Search or add a thesis

Advanced Search (Beta)
Home > Optical Character Recognition for Printed Urdu Nastaliq Font

Optical Character Recognition for Printed Urdu Nastaliq Font

Thesis Info

Access Option

External Link

Author

Din, Israrud

Program

PhD

Institute

Bahria University

City

Islamabad

Province

Islamabad

Country

Pakistan

Thesis Completing Year

2019

Thesis Completion Status

Completed

Subject

Computer Science

Language

English

Link

http://prr.hec.gov.pk/jspui/bitstream/123456789/14662/1/Israrud-Din%20computer%20engg%202019%20bahria%20isb%20prr.pdf

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676727801984

Similar


Optical Character Recognition (OCR) is one of the most investigated pattern classification problems that has received remarkable research attention for more than half a century. From the simplest systems recognizing isolated digits to end-to-end recognition systems, applications of OCRs vary from postal mail sorting to reading systems in scene images facilitating autonomous navigation or assisting the visually impaired. Despite tremendous research endeavors and availability of commercial recognition engines for many scripts, recognition of cursive scripts still remains an open and challenging research problem mainly due to the complexity of script, segmentation issues and large number of classes to recognize. Among these, Urdu makes the subject of our study. More specifically, this study investigates the recognition of printed Urdu text in Nastaliq style, the most widely employed script for Urdu text that is more complex than the Naskh style of Arabic. This work presents a holistic (segmentation-free) technique that exploits ligatures (partial words) as units of recognition. Urdu has a total of more than 26,000 unique ligatures, many of the ligatures, however, share the same main body (primary ligature) and differ only in the number and position of dots and diacritics (secondary ligatures). We exploit this idea to separately recognize the primary and secondary ligatures and later re-associate the two to recognize the complete ligature. Recognition is carried out using two techniques; the first of these is based on hand-crafted statistical features using hidden Markov models (HMMs). Features extracted using sliding windows are used to train a separate model for each ligature class. Feature sequences of the query ligature are fed to all the models and recognition is carried out through the model that reports the maximum probability. The second technique employs Convolutional Neural Networks (CNNs) to automatically extract useful feature representations from the classes and recognize the ligatures. We investigated the performance of a number of pre-trained networks using transfer learning techniques and trained our own set of networks from scratch as well. Experimental study of the system is carried out on two benchmark datasets of Urdu text, the ‘Urdu Printed Text Images’ (UPTI) database and the ‘Center of Language Engineering’ (CLE) database. A number of experimental scenarios are considered for system evaluation and the realized recognition rates are compared with state-of-the-art recognition systems for printed Urdu text. An interesting aspect of experimental study is the combination of unique ligatures in the two datasets to generate a large set of around 2800 unique primary and secondary ligatures covering a major proportion of the Urdu corpus. The system reports high classification rates (88.10% and 94.78% on CLE and UPTI query ligatures respectively) demonstrating the effectiveness of the proposed recognition techniques which can be adapted for other cursive scripts as well. The findings of this study are expected to be useful for the document recognition community in general and researchers targeting cursive scripts in particular.
Loading...
Loading...

Similar Books

Loading...

Similar Chapters

Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...

109. Al-Kafirun/Those who knowingly deny

109. Al-Kafirun/Those who knowingly deny

I/We begin by the Blessed Name of Allah

The Immensely Merciful to all, The Infinitely Compassionate to everyone.

109:01
a. Say:
b. ‘O you who knowingly deny the truth of Islam!

109:02
a. I will not submit in awe and worship to what you worship,

109:03
a. nor will you worship the One WHOM I submit in awe and worship – Allah, The One and Only God.

109:04
a. And neither have I ever submitted in awe and worship to what you worship,

109:05
a. nor have you ever worshiped the One WHOM I submit in awe and worship,

109:06
a. Therefore, for you, is your religion and its consequential ramifications in the Sight of Allah,
and
b. for me, is my religion’ and its consequential ramifications in the Sight of Allah.

جماعت احمدیہ کے مولوی عبد اللطیف بہاولپوری کی چار قرآنی سورتوں کی تفاسیر کا تحقیقی و تنقیدی جائزہ

This informative article is a vital as well as analytical analyze of the several Sūrʼas translated as well as defined by Mūlvi Abdul Latīf around the facets of the guidelines connected with Translation as well as Tafsīr set by Mirza Ghulām Ahmad Qādyāni founder of Jamʽat-e-Āḥmadiya. Who offered a brand new principle connected with Tafsīr to verify the inappropriate beliefs as well as his views that are total contrary to the principles set by authentic former Muslim scholars. Many Qādyāni Mufasrīn implemented those principles within their books connected with Tafsīr. Most notable ended up being Mūlvi Abdul Latīf Bahāwalpūri who had written this Translation as well as Tafsīr of 5 Sūrʼas i. ESūrʼa Banī ʼisraeel, Sūrʼa Kahaf, Sūrʼa Yāseen, Sūrʼa Qiyāmah and Sūrʼa Dahar. He implemented the guidelines set by Mirza Ghulām Ahmad Qādyāni. Throughout his work he created a number of alterations not only with Translation but with Tafsīr too. This article is an eye bird review of the principles of the Translation as well as Tafsīr connected with Holy Qurʼan set by authentic former scholars.

Comparative Evaluation of Dris Diagnosis and Recommendation Integrated System and Cnl Critical Nutrient Level Approaches for Nutrient Deficiency Diagnosis in Sugarcane

Sugarcane crop and sugar industry play pivotal role in the economy. Imbalanced nutrition limits the crop yield. Sugarcane nutrition tools based on soil test and tissue nutrient concentrations using traditional Critical Nutrient Levels (CNL) approach has limited applicability. Plant nutritional status interpretation, based on comparison of nutrient pairs with norms developed for higher yielding populations known as Diagnosis and Recommendation Integrated System (DRIS) could prove better for diagnosis of nutrient imbalances and optimizing sugarcane nutrition. One hundred twenty three observations on macro- and micronutrient concentration and associated yield with known soil type and sugarcane varieties were recorded through a field survey and DRIS norms were developed for sugarcane in lower Sindh (district Thatta). Third leaf and sheath (and fourth leaf for comparison) was sampled at the grand growth stage during mornings (7:00 am to 10:30 am) from each of the 123 sugarcane plantations. The plant samples were oven dried at 68 °C, ground, and digested in HNO3:HClO4 acids mixture (5:1). The digests were analyzed for phosphorus colorimetrically, potassium, copper, iron, manganese and zinc through atomic absorption spectroscopy, and nitrogen was measured by Kjeldahl’s method; and plant boron was determined through dry ashing and the concentration was measured colorimetrically. Surface soil samples were collected from each of the 123 sites, air dried and crushed to pass through 2mm sieve, and analyzed for EC, pH, calcium carbonate, particle size distribution, nitrate nitrogen, plant available phosphorus, potassium, zinc, copper, iron, boron and manganese. Nitrate was extracted in 2 M KCl and measured colorimetrically. Plant available phosphorus zinc, copper, iron, and manganese were extracted by ABDTPA, and boron with hot boiling water. The plant nutrient concentration data was fitted to DRIS using Beaufils methodology. Yield was recorded by harvesting 5x3 m2 area by each location. The data were organized on soil association and variety basis for each location, and analyzed for variance using general linear model implemented through SAS version 9.2. The soils were calcareous with pH 7.7 to 8.7 (mean 8.2), low in soil test nitrogen, low to medium in extractable P, and adequate in extractable potassium. Among the micronutrients, zinc was low, boron was medium and copper and iron were adequate. The soil test level differed a little with soil association except for plant available iron. Selected soil nutrients were found spatially variable. The soil zinc was lower in Mirpur Sakro and Thatta sub districts (Talukas) and high soil zinc was towards Sujawal-Jati sub districts. Similar spatial pattern existed for plant available iron, potassium, and boron which was related with soil type; and the land capability map further helped to understand the spatial variation in the nutrient status in the sugarcane growing area. Plant index tissue nutrients differed significantly (p < 0.01) with the soil type except for nitrogen and phosphorus. The highest accumulation of potassium in plant was from Borium, Gungro, and Arib soils and lowest from Gujo and Katiar soils. The highest copper concentration was observed in sugarcane grown on Gujo soil association while the lowest copper concentration was found in sugarcane grown on Bulri soil associations. Plant nutrient concentration also differed significantly with variety. The sugarcane varieties BL-4, Thatta-10 and Triton had nitrogen and phosphorus contents below their critical value and out of the optimum range. Potassium was above the critical value, even greater than the optimum range. Zinc was in the optimum range in sugarcane grown on Gujo, Gujo-Shahdara complex, Daro, and Rustam, and below the optimum range in other soils. Boron was lower than the optimum range on Borium and Bulri soils. Copper was optimum on all the soils and iron was deficient on all the soils. This suggested that edaphic factors influenced nutrient levels in plant. Also, Triton which normally had greater nitrogen content than BL-4, had lower N content when grown on Arib soil. Overall, N and K nutrients in Triton were below the critical value and optimum range. BL-4 grown on Daro soil had more N than the sugarcane varieties Triton and Thatta-10. In Gungro soil, BL-4 had lesser N than Thatta-10 and Triton. Similar is the case in Rustam soil. Therefore some varieties were better accumulator of some micronutrients on certain soils than the other varieties grown on the same soil. The mean and range of individual nutrients were different in low and high yielding populations of the varieties. High yielding population had greater nitrogen, phosphorus, potassium and manganese than the low yielding population of respective varieties. The magnitude of difference for zinc and boron was far greater while copper and iron concentration difference between low and high yielding populations was negligible. Low yielding population had wider nitrogen phosphorus ratio than high yielding population of corresponding varieties, while nitrogen to potassium ratio had opposite trend. Similarly the ratio of nitrogen to micronutrients was wider in low yielding population suggesting more nitrogen than micronutrients. The contrasts for the nutrient ratios between low and high yielding populations are discussed. DRIS indices derived from sugarcane fields of lower Sindh revealed that nitrogen ranged from 2.96 to 4.51which indicated that nitrogen was sufficient under the current practice of fertilization. The phosphorus index value of -6.23 to 3.58 indicated deficiency and the need for additional phosphorus application was identified for certain varieties. The potassium indices of 2.57 to 8.10 indicated high level of potassium in the sugarcane plant tissue. This level of potassium as determined by the DRIS reflected the luxurious uptake of potassium by the sugarcane. The indices for zinc ranging from -12.23 to -8.92, magnitude of difference from zero of balanced nutrition showed the severity of deficiency. From the results and comparison with other studies it was apparent that the potential response of sugarcane is likely to be high to the application of zinc. The indices for boron ranged from -14.87 (deficient) to -0.26 (adequate) showing the most severe deficiency of this element indicating high probability of response to boron application. The average indices of copper ranged from 4.59 to16.17 and iron from 7.24 to 12.98, indicating high status of these nutrients in the sugarcane plant tissue. The study provides guidelines for sugarcane nutrition on a regional level, large commercial growers and policy makers can benefit from the findings.