets but also to the historian and the layperson interested in understanding cultural interactions in the ancient world.“
~ Subhash Kak, Regents Professor, Oklahoma State University. (Co-author with Georg Feuerstein and David Frawley of “In Search of the Cradle of Civilization”)

“The sequence he has discovered in the alphabet is really fascinating. I don't know how no-one noticed this sequential phonetics before.”
~ Shrikant G. Talageri (Author of “The Rigveda: A Historical Analysis”)

The Sindhī language has been written in numerous scripts throughout its history. However, in the twentieth century, Perso-Arabic and Devanāgarī emerged as the dominant scripts for the language. Today, Perso-Arabic is the sole official script for Sindhī in Pakistan, while both Perso-Arabic and Devanāgarī are in concurrent use for the language in India. This paper identifies and analyses areas of orthographic standardisation and variation in the Perso-Arabic and Devanāgarī scripts for Sindhī , focusing primarily on practices in the Indian context. It first classifies orthographic variation into that stemming from phonological ambiguity, and that which is purely graphematic. The former includes the representation of reduced vowels, gemination, vocalic endings, loanwords, consonant clusters and sounds of unclear phonemic status. The latter includes the shapes and positioning of diacritics, allographs, derivative graphemes and collation orders. The paper concludes by summarising the possible pedagogical implications of such orthographic standardisation and variation.

Many scholars, politicians, and the lay people alike believe that Urdu in North India symbolizes a Muslim identity and culture. Based on an eight-month long ethnographic study and quantitative language data collected in Old Delhi, this
article challenges this notion and shows that the symbolic meanings of Urdu have been mutating in post-colonial India. A cross-generational study involving both Muslims and Hindus shows that different generations assign different meanings to Urdu. Unlike the older generation, Muslim youth do not identify
themselves with Urdu. A study of the Urdu sounds /f/, /z/, /kẖ /, /gẖ /, and /q/ in the speech of Muslim youth further demonstrates that they are losing three of these sounds. Another transformation involves the adoption of the Devanagari
script to write Urdu by many Muslims. This change in the literacy practices of Muslims reinforces the shift in the symbolic meanings of Urdu. I argue that the transformation in the symbolic meanings of Urdu is reflective and constitutive of the sociopolitical changes that Muslims have undergone in the twentieth century.

La presentación se propone explicar el desarrollo de la escritura en el subcontinente indio. Se desarrolla en cinco secciones: la primera corresponde al período denominado "proto-índico", luego se describe el período alfábetico en el cual se adaptaron escrituras de origen semítico. En tercer lugar, se describe el desarrollo de la escritura brahmi en el SE asiático. A continuación, se explica en detalle el alfabeto devanagari y las reglas fonéticas que éste implica. Por último se incluye una sección que desarrolla el concepto de "sandhi" y se explican algunas de sus reglas.

Print version available.
Please email wimjborsboom@gmail.com

A collection of essays, conversations and ruminations on:
(1) 'Innate Freedom' and the reclamation of it, if seemingly lost.
(2) Identity, ego and pseudo-self,
(3) Awareness, reality and illusion,
(4) The Buddha's Triple Gem and the Eightfold Path,
(5) Impermanence and suffering,
(6) Pain, suffering and illusion,
(7) Pleasure and pain,
(8) Entrapment, subservience and apparent loss of freedom,
(9) Kundalini.

A surprise awaits the reader who accompanies the writer on his investigation into the dynamics and language of entrapment, dependency and suffering while he endeavors to uncover the causes of apparent loss of human freedom.
Witness how he digs up clue after clue on how this innate and inherent freedom came to be treated as only conditionally available.
You are invited to look over his shoulders and find out with him how humankind was made to believe that dependence and suffering define life more than freedom and happiness do, and to discover that the claim on unconditional freedom and happiness is not only warranted but realizable for everyone.

Human Interactive Proofs (HIPs) are automatic reverse Turing tests designed to distinguish between
various groups of users. Completely Automatic Public Turing test to tell Computers and Humans Apart
(CAPTCHA) is a HIP system that distinguish between humans and malicious computer programs. Many
CAPTCHAs have been proposed in the literature that text-graphical based, audio-based, puzzle-based
and mathematical questions-based. The design and implementation of CAPTCHAs fall in the realm of
Artificial Intelligence. We aim to utilize CAPTCHAs as a tool to improve the security of Internet based
applications. In this paper we present a framework for a text-based CAPTCHA based on Devanagari
script which can exploit the difference in the reading proficiency between humans and computer
programs. Our selection of Devanagari script-based CAPTCHA is based on the fact that it is used by a
large number of Indian languages including Hindi which is the third most spoken language. There is
potential for an exponential rise in the applications that are likely to be developed in that script thereby
making it easy to secure Indian language based applications.

In this thesis, a classi fier based on local sub-unit level and global character level representations of a character, using stroke direction and order variation independent features, is developed for the recognition of Devanagari online handwritten characters. It is shown that online character corresponding to Devanagari ideal character can be analyzed and uniquely represented in terms of homogeneous sub-structures called the sub-units. These sub-units can be extracted using direction property of online strokes in an ideal character. A method for extraction of sub-units from a handwritten character is developed, such that the extracted sub-units are similar to the sub-units of the corresponding ideal character. Features are developed that are independent of variations in order and direction of strokes in characters. The features are called histograms of points, orientations, and dynamics of orientations (HPOD) features. The method for extraction of these features spatially maps co-ordinates of points and orientations and dynamics of orientations of strokes at these points. Histograms of these mapped features are computed in different regions into which the spatial map is divided. HPOD features extracted from the sub-units represent the character locally; and those extracted from the character as a whole represent it globally. A classifi er is developed that models handwritten characters in terms of the joint distribution of the local and global HPOD features of the characters and the number of sub-units in the characters. The classi fier uses latent variables to model the structure of the the sub-units. The parameters of the model are estimated using the maximum likelihood method. The use of HPOD features and the assumption of independent generation of the sub-units given the number of sub-units, make the classi fier independent of variations in the direction and order of strokes in the characters. This sub-unit based classi fier is called SUB classi fier. Datasets for training and testing the classi fiers consist of handwritten samples of Devanagari vowels, consonants, half consonants, nasalization sign, vowel omission sign, vowel signs, consonant with vowel sign, conjuncts, consonant clusters, and three more short strokes with different shapes. In all, there are 96 different characters or symbols that have been considered for recognition. The average number of samples per character class in the training and the test sets are, respectively, 133 and 29. The smallest and the largest dimensions of the extracted feature vectors are, respectively, 258 and 786. Since the size of the training set per class is not large compared to the dimension of the extracted feature vectors, the training set is small from the perspective of training any classi fier. Classifi ers that can be trained on a small data set are considered for performance comparison with the developed classi fier. Second order statistics (SOS), sub-space (SS), Fisher discriminant (FD), feedforward neural network (FNN), and support vector machines (SVM) are the other classi fiers considered that are trained with the other features like spatio-temporal (ST), discrete Fourier transform (DFT), discrete cosine transform (DCT), discrete wavelet transform (DWT), spatial (SP), and histograms of oriented gradients (HOG) features extracted from the samples of the training set. These classi fiers are tested with these features extracted from the samples of the test set. SVM classifi er trained with DFT features has the highest accuracy of 90.2% among the accuracies of the other classi fiers trained with other features extracted from the test set. The accuracy of SUB classi fier trained with HPOD features is 92.9% on the test set, which is the highest among the accuracies of all the classi fiers. The accuracies of the classi fiers SOS, SS, FD, FNN, and SVM increase when trained with HPOD features. The accuracy of SVM classi fier trained with HPOD features is 92.9%, which is the highest among the accuracies of the other classifi ers trained with HPOD features. SUB classi fier using HPOD features has the highest accuracy among the considered classi fiers trained with the considered features on the same training set and tested on the same test set. The better character discriminative capability of the designed HPOD features is reflected by the increase in the accuracies of the other classi fiers when trained with these features.

This thesis explores the visual similarity that underlies the coherence in the design of individual typefaces. Typeface designers aim to achieve a unifying coherence in their typefaces, so that characters can be identified individually as well as belonging together giving rise to an overall style. The objective is to determine whether the coherence perceived by readers differs from the coherence intended by designers. The research is cross-disciplinary, combining empirical studies of readers’ perceptions with a computational model that is based on relevant typeface design knowledge.

Character similarity is studied in multiple different typefaces (fonts) intended for continuous reading in Cyrillic, Devanagari, and Latin scripts. The studies were conducted online to collect a large number of responses. The participants were presented with a sequence of character triplets. They were asked to identify the odd one out in each of these triplets judging by their visual similarity, thus making a statement about the similarity of the two complementary characters. This method studies the similarity in context, which provides more refined details about participants’ similarity judgements.

The model interprets characters using two kinds of features: more specific parts and more general roles. The model learns the relative saliences of these features from a subset of the data collected in the studies. This allows the model to predict participants’ responses to the triplets from the studies and for other, unseen triplets. Additionally, the model can provide explanations of the criteria participants used in their similarity judgements and can generate similarity matrices.

The model achieved high scores when predicting response probabilities and identifying the overall odd ones out. A view of coherence that is supported by readers’ perception can be used to assist designers in their creative process, help with fonts’ quality assessments, and contribute to readability research and multi-script typography.

We report our recent work on the recognition of scene text captured by mobile cameras, which we have named Kannada Pado. The text region is currently manually cropped using a user-friendly interface, which permits repeated croppings from the captured image in a hierarchical fashion. The scene text segment is then binarized using the algorithm, midline analysis, and propagation for segmentation. The segmented binary text image is recognized using Lipi Gnani Kannada OCR. The recognized text can be transcribed in Roman, Devanagari, and other principal Indian scripts. Such tools will be of immense use in metropolitan cities such as Bengaluru for business visitors and tourists to be able to read important textual information using their mobile itself. The entire implementation is of low computational complexity and hence, runs fully on the mobile itself, without any backend computation. Currently, text recognition accuracy is the bottleneck, which, when improved, will make the app immediately usable by people. Then, it will be made available to the public from Google Playstore.

The study of character research is an active area for research as it pertains a lot of challenges. Various pattern recognition techniques are being used every day. As there are so many writing styles available, development of OCR (Optical Character Recognition) for handwritten text is difficult. Therefore, several measures have to be taken to improve the recognition process so that the burden of computation can be decreased and the accuracy for pattern recognition can be increased. The main objective of this review was to recognize and analyze handwritten document images. In this paper, we present a scheme to identify different Indian scripts like Devanagari and Gurumukhi.

Devanagari ancient manuscript recognition framework is drawing a lot of considerations from researchers nowadays. Devanagari ancient manuscripts are rare and delicate documents. To exploit the priceless information included in these documents, these documents are being digitized. Optical character recognition process is being used for the recognition of these documents. This paper presents a system for improvement in recognition of Devanagari ancient manuscripts using AdaBoost and Bagging methodologies. Discrete cosine transform (DCT) zigzag is used for feature extraction. Decision tree, Naïve Bayes and support vector machine classifiers are used for the recognition of basic characters segmented from Devanagari ancient manuscripts. A dataset of 5484 pre-segmented characters of Devanagari ancient documents is considered for experimental work. Maximum recognition accuracy of 90.70% has been achieved using DCT zigzag features and RBF-SVM classifier. AdaBoost and Bagging ensemble methods are used with the base classifiers to improve the accuracy. Maximum accuracy of 91.70% is achieved for adaptive boosting (AdaBoost) with RBF-SVM. Various parameters for performance measures such as precision, recall, F-measure, false acceptance rate, false rejection rate and RMSE are used for assessing the quality of the ensemble methods.

The study of character research is an active area for research as it pertains a lot of challenges. Various pattern recognition techniques are being used every day. As there are so many writing styles available, development of OCR (Optical Character Recognition) for handwritten text is difficult. Therefore, several measures have to be taken to improve the recognition process so that the burden of computation can be decreased and the accuracy for pattern recognition can be increased. The main objective of this review was to recognize and analyze handwritten document images. In this paper, we present a scheme to identify different Indian scripts like Devanagari and Gurumukhi.

In this paper a number of classical methodologies together with the recent works are considered for comparison and study of the concept of segmentation for both document and graphic images. we propose segmentation method for noisy document. The segmentation method is proposed for Devnagari document by considering the structure of Devnagari script .We have also consider the irregularities of Devnagari writing style ,Yuktaksha and numbers. We are not removing shirorekha. It separates the image text documents into lines, words and characters. We are getting 100% accuracy at line and word level but at character level it depend on the type of character, whether it is single ,Yuktakshar or triakshar.