A Rank based Ensemble Classifier for Image Classification using Color and Texture Features
Publish Year: 1392
نوع سند: مقاله کنفرانسی
زبان: English
View: 1,202
This Paper With 6 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
ICMVIP08_179
تاریخ نمایه سازی: 9 بهمن 1392
Abstract:
This paper presents a color image classificationmethod using rank based ensemble classifier. In this paper, weuse color histogram in different color spaces and Gaborwavelet to extract color and texture features respectively.These features are classified by two classifiers: NearestNeighbor (NN) and Multi Layer Perceptron (MLP). In theproposed approach, each set of features are classified by eachclassifier to generate a rank list of length three. Therefore, wehave some rank list for different combination of feature setsand classifiers. The generated rank lists present an ordered listof class labels that the classifier believes the input image isrelated to those classes in order of priority. To combine theoutputs (rank list) of each classifier, simple and weightedmajority vote are used. Experiments show the proposed systemwith weighted majority vote achieves a recall and precision of86.2 % and 86.16 % respectively. Our proposed system hashigher efficiency in comparison of other system
Keywords:
Authors
Fatemeh Ahmadi
Department of Mathematics and Computer Science Amirkabir University of Technology
Mohamad Hoseyn Sigari
Control and Intelligent Processing Center of Excellence Electrical and Computer Engineering Department
Mohamad Ebrahim Shiri
Department of Mathematics and Computer Science Amirkabir University of Technology (Tehran Polytechnic)
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :