System and method for predicting crop yield

A non-transitory, tangible, computer-readable media having computer-readable instructions stored thereon, for use with a computer, that when executed by the computer, cause the computer to:

• receive image data of a geographic region, the image data including pixels;
• generate an array of vegetation indices based on the received image data utilizing a plurality of vegetation index generating components, each operable to generate a respective individual vegetation index based on the received image data, wherein each vegetation index in the array of vegetation indices is associated with a respective weighting factor;
• receive crop data comprising training data for classifying one or more of the pixels as a type of crop;
• generate a masked vegetation index based on the array of vegetation indices and the received crop data, such that the pixels are associated with vegetation index data from each vegetation index in the array of vegetation indices and such that the pixels are associated with a classification as a type of crop;
• receive historical crop yield data;
• and
• generate, utilizing multivariate regression, a first predicted crop yield based on the masked vegetation index, and a second predicted crop yield based on the historical crop yield data, wherein the multivariate regression includes utilizing the weighting factors of the array of vegetation indices and includes modifying the weighting factors of the array of vegetation indices based on past accuracy of results of individual ones of the array of vegetation indices, such that accuracy of the second predicted crop yield in relation to the historical crop yield data is increased.