Sarod: Efficient End-To-End Object Detection On SAR Images With Reinforcement Learning

Abstract

Generally, object detection on Synthetic-Aperture Radar (SAR) images is known to be more challenging than that in Electro-Optical (EO) satellite images because SAR images have non-negligible speckle noise and require extensive data pre-processing. Nevertheless, object detection in SAR images is important, as SAR imagery can be obtained under severe weather and time conditions. While many recent object detection approaches on SAR imagery focus on improving detection accuracy, few studies focus on improving processing efficiency. In fact, there are significant challenges and trade-offs to achieve both high accuracy and efficiency at the same time. In this work, we introduce SAROD, a novel efficient end-to-end object detection framework on SAR images based on Reinforcement Learning (RL) to balance the tradeoffs. Our proposed model consists of two detectors, coarse and fine-grained detectors, with an RL agent, where RL has not yet been utilized for object detection on SAR imagery. Our model was evaluated on a challenging SAR imagery dataset, achieving performance comparable to state-of-the-art detectors while maintaining high efficiency of source data usage.

Ranking Ship Detection Methods Using SAR Images Based on Machine Learning and Artificial Intelligence

“…[21] Ship size, sea condition, accuracy, cost [37] Gradient explosion, robustness, speed, detection accuracy [22] Ship detection, efficiency, robustness, sea-land segmentation [38] Deep learning features, ship target, detection performance [26] Detection accuracy, false alarm rate, performance, position [39] Verification accuracy, testing accuracy, ship classification, false alarm [27] Detection rate, speed, detection accuracy, ship's target [40] Ship detection, ship size, performance, robustness [28] Real-time observation, rescue, detection accuracy, faster [41] Scene classification, ship detection, accuracy, efficiency [29] Missed detections, accuracy, densely arranged ships, scale sensitivity [42] Mean average precision, accuracy, dataset, performance [30] Multi-scene detection, false alarm, performance [43] Small targets, computational efficiency, detection performance, ship management [31] Training speed, accuracy, performance, ship detection [44] Extraction and classification of candidate regions, robustness, adaptability [32] Speed, accuracy, performance, ship detection, cost [45] Ship detection, image recognition, automatic, time [33] Lost ships, open-source, fast, cost [46] Small ships, computational efficiency, pixels, precision, classification [34] Accuracy, ship detection, mean average precision, unique [64] Processing speed, accuracy, object detection, unique [35] Detection, segmentation, accuracy, pixel level [65] Object detectors, land-ocean segmentation, performance [36] Automatic, accuracy, speed, loss function…”