这里写目录标题
Abstract
Following the great success of deep convolutional
neural networks (CNNs) in computer vision, this paper proposes a complex-valued CNN (CV-CNN) specifically for synthetic aperture radar (SAR) image interpretation. It utilizes both amplitude and phase information of complex SAR imagery. All elements of CNN including input–output layer, convolution layer, activation function, and pooling layer are extended to the complex domain. Moreover, a complex backpropagation algorithm based on stochastic gradient descent is derived for CV-CNN training. The proposed CV-CNN is then tested on the typical polarimetric SAR image classification task which classifies each pixel into known terrain types via supervised training. Experiments with the benchmark data sets of Flevoland and Oberpfaffenhofen show that the classification error can be further reduced if employing CV-CNN instead of conventional real-valued CNN with the same degrees of freedom. The performance of CV-CNN is comparable to that of existing state-of-the-art methods in terms of overall classification accuracy.
Method
This paper first presents a CV-CNN, which not only takes complex data as input, but also propagates the phase information through all layers. It is necessary because the entire hierarchic representation of SAR image has to be constructed using CV data. The proposed CV-CNN extends the entire network including both data and parameters into the complex domain, which means that all mathematical operations are extended under complex analysis theory. It also includes a complex backpropagation algorithm for network training. The performance of CV-CNN when applied to POLSAR classification is evaluated with several benchmark data sets and compared with conventional real-valued counterparts with the same configurations. Significant improvements in terms of classification accuracy are achieved.
Results
CNN has achieved great success in computer vision areas.
This study aims to apply CNN to SAR image processing. In order to take advantage of phase information, we present the CV-CNN which extends RV-CNN to complex domain. Both neurons and weights are represented by complex numbers. A complex version of backpropagation is also proposed, which can effectively train deep CV-CNN with massive complex SAR images. A novel POLSAR image classification scheme based on CV-CNN is then proposed and evaluated with real SAR data. Results show that a significant improvement in terms of classification accuracy can be achieved by converting RV-CNN to CV-CNN. It opens up a wide range of applications for CV-CNN in SAR image interpretation.
Note
NOTE:This paper is originated from《Complex-Valued Convolutional Neural Network and Its Application in Polarimetric SAR Image Classification》