An Unmanned Autonomous Octocopter equipped with a multispectral camera was used to take imagery of peanut plant cover at different stages of maturity. Vegetation indexes (Normalized Difference Vegetation Index, Transformed Difference Vegetation Index, Modified Soil Adjusted Vegetation Index, Modified Chlorophyll Absorption Ratio Index and the Modified Triangular Vegetation Index) were stacked and used to mask out the peanut canopy cover from background soil, shadows and any other surficial materials. Masked peanut canopy was used to develop a peanut maturity-spectral reflectance prediction model. The model was built using partial least squares. Comparison between the model- predicted and real values showed that the model does not give an accurate estimate of maturity up to 60 days after planting, but the accuracy of the model increases with time. This may since the difference between chlorophyll a and become more significant in mature peanuts more than immature ones.The overall assessment of the model indicates that the model needs to be calibrated for more precise prediction of the peanut maturity. This could be achieved by expanding the peanut maturity, versus peanut leaf spectra, database by taking data more frequently, especially close to harvest. Data collection should be started two months after planting when the ratios between chlorophyll a and b become more detectable in the leaf reflectance spectra.