Arabic video text detection for multimedia web processing

مقدم أطروحة جامعية

al-Qatami, Ahlam Abd al-Latif

مشرف أطروحة جامعية

al-Atum, Jalal Umar

أعضاء اللجنة

al-Khasawinah, Baha al-Din
Ahmad, Ashraf Hashim
Hirz Allah, Nail B.

الجامعة

جامعة الأميرة سمية للتكنولوجيا

الكلية

كلية الملك الحسين لعلوم الحوسبة

القسم الأكاديمي

قسم علم الحاسوب

دولة الجامعة

الأردن

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2012

الملخص الإنجليزي

The detection of scene and caption Arabic text from unconstrained, general-purpose video is an important research problem in the context of content-based retrieval and summarization of visual information.

Most published methods work on Latin text on the video frame.

Also, these methods have been developed for extracting Latin text from images that have been applied to video frames.

They do not satisfy Arabic text feature to disc, it to deter du accurate Arabic text area.

This thesis presents a-effable, T-stem for detecting Arabic text from unconstrained, general-purpose video.

In developing methods for extraction of text from video it was observed that no single algorithm could detect all forms of text so we propose an efficient Arabic text detection algorithm based on the Palladian operator in the frequency domain.

The zero crossing value is computed for each pixel in the Palladian-filtered image after all edges in four directions are detecting.

Classifier is then used to classify all the pixels of the filtered image into two clusters: text and non-text.

For each candidate text region, the corresponding region in the edge map of the input image undergoes projection profile analysis to determine the boundary of the text blocks.

Then, we employ empirical rules to eliminate false positives based on the geometrical properties.

Experimental results show that the proposed algorithm is able to detect texts of different fonts, contrasts and backgrounds.

Moreover, it outperforms three existing algorithms in terms of detection and false positive rates.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب

الموضوعات

• معالجة البيانات
• الإلكترونيات الرقمية
• اللغة العربية

عدد الصفحات

70

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-307152