scholarly journals Nonlocal CNN SAR Image Despeckling

2020 ◽  
Vol 12 (6) ◽  
pp. 1006 ◽  
Author(s):  
Davide Cozzolino ◽  
Luisa Verdoliva ◽  
Giuseppe Scarpa ◽  
Giovanni Poggi
Keyword(s):  
Deep Learning ◽  
Weighted Average ◽  
Sar Image ◽  
Sar Images ◽  
Nonlocal Means ◽  
Analysis Window ◽  
Account Information ◽  
Measure Of Similarity ◽  
Image Despeckling ◽  
Insight Into

We propose a new method for SAR image despeckling, which performs nonlocal filtering with a deep learning engine. Nonlocal filtering has proven very effective for SAR despeckling. The key idea is to exploit image self-similarities to estimate the hidden signal. In its simplest form, pixel-wise nonlocal means, the target pixel is estimated through a weighted average of neighbors, with weights chosen on the basis of a patch-wise measure of similarity. Here, we keep the very same structure of plain nonlocal means, to ensure interpretability of results, but use a convolutional neural network to assign weights to estimators. Suitable nonlocal layers are used in the network to take into account information in a large analysis window. Experiments on both simulated and real-world SAR images show that the proposed method exhibits state-of-the-art performance. In addition, the comparison of weights generated by conventional and deep learning-based nonlocal means provides new insight into the potential and limits of nonlocal information for SAR despeckling.

scholarly journals An Advanced SAR Image Despeckling Method by Bernoulli-Sampling-Based Self-Supervised Deep Learning

2021 ◽  
Vol 13 (18) ◽  
pp. 3636
Author(s):  
Ye Yuan ◽  
Yanxia Wu ◽  
Yan Fu ◽  
Yulei Wu ◽  
Lidan Zhang ◽  
...  
Keyword(s):  
Big Data ◽  
Deep Learning ◽  
Noise Suppression ◽  
Speckle Noise ◽  
Training Data ◽  
Superior Performance ◽  
Sar Image ◽  
Sar Images ◽  
Image Despeckling ◽  
Image Pairs

As one of the main sources of remote sensing big data, synthetic aperture radar (SAR) can provide all-day and all-weather Earth image acquisition. However, speckle noise in SAR images brings a notable limitation for its big data applications, including image analysis and interpretation. Deep learning has been demonstrated as an advanced method and technology for SAR image despeckling. Most existing deep-learning-based methods adopt supervised learning and use synthetic speckled images to train the despeckling networks. This is because they need clean images as the references, and it is hard to obtain purely clean SAR images in real-world conditions. However, significant differences between synthetic speckled and real SAR images cause the domain gap problem. In other words, they cannot show superior performance for despeckling real SAR images as they do for synthetic speckled images. Inspired by recent studies on self-supervised denoising, we propose an advanced SAR image despeckling method by virtue of Bernoulli-sampling-based self-supervised deep learning, called SSD-SAR-BS. By only using real speckled SAR images, Bernoulli-sampled speckled image pairs (input–target) were obtained as the training data. Then, a multiscale despeckling network was trained on these image pairs. In addition, a dropout-based ensemble was introduced to boost the network performance. Extensive experimental results demonstrated that our proposed method outperforms the state-of-the-art for speckle noise suppression on both synthetic speckled and real SAR datasets (i.e., Sentinel-1 and TerraSAR-X).

scholarly journals deSpeckNet: Generalizing Deep Learning-Based SAR Image Despeckling

2020 ◽  
pp. 1-15
Author(s):  
Adugna G. Mullissa ◽  
Diego Marcos ◽  
Devis Tuia ◽  
Martin Herold ◽  
Johannes Reiche
Keyword(s):  
Deep Learning ◽  
Sar Image ◽  
Image Despeckling

scholarly journals Two-Phase Object-Based Deep Learning for Multi-Temporal SAR Image Change Detection

2020 ◽  
Vol 12 (3) ◽  
pp. 548 ◽  
Author(s):  
Xinzheng Zhang ◽  
Guo Liu ◽  
Ce Zhang ◽  
Peter M. Atkinson ◽  
Xiaoheng Tan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Change Detection ◽  
Speckle Noise ◽  
Sar Image ◽  
Phase Object ◽  
Sar Images ◽  
Two Phase ◽  
Object Based ◽  
Multi Temporal ◽  
Image Change Detection

Change detection is one of the fundamental applications of synthetic aperture radar (SAR) images. However, speckle noise presented in SAR images has a negative effect on change detection, leading to frequent false alarms in the mapping products. In this research, a novel two-phase object-based deep learning approach is proposed for multi-temporal SAR image change detection. Compared with traditional methods, the proposed approach brings two main innovations. One is to classify all pixels into three categories rather than two categories: unchanged pixels, changed pixels caused by strong speckle (false changes), and changed pixels formed by real terrain variation (real changes). The other is to group neighbouring pixels into superpixel objects such as to exploit local spatial context. Two phases are designed in the methodology: (1) Generate objects based on the simple linear iterative clustering (SLIC) algorithm, and discriminate these objects into changed and unchanged classes using fuzzy c-means (FCM) clustering and a deep PCANet. The prediction of this Phase is the set of changed and unchanged superpixels. (2) Deep learning on the pixel sets over the changed superpixels only, obtained in the first phase, to discriminate real changes from false changes. SLIC is employed again to achieve new superpixels in the second phase. Low rank and sparse decomposition are applied to these new superpixels to suppress speckle noise significantly. A further clustering step is applied to these new superpixels via FCM. A new PCANet is then trained to classify two kinds of changed superpixels to achieve the final change maps. Numerical experiments demonstrate that, compared with benchmark methods, the proposed approach can distinguish real changes from false changes effectively with significantly reduced false alarm rates, and achieve up to 99.71% change detection accuracy using multi-temporal SAR imagery.

Impact of Method Noise on SAR Image Despeckling

2020 ◽  
Vol 15 (1) ◽  
pp. 52-63 ◽  
Author(s):  
Prabhishek Singh ◽  
Raj Shree
Keyword(s):  
Performance Metrics ◽  
Synthetic Aperture ◽  
Sar Image ◽  
Sar Images ◽  
Standard Methods ◽  
Residual Image ◽  
Method Noise ◽  
Image Despeckling ◽  
The Difference ◽  
And Performance

This article introduces the concept, use and implementation of method noise in the field of synthetic aperture radar (SAR) image despeckling. Method noise has the capability to enhance the efficiency and performance of any despeckling algorithm. It is easy, efficient and enhanced way of improving the results. The difference between speckled image and despeckled image contains some residual image information which is due to the inefficiency of the denoising algorithm. This article will compare the results of some standard methods with and without the use of method noise and prove its efficiency and validity. It also shows its best use in different ways of denoising. The results will be compared on the basis of performance metrics like PSNR and SSIM. The concept of method noise is not restricted to only SAR images. It has vast usage and application. It can be used in any denoising procedure such as medical images, optical image etc. but this paper shows the experimental results only on the SAR images.

scholarly journals SPECKLE NOISE REDUCTION IN SAR IMAGES USING INFORMATION THEORY

2020 ◽  
Vol XLII-3/W12-2020 ◽  
pp. 141-146
Author(s):  
D. Chan ◽  
J. Gambini ◽  
A. C. Frery
Keyword(s):  
Signal To Noise Ratio ◽  
Speckle Noise ◽  
Statistical Test ◽  
Signal To Noise ◽  
Sar Images ◽  
Nonlocal Means ◽  
Two Samples ◽  
The Mean ◽  
Measure Of Similarity ◽  
Speckle Noise Reduction

Abstract. In this work, a new nonlocal means filter for single-look speckled data using the Shannon and Rényi entropies is proposed. The measure of similarity between a central window and patches of the image is based on a statistical test for comparing if two samples have the same entropy and hence have the same distribution.The results are encouraging, as the filtered image has better signal-to-noise ratio, it preserves the mean, and the edges are not severely blurred.

Tapping the Potential of Earth Observation - Calving Front Detection in SAR Images using Deep Learning Techniques

2021 ◽  
Author(s):  
Nora Gourmelon ◽  
Thorsten Seehaus ◽  
AmirAbbas Davari ◽  
Matthias Braun ◽  
Andreas Maier ◽  
...  
Keyword(s):  
Deep Learning ◽  
Hydrological Cycle ◽  
Class Imbalance ◽  
Earth Observation ◽  
Learning Approaches ◽  
Sar Image ◽  
Sar Images ◽  
Distance Map ◽  
Learning Techniques ◽  
Front Detection

<p>The calving fronts of lake or marine terminating glaciers provide information about the state of glaciers. A change in its position can affect the flow of the entire glacier system, and the loss of ice mass as icebergs calve-off and discharge into the ocean has a multi-scale impact on the global hydrological cycle. The calving fronts can be manually delineated in Synthetic Aperture Radar (SAR) images. However, this is a time-consuming, tedious and expensive task. As deep learning approaches have achieved tremendous success in various disciplines, such as medical image processing and computer vision, the project Tapping the Potential of Earth Observation (TAPE) is amongst other things dedicated to applying deep learning techniques to calving front detection. So far, all our experiments have employed U-Net based architectures, as the U-Net is state-of-the-art in semantic image segmentation. A major challenge of front detection is the class imbalance: The front has significantly fewer pixels than the remaining parts of the SAR image. Hence, we developed variants of the U-Net specifically addressing this challenge including an Attention U-Net, a probabilistic Bayesian U-Net, as well as a U-Net with a distance map-based binary cross-entropy (BCE) loss function and a Mathews correlation coefficient (MCC) as early stopping criterion. In future work, we plan to investigate multi-task learning and a segmentation of the SAR image into different classes (i.e. ocean, glacier and rocks) to enhance the quality and efficiency of the front detection.</p>

Sign in / Sign up

Export Citation Format

Share Document