Blooming charge assessment in apple orchards for automatic thinning activities

This work aims to develop an automatic system capable of providing objective information about the bloom charge in an apple orchard in order to manage flower-thinning activities. The article presents and discusses the use of a mobile lab (ByeLab) equipped with several optical sensors to carry out a site-specific bloom charge assessment in apple trees. The data collected by the sensors were processed by a specific algorithm implemented in MatLab^®. Investigations of the flower reflectance signature indicated that the Normalized Difference Vegetation Index (NDVI) is the most suitable parameter to distinguish leaves from flowers. Pure flowers produce NDVI values slightly negative or at least very near to 0. Despite the homogeneous behavior of the NDVI flower response, OptRx™ sensors, which provide an average assessment of an area, were not able to highlight a significant correlation between the number of flowers and the NDVI values. In the future, further studies will be conducted to assess if other techniques based on image analyses can provide better and more sensitive results regarding the bloom charge assessment. Such results could then be used as a reference in automating machines for thinning operations according to a site-specific approach.