Share Email Print
cover

Proceedings Paper

SRNet: a cascade network to speckle reduction of SAR image
Author(s): Konghuai Shen; Xinglong Wang; Long Huang; Zhe Hu; Weidong Yang Jr.
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Speckle noise limits the usage of synthetic aperture radar (SAR) for object recognition and segmentation tasks. Most traditional methods sacrifice useful image information to achieve speckle reduction. The classic method based on local sliding window filtering has obvious side effect of erasing object edges and blurring texture information comparing with ground truth image. Another widely used method is convolutional neural network based on mean squared error, the visual effect of denoised image is not satisfactory even though MSE loss can have higher peak signal-to-noise ratio (PSNR) performance. In this paper, we present a cascade network to address this problem, namely SRNet, which employs an asymmetric architecture for the task of speckle noise reduction. The cascade architecture can supervise the network to revise on both pixel-wise level and feature-wise level by calculating correlation coefficient loss on the feature maps. In the meanwhile, we utilize the auxiliary loss on the intermediate results to accelerate the convergence of the network. The proposed network preserves the edge texture details much better than other compared methods.

Paper Details

Date Published: 14 February 2020
PDF: 7 pages
Proc. SPIE 11428, MIPPR 2019: Multispectral Image Acquisition, Processing, and Analysis, 114280O (14 February 2020); doi: 10.1117/12.2539382
Show Author Affiliations
Konghuai Shen, Huazhong Univ. of Science and Technology (China)
Xinglong Wang, System Design Institute of Hubei Aerospace Technology Academy (China)
Long Huang, System Design Institute of Hubei Aerospace Technology Academy (China)
Zhe Hu, System Design Institute of Hubei Aerospace Technology Academy (China)
Weidong Yang Jr., Huazhong Univ. of Science and Technology (China)


Published in SPIE Proceedings Vol. 11428:
MIPPR 2019: Multispectral Image Acquisition, Processing, and Analysis
Xinyu Zhang; Chao Pan; Hongshi Sang, Editor(s)

© SPIE. Terms of Use
Back to Top
PREMIUM CONTENT
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?
close_icon_gray