A Comparison of Deep Learning Object Detection Models for Satellite Imagery
This work addresses the problem of selecting efficient object detection models for satellite imagery analysis, particularly for oil and gas fracking wells and small cars, but it is incremental as it compares existing methods without introducing new techniques.
The paper compared deep learning object detection models for satellite imagery, finding that single-stage detectors matched detection performance for fracking well pads with superior speed, while two-stage and multi-stage models provided higher accuracies for small cars at a speed cost.
In this work, we compare the detection accuracy and speed of several state-of-the-art models for the task of detecting oil and gas fracking wells and small cars in commercial electro-optical satellite imagery. Several models are studied from the single-stage, two-stage, and multi-stage object detection families of techniques. For the detection of fracking well pads (50m - 250m), we find single-stage detectors provide superior prediction speed while also matching detection performance of their two and multi-stage counterparts. However, for detecting small cars, two-stage and multi-stage models provide substantially higher accuracies at the cost of some speed. We also measure timing results of the sliding window object detection algorithm to provide a baseline for comparison. Some of these models have been incorporated into the Lockheed Martin Globally-Scalable Automated Target Recognition (GATR) framework.