Oil Spill Identification from SAR Images for Low Power Embedded Systems Using CNN

Oil spills represent one of the major threats to marine ecosystems. Satellite synthetic-aperture radar (SAR) sensors have been widely used to identify oil spills due to their ability to provide high resolution images during day and night under all weather conditions. In recent years, the use of artificial intelligence (AI) systems, especially convolutional neural networks (CNNs), have led to many important improvements in performing this task. However, most of the previous solutions to this problem have focused on obtaining the best performance under the assumption that there are no constraints on the amount of hardware resources being used. For this reason, the amounts of hardware resources such as memory and power consumption required by previous solutions make them unsuitable for remote embedded systems such as nano and micro-satellites, which usually have very limited hardware capability and very strict limits on power consumption. In this paper, we present a CNN architecture for semantically segmenting SAR images into multiple classes. The proposed CNN is specifically designed to run on remote embedded systems, which have very limited hardware capability and strict limits on power consumption. Even if the performance in terms of results accuracy does not represent a step forward compared with previous solutions, the presented CNN has the important advantage of being able to run on remote embedded systems with limited hardware resources while achieving good performance. The presented CNN is compatible with dedicated hardware accelerators available on the market due to its low memory footprint and small size. It also provides many additional very significant advantages, such as having shorter inference times, requiring shorter training times, and avoiding transmission of irrelevant data. Our goal is to allow embedded low power remote devices such as satellite systems for remote sensing to be able to directly run CNNs on board, so that the amount of data that needs to be transmitted to ground and processed on ground can be substantially reduced, which will be greatly beneficial in significantly reducing the amount of time needed for identification of oil spills from SAR images.

[1]  Jing Li,et al.  A robot vision navigation method using deep learning in edge computing environment , 2021, EURASIP J. Adv. Signal Process..

[2]  Xiaofeng Li,et al.  Ocean Oil Spill Classification with RADARSAT-2 SAR Based on an Optimized Wavelet Neural Network , 2017, Remote. Sens..

[3]  Ayman El-Baz,et al.  A Deep-Learning Framework for the Detection of Oil Spills from SAR Data , 2021, Sensors.

[4]  Jan Dirk Wegner,et al.  Crop mapping from image time series: deep learning with multi-scale label hierarchies , 2021, ArXiv.

[5]  M. Baze,et al.  A digital CMOS design technique for SEU hardening , 2000 .

[6]  Robert B. Fisher,et al.  Segmentation of Oil Spills on Side-Looking Airborne Radar Imagery with Autoencoders , 2018, Sensors.

[7]  Jian Li,et al.  Oil Spill Detection in Quad-Polarimetric SAR Images Using an Advanced Convolutional Neural Network Based on SuperPixel Model , 2020, Remote. Sens..

[8]  Fabio Del Frate,et al.  Neural networks for oil spill detection using ERS-SAR data , 2000, IEEE Trans. Geosci. Remote. Sens..

[9]  Aubrey K. Dunne,et al.  Eyes of Things , 2017, 2017 IEEE International Conference on Cloud Engineering (IC2E).

[10]  Luca Fanucci,et al.  An FPGA-Based Hardware Accelerator for CNNs Using On-Chip Memories Only: Design and Benchmarking with Intel Movidius Neural Compute Stick , 2019, Int. J. Reconfigurable Comput..

[11]  Filippo Maria Bianchi,et al.  Large-scale detection and categorization of oil spills from SAR images with deep learning , 2020, Remote. Sens..

[12]  Yiannis Kompatsiaris,et al.  Oil Spill Identification from Satellite Images Using Deep Neural Networks , 2019, Remote. Sens..

[13]  Khamaruzaman Wan Yusof,et al.  A novel deep learning instance segmentation model for automated marine oil spill detection , 2020 .

[14]  Yixiao Wang,et al.  A Deep Convolutional Neural Network for Oil Spill Detection from Spaceborne SAR Images , 2020, Remote. Sens..

[15]  K. Topouzelis,et al.  Detection and discrimination between oil spills and look-alike phenomena through neural networks , 2007 .

[16]  Tiago Gomes,et al.  Lock-V: A heterogeneous fault tolerance architecture based on Arm and RISC-V , 2021 .

[17]  Luca Fanucci,et al.  CloudScout: A Deep Neural Network for On-Board Cloud Detection on Hyperspectral Images , 2020, Remote. Sens..

[18]  Peng Liu,et al.  A survey of remote-sensing big data , 2015, Front. Environ. Sci..

[19]  Ting Liu,et al.  Recent advances in convolutional neural networks , 2015, Pattern Recognit..

[20]  Chen Feng,et al.  A Deep Learning Fusion Recognition Method Based On SAR Image Data , 2019 .

[21]  Guo Wei,et al.  Dark Spot Detection in SAR Images of Oil Spill Using Segnet , 2018, Applied Sciences.

[22]  George Papandreou,et al.  Rethinking Atrous Convolution for Semantic Image Segmentation , 2017, ArXiv.

[23]  Zeyu Jiao,et al.  A new approach to oil spill detection that combines deep learning with unmanned aerial vehicles , 2019, Comput. Ind. Eng..

[24]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[25]  Carlos Dafonte,et al.  Oil spill segmentation in SAR images using convolutional neural networks. A comparative analysis with clustering and logistic regression algorithms , 2019, Appl. Soft Comput..

[26]  Vladlen Koltun,et al.  Multi-Scale Context Aggregation by Dilated Convolutions , 2015, ICLR.

[27]  Jubai An,et al.  Discrimination of Oil Slicks and Lookalikes in Polarimetric SAR Images Using CNN , 2017, Sensors.

[28]  Guangmin Sun,et al.  Application of Deep Networks to Oil Spill Detection Using Polarimetric Synthetic Aperture Radar Images , 2017 .

[29]  Ying Li,et al.  A Spectral Feature Based Convolutional Neural Network for Classification of Sea Surface Oil Spill , 2019, ISPRS Int. J. Geo Inf..

[30]  Konstantinos Topouzelis,et al.  Oil spill feature selection and classification using decision tree forest on SAR image data , 2012 .

[31]  Luca Fanucci,et al.  An FPGA-Based Hardware Accelerator for CNNs Inference on Board Satellites: Benchmarking with Myriad 2-Based Solution for the CloudScout Case Study , 2021, Remote. Sens..