Ship Detection: Parameter Server Variant

TL;DR

This paper compares single node and parameter server architectures for deep learning ship detection in satellite imagery, showing the parameter server approach improves class accuracy on challenging datasets.

Contribution

It introduces a parameter server variant for ship detection that enhances accuracy and provides a systematic performance comparison with traditional single node methods.

Findings

Parameter server variant achieves 73% accuracy on target dataset.

Custom U-Net reaches 92% accuracy on validation dataset.

Systematic comparison supports the effectiveness of the parameter server approach.

Abstract

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.