Picture of rolled up £5 note

Open Access research that shapes economic thinking...

Strathprints makes available scholarly Open Access content by the Fraser of Allander Institute (FAI), a leading independent economic research unit focused on the Scottish economy and based within the Department of Economics. The FAI focuses on research exploring economics and its role within sustainable growth policy, fiscal analysis, energy and climate change, labour market trends, inclusive growth and wellbeing.

The open content by FAI made available by Strathprints also includes an archive of over 40 years of papers and commentaries published in the Fraser of Allander Economic Commentary, formerly known as the Quarterly Economic Commentary. Founded in 1975, "the Commentary" is the leading publication on the Scottish economy and offers authoritative and independent analysis of the key issues of the day.

Explore Open Access research by FAI or the Department of Economics - or read papers from the Commentary archive [1975-2006] and [2007-2018]. Or explore all of Strathclyde's Open Access research...

Dual-branch deep convolution neural network for polarimetric SAR image classification

Gao, Fei and Huang, Teng and Wang, Jun and Sun, Jinping and Hussain, Amir and Yang, Erfu (2017) Dual-branch deep convolution neural network for polarimetric SAR image classification. Applied Sciences, 7 (5). ISSN 2076-3417

[img]
Preview
Text (Gao-etal-AS-2017-Dual-branch-deep-convolution-neural-network-for-polarimetric-SAR-image-classification)
Gao_etal_AS_2017_Dual_branch_deep_convolution_neural_network_for_polarimetric_SAR_image_classification.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo

Download (3MB) | Preview

Abstract

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 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 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 complex coherency matrix. The other is utilized to extract the spatial features of Pauli RGB 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 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 the state of the art methods.