Chile is one of the main exporters of sweet cherries in the world and one of the few in the southern hemisphere, being their harvesting between October and January. Hence, Chilean cherries have gained market in the last few years and positioned Chile in a strategic situation which motivates to undergo through a deep innovation process in the field. Currently, cherry crop estimates have an error of approximately 45%, which propagates to all stages of the production process. In order to mitigate such error, we develop, test and evaluate a deep neural-based approach, using a portable artificial vision system to enhance the cherries harvesting estimates. Our system was tested in a cherry grove, under real field conditions. It was able to detect cherries with up to 85% of accuracy and to estimate production with up to 25% of error. In addition, it was able to classify cherries into four sizes, for a better characterization of the production for exportation.

Fuel moisture content (FMC) proved to be one of the most relevant parameters for controlling fire behavior and risk, particularly at the wildland-urban interface (WUI). Data relating FMC to spectral indexes for different species are an important requirement identified by the wildfire safety community. In Valparaíso, the WUI is mainly composed of Eucalyptus Globulus and Pinus Radiata—commonly found in Mediterranean WUI areas—which represent the 97.51% of the forests plantation inventory. In this work we study the spectral signature of these species under different levels of FMC. In particular, we analyze the behavior of the spectral reflectance per each species at five dehydration stages, obtaining eighteen spectral indexes related to water content and, for Eucalyptus Globulus, the area of each leave—associated with the water content—is also computed. As the main outcome of this research, we provide a validated linear regression model associated with each spectral index and the fuel moisture content and moisture loss, per each species studied.

The vegetation indices derived from spectral reflectance have served as an indicator of vegetation’s biophysical and biochemical parameters. Some of these indices are capable of characterizing more than one parameter at a time. This study examines the feasibility of retrieving several spectral vegetation indices from a single index under the assumption that all these indices are correlated with water content. The models used are based on a linear regression adjusted with least squares. The spectral signatures of Eucalyptus globulus and Pinus radiata, which constitute 97.5% of the forest plantation in Valparaiso region in Chile, have been used to test and validate the proposed approach. The linear models were fitted with an independent data set from which their performance was assessed. The results suggest that from the Leaf Water Index, other spectral indices can be recovered with a root mean square error up to 0.02, a bias of 1.12%, and a coefficient of determination of 0.77. The latter encourages using a sensor with discrete wavelengths instead of a continuum spectrum to estimate the forestry’s essential parameters.