Resource-Constrained Simultaneous Detection and Labeling of Objects in High-Resolution Satellite Images
This work addresses resource-limited object detection in satellite imagery, which is incremental as it builds on existing methods like CNNs and single shot detectors for a specific domain.
The paper tackles the problem of detecting and classifying man-made objects in high-resolution satellite images under computational constraints, achieving a third-place ranking on the xView challenge with a method using parallel CNN pipelines and improved region merging.
We describe a strategy for detection and classification of man-made objects in large high-resolution satellite photos under computational resource constraints. We detect and classify candidate objects by using five pipelines of convolutional neural network processing (CNN), run in parallel. Each pipeline has its own unique strategy for fine tunning parameters, proposal region filtering, and dealing with image scales. The conflicting region proposals are merged based on region confidence and not just based on overlap areas, which improves the quality of the final bounding-box regions selected. We demonstrate this strategy using the recent xView challenge, which is a complex benchmark with more than 1,100 high-resolution images, spanning 800,000 aerial objects around the world covering a total area of 1,400 square kilometers at 0.3 meter ground sample distance. To tackle the resource-constrained problem posed by the xView challenge, where inferences are restricted to be on CPU with 8GB memory limit, we used lightweight CNN's trained with the single shot detector algorithm. Our approach was competitive on sequestered sets; it was ranked third.