High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural Network
With the development of artificial intelligence, Synthetic-Aperture Radar (SAR) ship detection using deep learning technology can effectively avoid traditionally complex feature design and thereby greatly improve detection accuracy. However, most existing detection models often improve detection acc...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
China Science Publishing & Media Ltd. (CSPM)
2019-12-01
|
| Series: | Leida xuebao |
| Subjects: | |
| Online Access: | http://radars.ie.ac.cn/article/doi/10.12000/JR19111?viewType=HTML |
| _version_ | 1827612774039027712 |
|---|---|
| author | ZHANG Xiaoling ZHANG Tianwen SHI Jun WEI Shunjun |
| author_facet | ZHANG Xiaoling ZHANG Tianwen SHI Jun WEI Shunjun |
| author_sort | ZHANG Xiaoling |
| collection | DOAJ |
| description | With the development of artificial intelligence, Synthetic-Aperture Radar (SAR) ship detection using deep learning technology can effectively avoid traditionally complex feature design and thereby greatly improve detection accuracy. However, most existing detection models often improve detection accuracy at the expense of detection speed that limits some real-time applications of SAR such as emergency military deployment, rapid maritime rescue, and real-time marine environmental monitoring. To solve this problem, a high-speed and high-accuracy SAR ship detection method called SARShipNet-20 based on a Depthwise Separable Convolution Neural Network (DS-CNN) has been proposed in this paper, that replaces the Traditional Convolution Neural Network (T-CNN) and combines Channel Attention (CA) and Spatial Attention (SA). As a result, high-speed and high-accuracy SAR ship detection can be simultaneously achieved. This method has certain practical significance in the field of real-time SAR application, and its lightweight model is helpful for future FPGA or DSP hardware transplantation. |
| first_indexed | 2024-03-09T08:29:10Z |
| format | Article |
| id | doaj.art-cf7e9ffb87944019bd06a99c4e5b3986 |
| institution | Directory Open Access Journal |
| issn | 2095-283X 2095-283X |
| language | English |
| last_indexed | 2024-03-09T08:29:10Z |
| publishDate | 2019-12-01 |
| publisher | China Science Publishing & Media Ltd. (CSPM) |
| record_format | Article |
| series | Leida xuebao |
| spelling | doaj.art-cf7e9ffb87944019bd06a99c4e5b39862023-12-02T20:27:50ZengChina Science Publishing & Media Ltd. (CSPM)Leida xuebao2095-283X2095-283X2019-12-018684185110.12000/JR19111High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural NetworkZHANG Xiaoling0ZHANG Tianwen1SHI Jun2WEI Shunjun3(School of Electronic Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China)(School of Electronic Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China)(School of Electronic Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China)(School of Electronic Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China)With the development of artificial intelligence, Synthetic-Aperture Radar (SAR) ship detection using deep learning technology can effectively avoid traditionally complex feature design and thereby greatly improve detection accuracy. However, most existing detection models often improve detection accuracy at the expense of detection speed that limits some real-time applications of SAR such as emergency military deployment, rapid maritime rescue, and real-time marine environmental monitoring. To solve this problem, a high-speed and high-accuracy SAR ship detection method called SARShipNet-20 based on a Depthwise Separable Convolution Neural Network (DS-CNN) has been proposed in this paper, that replaces the Traditional Convolution Neural Network (T-CNN) and combines Channel Attention (CA) and Spatial Attention (SA). As a result, high-speed and high-accuracy SAR ship detection can be simultaneously achieved. This method has certain practical significance in the field of real-time SAR application, and its lightweight model is helpful for future FPGA or DSP hardware transplantation.http://radars.ie.ac.cn/article/doi/10.12000/JR19111?viewType=HTMLconvolution neural network (cnn)depthwise separable convolution neural network (ds-cnn)sarship detectionattention mechanism |
| spellingShingle | ZHANG Xiaoling ZHANG Tianwen SHI Jun WEI Shunjun High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural Network Leida xuebao convolution neural network (cnn) depthwise separable convolution neural network (ds-cnn) sar ship detection attention mechanism |
| title | High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural Network |
| title_full | High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural Network |
| title_fullStr | High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural Network |
| title_full_unstemmed | High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural Network |
| title_short | High-speed and High-accurate SAR Ship Detection Based on a Depthwise Separable Convolution Neural Network |
| title_sort | high speed and high accurate sar ship detection based on a depthwise separable convolution neural network |
| topic | convolution neural network (cnn) depthwise separable convolution neural network (ds-cnn) sar ship detection attention mechanism |
| url | http://radars.ie.ac.cn/article/doi/10.12000/JR19111?viewType=HTML |
| work_keys_str_mv | AT zhangxiaoling highspeedandhighaccuratesarshipdetectionbasedonadepthwiseseparableconvolutionneuralnetwork AT zhangtianwen highspeedandhighaccuratesarshipdetectionbasedonadepthwiseseparableconvolutionneuralnetwork AT shijun highspeedandhighaccuratesarshipdetectionbasedonadepthwiseseparableconvolutionneuralnetwork AT weishunjun highspeedandhighaccuratesarshipdetectionbasedonadepthwiseseparableconvolutionneuralnetwork |