Yyield prediction in a commercial apple orchard by analyzing RGB and multi-spectral images of trees during flowering period

Abstract

The objective of the research was to study the spatial variability in an apple orchard and to develop methods for predicting the variation of the final yield. RGB and multi-spectral images of part of the trees were taken during the flowering period. Software was used to isolate and measure the flowers from the whole images. The goals were to a) study the flower variability in an apple orchard, b) to examine the correlation between the flower variability and the variability of the final yield and c) to evaluate the yield prediction based on RGB and multispectral camera images. The results showed that in both methods, the estimated percent of flowers was highly correlated with the final yield and the method which used multi-spectral images gave the best results.