Expert Level Imagery Scientist (SAR)

• 6+ years as a SAR expert with understanding of collection, phenomenology, image formation process, and exploitation products.\n
• Bachelor’s Degree in related field. Additional years of experience can be considered in lieu of degree\n
• Exhibit experience using SARPY and Matlab SAR toolbox\n
• Experience in RNIIRS, information theoretic-based image quality metrics, SAR imagery quality metrics (e.g., Integrated Sidelobe Ratio, Multiplicative Noise Ratio), and sensor metadata describing impacts of geometry on phenomenology (e.g., graze, squint, azimuth)\n
• Exhibit experience exploiting SAR to determine the occurrence and location of objects of interest\n
• Exhibit experience developing, testing, and evaluating new algorithms, processes, methodologies, and products using SAR imagery including experience utilizing advanced processing tools (e.g., Python, MATLAB, Google Earth Engine, etc.) to automate scientific processes of SAR imagery data to aid analysis\n
• Exhibit experience communicating with a variety of technical and non-technical audiences on availability and capabilities of SAR imagery products, methodologies, procedures, and algorithms to enhance analysis.\n
• Exhibit a deep understanding of the principles of remote sensing and imagery processing and advanced exploitation methods\n
• Experience with Synthetic Aperture Radar imagery analytical products and conducting multi-INT analysis\n
• Utilize structured, unstructured, and semi-structured data and visualization tools to exploit data through use of programming and scripting, including advanced geospatial skills and an excellent understanding of how to apply evolving technology and methodologies to related issues\n
• Active TS/SCI clearance with ability to be approved for a Poly\n, * Exhibit experience applying CV and machine learning (ML) techniques to SAR imagery and data to address intelligence problems\n

Leidos has an exciting opportunity for an Expert Level Imagery Scientist (SAR) - to join our team in Alexandria, VA.\n \n The SAR Geospatial Analyst shall provide geospatial and imagery expertise and quantitative analysis to make recommendations that improve data curation and development in support of Machine Learning algorithm testing and evaluation.\n \n The Imagery Scientist is the subject matter expert on their respective imagery modality (e.g., Synthetic Aperture Radar). They shall provide technical direction and conduct the work necessary to acquire and prepare imagery of the necessary quality, standards, and requirements provided by the Government. Developed solutions should be informed by specific phenomenology limitations and advantages of the sensors and platforms in mind.\n \n \nPrimary Responsibilities\n \n \n

• Conduct assessment of potential differences between new sensor characteristics and capabilities compared to currently utilized platforms\n
• Assess potential differences in metadata, data format, and data structure characteristics in regards to changes to databases, schemas, APIs, and other ETL related processes for the ingestion and movement of data when integrating into the existing data operations pipeline\n
• Determine how to acquire new data, potential latency associated with acquisition, data formats, and security domains\n
• Determine how to pre-process and standardize the data to match existing data standards or to be transformed into a usable state for labeling and model testing purposes. This may involve converting between file format types or tiling full-size images into specified sizes or geospatial bounds.\n
• Investigate any gaps in the emerging sensor capability that may need to be supplemented by other sources\n
• Explore options for coincident imagery collects from other imagery platforms (e.g., EO platfoms) that align with areas of emerging sensor’s collection. Determine other platforms with similar geographic and temporal coverage.\n
• Leverage multi-INT data, with an emphasis on SAR, to curate imagery from the emergent sensors. Curated imagery will be prioritized for creation of labeled data to support the model evaluation and accreditation. Curated imagery should have objects of interest or needed characteristics (e.g., geographic coverage, scene attributes) per Government prioritization.\n

\n \u2022 Based on the characteristics and limitations of the emergent sensor, recommend curation strategies and identify data gaps to develop a diverse, representative dataset for Machine Learning model evaluation\n \u2022 Assess potential differences in metadata, data format, and data structure characteristics in regard to changes to databases, schemas, APIs, and other ETL related processes for the ingestion and movement of data when integrating into the existing data operations pipeline\n \u2022 Determine how to acquire new data, potential latency associated with acquisition,\n \n