نویسندگان
1 دانشگاه بیرجند- دانشجوی دکتری مخابرات
2 دانشگاه بیرجند
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Polarimetric synthetic aperture radar (PolSAR) images contain so much information about the characteristics of the targets of the desired area with high resolution. Nowadays, using this data for terrain classification is known as a hot topic of interest for researchers. Recently, sparse representation-based technique as a powerful tool in the field of signal processing, has attracted a lot of attention. Therefore, in the first step, the structure of a sparse representation-based classifier is proposed. On the other hand, according to recent research results, ensemble classifier as an effective approach has more capabilities compare to single-classifiers. Therefore, in the next step, an ensemble classifier with Naïve Bayes combination rule is presented by using the sparse representation-based classifier and other diverse single-classifiers. Finally, an optimum ensemble classifier is proposed by using multiple objective particle swarm optimization (MOPSO) and considering accuracy and reliability as objective functions. The experimental results over a benchmark PolSAR image demonstrate the effectiveness of the proposed algorithms compared to the existing techniques.
کلیدواژهها [English]
)