Ship Detection: Parameter Server Variant
This work provides an incremental improvement in ship detection accuracy for remote sensing applications, which is important for maritime surveillance and logistics.
This paper addresses the problem of ship detection in satellite optical imagery, which is challenged by false positives from clouds, landmasses, and man-made objects. The authors developed a custom U-Net that achieved 92% class accuracy on a validation dataset and 68% on a target dataset, improving upon typical scores of 88%. A parameter server variant further boosted performance on the target dataset to 73% class accuracy.
Deep learning ship detection in satellite optical imagery suffers from false positive occurrences with clouds, landmasses, and man-made objects that interfere with correct classification of ships, typically limiting class accuracy scores to 88\%. This work explores the tensions between customization strategies, class accuracy rates, training times, and costs in cloud based solutions. We demonstrate how a custom U-Net can achieve 92\% class accuracy over a validation dataset and 68\% over a target dataset with 90\% confidence. We also compare a single node architecture with a parameter server variant whose workers act as a boosting mechanism. The parameter server variant outperforms class accuracy on the target dataset reaching 73\% class accuracy compared to the best single node approach. A comparative investigation on the systematic performance of the single node and parameter server variant architectures is discussed with support from empirical findings.