Event Date
Tech-savvy growers use field imagery and plot-level data to build predictive models to help them in planning and decision-making, but these methods come with challenges. Physically visiting the fields to collect data and manually labeling field imagery take time and resources.
Join the AI Institute for Next Generation Food Systems (AIFS) on the UC Davis campus or Zoom to learn how Dr. James Sayre's work revolutionizes this process by using remote sensing and machine learning to develop accurate models with data collected more efficiently and frequently.
Leveraging publicly available survey data and large-scale satellite imagery, Dr. Sayre and coauthors (Dr.) Joel Ferguson and Kangogo Sogomo introduce a novel dimensionality reduction technique. This allows them to train machine learning models with only aggregated data while maintaining the crucial temporal and spatial features for accurate crop discrimination.
This novel approach can predict crop yields and other agricultural outcomes at the field-level with comparable accuracy rates to other models trained on more granular data. Furthermore, this method has the potential to enhance the reliability and periodicity of publicly available survey data, offering an up-to-date and accurate view of our food systems.
This event is available in hybrid format to attend in person or online. Lunch will be provided for in-person attendees.
Learn more about the event here.