Abstract
The deep convolution neural network (CNN), which has prominent advantages in feature learning, can learn and extract features from data automatically. Existing polarimetric synthetic aperture radar (PolSAR) image classification methods based on the CNN only consider the polarization information of the image, instead of incorporating the image’s spatial information. In this paper, a novel method based on a dual-branch deep convolution neural network (Dual-CNN) is proposed to realize the classification of PolSAR images. The proposed method is built on two deep CNNs: one is used to extract the polarization features from the 6-channel real matrix (6Ch) which is derived from the complex coherency matrix. The other is utilized to extract the spatial features of a Pauli RGB (Red Green Blue) image. These extracted features are first combined into a fully connected layer sharing the polarization and spatial property. Then, the Softmax classifier is employed to classify these features. The experiments are conducted on the Airborne Synthetic Aperture Radar (AIRSAR) data of Flevoland and the results show that the classification accuracy on 14 types of land cover is up to 98.56%. Such results are promising in comparison with other state-of-the-art methods.
Highlights
Polarimetric synthetic aperture radar (PolSAR) is a kind of high resolution imaging system, which can work under all weather, day-and-night conditions
Aperture Radar (AIRSAR) data of Flevoland and the results show that the classification accuracy on 14 types of land cover is up to 98.56%
To solve the above problems, this paper proposes a dual-branch deep convolution neural network (Dual-CNN) method for PolSAR image classification
Summary
Polarimetric synthetic aperture radar (PolSAR) is a kind of high resolution imaging system, which can work under all weather, day-and-night conditions. Zhou et al converted the complex matrix of the PolSAR image into a real matrix of six channels to suit the input of the neural network, and designed two, cascaded, fully connected networks to map the features to a certain classifier [15] This algorithm further improves the accuracy of PolSAR image classification. The proposed method is composed of two CNNs: one is used to extract the polarization features of the real matrix of the six channels (6Ch-CNN) and the other is used to extract the spatial features of the Pauli RGB image (PauliRGB-CNN) These two kinds of features are fed into a fully connected layer to achieve mutual harmony, and the Softmax classifier is followed immediately to complete the classification work.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have