Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNet

As a branch application of the synthetic aperture radar (SAR), coherent change detection (CCD) is commonly utilized to help differentiate specific changes from the unchanged scene. To exactly pick out manmade changes of interest from change maps, the introduction of artificial neural networks (ANNs)...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinyi Ye, Yanxing Liu, Zihan Wang, Yizhe Fan, Bingchen Zhang
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11050980/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849319104795639808
author Xinyi Ye
Yanxing Liu
Zihan Wang
Yizhe Fan
Bingchen Zhang
author_facet Xinyi Ye
Yanxing Liu
Zihan Wang
Yizhe Fan
Bingchen Zhang
author_sort Xinyi Ye
collection DOAJ
description As a branch application of the synthetic aperture radar (SAR), coherent change detection (CCD) is commonly utilized to help differentiate specific changes from the unchanged scene. To exactly pick out manmade changes of interest from change maps, the introduction of artificial neural networks (ANNs) has significantly enhanced detection performance but taken up great computational resources, making it challenging to deploy on edge devices with limited resources and a strict power budget. Spiking neural networks (SNNs) not only infer results more energy-efficiently but also have the potential to drive ultralow-power neuromorphic computers which ANNs lack. In this study, to detect coherent changes precisely and with less energy consumption, a robust SNN-based change analysis method is proposed to make it available to implement CCD tasks via the &#x201C;WenTian I&#x201D; brain-like computer. First, we adopt a Poisson spike encoding method based on fuzzy <italic>c</italic>-means clustering to preclassify pixels and translate continuous change maps into spikes. Second, we input the spike images into a novel deep SNN, multithreshold Spiking-UNet. By representing information in the form of sparse spike flows, our proposed change detection model can reduce the energy consumption of CCD while maintaining excellent detection capabilities. The corresponding experimental results validate the comparable detection effectiveness and superior energy efficiency of our SNN-based method used for SAR CCD.
format Article
id doaj-art-8a3a358336a64f76b66e96208a3a4437
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-8a3a358336a64f76b66e96208a3a44372025-08-20T03:50:38ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118171391715310.1109/JSTARS.2025.358305811050980Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNetXinyi Ye0https://orcid.org/0009-0009-0227-5482Yanxing Liu1https://orcid.org/0009-0007-8604-933XZihan Wang2https://orcid.org/0009-0007-3085-8621Yizhe Fan3https://orcid.org/0009-0005-7551-2275Bingchen Zhang4Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAs a branch application of the synthetic aperture radar (SAR), coherent change detection (CCD) is commonly utilized to help differentiate specific changes from the unchanged scene. To exactly pick out manmade changes of interest from change maps, the introduction of artificial neural networks (ANNs) has significantly enhanced detection performance but taken up great computational resources, making it challenging to deploy on edge devices with limited resources and a strict power budget. Spiking neural networks (SNNs) not only infer results more energy-efficiently but also have the potential to drive ultralow-power neuromorphic computers which ANNs lack. In this study, to detect coherent changes precisely and with less energy consumption, a robust SNN-based change analysis method is proposed to make it available to implement CCD tasks via the &#x201C;WenTian I&#x201D; brain-like computer. First, we adopt a Poisson spike encoding method based on fuzzy <italic>c</italic>-means clustering to preclassify pixels and translate continuous change maps into spikes. Second, we input the spike images into a novel deep SNN, multithreshold Spiking-UNet. By representing information in the form of sparse spike flows, our proposed change detection model can reduce the energy consumption of CCD while maintaining excellent detection capabilities. The corresponding experimental results validate the comparable detection effectiveness and superior energy efficiency of our SNN-based method used for SAR CCD.https://ieeexplore.ieee.org/document/11050980/Coherent change detection (CCD)fuzzy <italic xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">c</italic>-means (FCM) clusteringPoisson spike encodingspiking neural network (SNN)synthetic aperture radar (SAR)U-Net
spellingShingle Xinyi Ye
Yanxing Liu
Zihan Wang
Yizhe Fan
Bingchen Zhang
Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNet
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Coherent change detection (CCD)
fuzzy <italic xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">c</italic>-means (FCM) clustering
Poisson spike encoding
spiking neural network (SNN)
synthetic aperture radar (SAR)
U-Net
title Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNet
title_full Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNet
title_fullStr Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNet
title_full_unstemmed Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNet
title_short Energy-Efficient SAR Coherent Change Detection Based on Deep Multithreshold Spiking-UNet
title_sort energy efficient sar coherent change detection based on deep multithreshold spiking unet
topic Coherent change detection (CCD)
fuzzy <italic xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">c</italic>-means (FCM) clustering
Poisson spike encoding
spiking neural network (SNN)
synthetic aperture radar (SAR)
U-Net
url https://ieeexplore.ieee.org/document/11050980/
work_keys_str_mv AT xinyiye energyefficientsarcoherentchangedetectionbasedondeepmultithresholdspikingunet
AT yanxingliu energyefficientsarcoherentchangedetectionbasedondeepmultithresholdspikingunet
AT zihanwang energyefficientsarcoherentchangedetectionbasedondeepmultithresholdspikingunet
AT yizhefan energyefficientsarcoherentchangedetectionbasedondeepmultithresholdspikingunet
AT bingchenzhang energyefficientsarcoherentchangedetectionbasedondeepmultithresholdspikingunet