Image Processing Scientist - Geometric

Role Description

We are seeking a Image Processing Scientist - Geometric to lead the design, development, and validation of geometric calibration and orthorectification pipelines for high-resolution and hyperspectral satellite imagery. You will be responsible for defining and implementing robust camera models, attitude reconstruction algorithms, and metadata correction techniques to ensure sub-pixel geolocation accuracy across single- and multi-detector focal plane arrays .

Responsibilities Duties

Geometric Calibration Orthorectification

• Develop and maintain Python-based libraries for camera intrinsics/extrinsics estimation, image-to-ground and ground-to-image transforms

• Implement orthorectification workflows, integrating DEM sources (e.g., Copernicus) and projection handling

• Design algorithms to reconstruct and correct pushbroom metadata (e.g., attitude, first_line_time, ephemeris) for congruent multi-band alignment

Sensor Characterization Accuracy Assessment

• Calibrate focal-plane arrays on orbiting satellites (single-detector and multi-detector architectures) to extract focal length, optical distortion, and mounting offsets

• Quantify and monitor pointing and geolocation errors (CE90, LE90) using GCPs, truth ortho references (e.g., Sentinel-2), and scatter-plot analysis

• Lead periodic assessments of geolocation performance and drive corrective action in attitude/orbit determination subsystems

Pipeline Development Automation

Architect end-to-end processing pipelines for pushbroom sensor data: metadata ingestion, NUC application, orthorectification, packaging as GeoTIFFs

Integrate external reference data services (e.g., Sentinel-2 archives) for truth comparisons and validation

Collaborate with DevOps to containerize and deploy processing services at scale

Algorithm Research Innovation

Invent and prototype new registration techniques (e.g., multi-channel registration, shear-based band alignment, scale correction)

Evaluate state-of-the-art photogrammetric and computer-vision methods to enhance geometric fidelity

Publish and present technical findings at industry conferences

Cross-Functional Collaboration Mentorship

Work closely with radiometric scientists, data engineers, and mission planners to align geometric requirements with end-user products

Mentor junior engineers in best practices for camera modeling, metadata handling, and quality assessment

Contribute to technical design reviews, documentation, and training materials

Required Skills:

• Ph.D. or M.S. in Photogrammetry, Remote Sensing, Geodesy, Physics, Computer Vision, or related field and hands-on experience in geometric calibration and orthorectification of satellite or aerial imagery

• Proficient in Python (experience with NumPy, SciPy, Rasterio/GDAL)

• Deep understanding of camera models (pinhole, pushbroom), sensor geometry, and projection systems (UTM, WGS84, custom CRS)

• Proven track record of implementing attitude reconstruction algorithms and metadata correction workflows

• Expertise in geolocation accuracy metrics (CE90, GCP workflows) and DEM integration for terrain correction

• Strong analytical and problem-solving skills, with ability to validate complex algorithms quantitatively

Preferred Skills:

• Experience with hyperspectral sensor architectures and multi-detector focal plane arrays

• Familiarity with CEOS/OGC standards, RadCalNet sites, and ARD compliance requirements
• Background in developing automated, containerized processing pipelines (Docker, Kubernetes)
• History of conference presentations or peer-reviewed publications in photogrammetry
• Excellent communication skills and demonstrated ability to mentor cross-disciplinary teams

Candidate Acumen:

• A strong desire to work in an unstructured, high-growth, fast-paced start-up environment

Benefits:

• Health insurance coverage
• Unlimited leaves flexible working hours
• Role-based remote work and work-from-home benefit
• Relocation assistance
• Professional Mental Wellness services
• Creche facility for primary caregivers (limited to India)
• Employee Stock Options for all hires