Planetary Scale Location-based Insights

Abstract: The recent explosion of new data streams from varied satellite streams necessitates the use of automation techniques that allow for processing and refining raw sensor data to streams of foundational feature information for business insights. Deep learning provides new avenues for deriving insights from imagery. Recent approaches allow for feature detection and localization over varied context and different imaging conditions. Overhead satellite imagery is varied and rich, but fundamentally different from other data sources and introduces unique challenges towards scalable location insights.

We will explore the use of modern deep learning approaches to object detection, semantic segmentation towards feature extraction and meaningful change detection in satellite imagery. We identify techniques towards sampling data, leveraging heterogeneous datasets, performance benchmarking and tuning models towards generalized performance. We will walk through the technical challenges involved with serving results from these deep learning models in a scalable manner along with essential metadata for interpretability and usability for our customers. We then explore approaches with human in the loop towards improved performance over time towards scalable location-based intelligence served across the entirety of the Earth’s landmass and discuss the challenges and opportunities at the frontiers of geospatial data and location-based insights. Specific applications of the Planet Analytics API towards geospatial intelligence will be discussed including data fusion from multiple geospatial data sources.

By imaging the Earth every day at 3.7 m resolution and enabling on-demand follow up imagery at 72 cm resolution, Planet offers a uniquely valuable dataset for creating datasets for imagery analytics over varied context. We use the above approaches towards creating foundational analytic feeds serving new business applications – the technical underpinnings for Planet Analytics.

Bio: Gopal manages the Analytics Engineering team at Planet, an integrated aerospace and data analytics company that operates history's largest commercial fleet of satellites. His background is in the development of foundational technology that powers products in the imaging and machine learning space. He is known for agile engineering execution from concept to scalable high quality products. Gopal’s recent experience is in industry-leading analytics products from early concept demonstrations to multiple customers at Captricity and Harvesting, where he advised the CEO. Previously he led the algorithm engineering development of Dolby’s first imaging display product, the Emmy Award winning Dolby Professional Reference Monitor and technologies for high dynamic range video reproduction in Dolby Vision, now in the iPhone X. Gopal holds an MS in Electrical Engineering from University of Southern California and completed the Ignite Program, connecting technologists with new commercial ventures, at the Stanford Graduate School of Business.