Satellite spectral imaging for crop monitoring

Chinese Academy of Sciences CAS, Northwest Agriculture and Forestry University and Beijing Normal University are prominent research organizations in the field. Recent work at these institutions includes satellite-based crop mapping, yield prediction, dynamic maize mapping and analysis of vegetation indices. The Chinese Academy of Sciences CAS also contributed to research on crop performance in wheat and sorghum systems.

FACTS records $480,532 across 5 grants since 2020. It separately lists 4 funding rounds totaling $407,392,475. These are the recorded funding figures associated with this topic in the available data; the listed organizations and funding entries should not be read as proof that every award exclusively supports satellite spectral crop imaging.

The latest listed entry is a $233,000,000 CSIRO grant dated December 17, 2025. Other recent entries include a $159,800,000 University of Wisconsin - Madison funding round dated July 21, 2025, and a $12,092,475 University of Guelph grant dated January 22, 2025. The records do not identify investors for these entries.

Nanjing Research Institute for Agricultural Mechanization Ministry of Agriculture and Carbon Applied each hold 1 listed patent. The data identifies 2 patents in the reporting window and does not show a patent holder with a higher count than the other. These counts describe the listed patent records, not the full scope of each organization’s intellectual property.

FACTS lists 2 patents in the 24-month window, but reports no patent growth percentage. The counts are therefore too limited to establish a direction of change from the supplied figures alone. Research and publication activity is more visible in the data, with 218 publications in the last year and publication growth of 46.3%.

The field has a substantial research base: FACTS lists 123 research organizations and 14 companies, with no startups recorded. Multispectral indices, time-series analysis and machine-learning prediction are established approaches, while recent studies continue to test earlier yield estimates, sensor fusion and field-scale applications. This points to an active research-to-deployment field rather than a uniform, fully mature product category.

In September 2026, Virginia Tech researchers reported a phenology-adaptive corn yield mapping method combining Sentinel-2 spectral imagery and accumulated heat units. That month, Kuva Space announced a completed pilot using hyperspectral satellites and AI for crop monitoring. Recent research also explores combining satellite imagery with radar, climate data and ground measurements to improve crop classification, stress detection and yield estimates.

FACTS identifies 5 approaches: multispectral indices, hyperspectral analysis, crop mapping and classification, time-series monitoring, and machine-learning prediction. Indices provide compact crop-condition signals; hyperspectral analysis captures finer spectral detail. Mapping classifies crops, time series follows seasonal change, and machine learning estimates outcomes from combined inputs. The main trade-offs are signal detail and model complexity versus calibration, cloud gaps and generalization.

The United States leads the recorded activity with 506 documents, followed by China with 398 and India with 245. Their respective shares are 32.1%, 25.3% and 15.6%. The United Kingdom and Germany also have substantial activity, with 155 and 141 documents. These figures reflect the country distribution in FACTS.

FACTS identifies food security monitoring, crop stress and damage assessment, agricultural statistics, precision farm management, insurance and risk assessment, and crop production forecasting. Food security monitoring has 203 associated documents, while crop stress and damage assessment has 99. The applications range from observing crop condition to supporting planning and risk decisions; forecast performance still depends on reliable local data and validation.