12601

11528

كارشناسي ارشد

مخابرات سيستم

اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر

1396

دوازده، 63ص.: مصور، جدول، نمودار

ص. ع. به فارسي و انگليسي

محمدرضا احمدزاده

رسول اميرفتاحي، مهدي مهدوي

1396/05/04

كتابنامه

برق و كامپيوتر

مهندسي برق و كامپيوتر

ID11528

2 6 6 توصيفگرهاي مقاوم به چرخش 66 2 6 2 توصيفگرهاي مقاوم به تغيير مقياس 06 2 6 0 توصيفگرهاي مقاوم به نويز 46 2 2 توصيفگرهاي مبتني بر ويژگيهاي محلي 56 2 2 6 الگوهاي باينري محلي 61 LBP 2 2 2 توصيفگر واريانس 23 LBPV LBP 2 2 0 توصيفگر 03 LTP 2 2 4 توصيفگر LBP كامل شده 06 CLBP 2 0 محدوديتهاي ويژگيهاي محلي براي توصيف بافت 10 فصل سوم روش جديد پيشنهادي 0 6 مقدمه 80 0 2 استخراج ويژگي با روش پيشنهادي 34 0 0 مقاومت نسبت به چرخش 54 0 4 مقاومت نسبت به نويز 54 0 5 چالشهاي موجود در روش پيشنهادي 74 فصل چهارم ارزيابي روش پيشنهادي و نتايج آزمايشها 4 6 نحوه ارزيابي روش پيشنهادي 34 4 6 6 پايگاه داده 01 TC و نتايج 35 4 6 2 پايگاه داده 11 TC و نتايج 25 4 6 0 پايگاه داده 21 TC و نتايج 25 4 2 بررسي نتايج 15 فصل پنجم جمعبندي و پيشنهادها 5 6 دستاوردهاي روش پيشنهادي 85 5 2 چالشهاي روش پيشنهادي 85 5 0 پيشنهادها براي ادامه تحقيق 85 مراجع 35 چكيده انگليسي 41 نه

A Novel Texture Descriptor Based on LBP Arash Ahmadi arash ahmadi@ec iut ac ir 2017 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Farsi Supervisor Prof Mohammad Reza Ahmadzadeh ahmadzadeh@cc iut ac irAbstractClassification is one of the most important topics in image processing and computer vision Proper classification of an image requires selection of suitable features that discriminatedifferent images In addition ideal features must be invariant to factors such as rotation scaling brightness and noise The main features that are used in image processing are thosethat reflect color shape and texture of an image Texture is an important feature among manytypes of images Different images ranging from remote sensing images to microscopyimages all contain textures and therefore texture classification is used in many applicationsincluding automated inspection image retrieval and medical image analysis So far manymethods have been proposed to describe texture including statistical methods structuralmethods methods based on filters and methods based on models In this thesis we proposea texture descriptor based on more local features compared to the conventional methods Theproposed method is also rotation invariant The proposed descriptor vector has a higherresistance against noise when compared to other methods based on local features Theproposed descriptor is named Local Patch Difference Vector and is based on the Euclideandistance of circular neighborhoods values around each pixel The classification results of theproposed descriptor is comparable to other conventional methods and also when noise isapplied to images classification rate using the proposed method leads to significantly higherresults when compared to conventional methods In Outex databases when applyingGaussian noise with variance of 0 015 the classification results have been improved by anaverage 4 38 comparing to conventional methods based on local features Keywords Computer Vision Image Processing Classification Texture Descriptor

محمدرضا احمدزاده

رسول اميرفتاحي، مهدي مهدوي