Convolutional Neural Networks with Different Dimensions for POLSAR Image Classification
عنوان مقاله: Convolutional Neural Networks with Different Dimensions for POLSAR Image Classification
شناسه ملی مقاله: CSCG04_123
منتشر شده در چهارمین کنفرانس بین المللی محاسبات نرم در سال 1400
شناسه ملی مقاله: CSCG04_123
منتشر شده در چهارمین کنفرانس بین المللی محاسبات نرم در سال 1400
مشخصات نویسندگان مقاله:
Maryam Imani - Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran
خلاصه مقاله:
Maryam Imani - Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran
Polarimetric Synthetic aperture radar (PolSAR) images contain polarimetric and spatial information of materials present in the scene. Three simple architectures of convolutional neural networks (CNNs) with different dimensions are proposed for PolSAR image classification in this work. A one dimensional CNN (۱D CNN) is suggested for polarimetric feature extraction. A ۲D CNN is presented for spatial feature extraction and a ۳D CNN is introduced for polarimetric-spatial feature extraction. The performance of CNNs are compared with morphological profile of PolSAR cube when fed to the support vector machine (SVM) and random forest (RF) classifiers. The experiments are done in two cases of using ۱% and ۵% training samples. The superiority of ۳D CNN compared to other methods is shown using different quantitative classification measures.
کلمات کلیدی: PolSAR, classification, feature extraction, CNN
صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/1418632/