Improved Deep Learning Method for Real-Time Ship Detection and Localization From SAR Image

The main problem in automatic ship detection from SAR images is false detection, mostly due to speckle presence. Therefore, we propose a new deep learning model with a novel preprocessing stage (SarNeDe method) to address this problem. We are introducing a lightweight deep learning architecture to detect and localize ships in the real-time SAR image. First, generate a three-channel SarNeDe image in the preprocessing step. Then, this SarNeDe image is used to train the model to predict the ship's position in the SAR image. We experimented on the public SAR ship detection dataset (SSDD) and Dataset of Ship Detection for Deep Learning under Complex Backgrounds (SDCD) to validate the proposed method's feasibility. We used python 3.5 for coding with the Keras framework in the NVIDIA Tesla K80 GPU hardware platform. The experimental results indicated that our proposed method's ship detection accuracy has increased with reduced false detection percentage, ship detection, SAR, deep learning, and convolution neural network (CNN).