New, Low-Cost, Hand-Held Multispectral Device for In-Field Fruit-Ripening Assessment

The state of ripeness at harvest is a key piece information for growers as it determines the market price yield. This has been traditionally assessed by destructive chemical methods, which lead to low-spatiotemporal resolution in monitorization crop development and poor responsiveness growers. These limitations have shifted focus remote-sensing, spectroscopy-based approaches. However, most research focusing on these approaches accomplished with expensive equipment, exorbitant users. To combat this issue, work presents low-cost, hand-held, multispectral device original hardware specially designed face complexity related in-field use. proposed based board (AS7265x, AMS AG) that three sensor chips spectral response eighteen channels range from 410 940 nm. was evaluated red-grape field experiment. Briefly, used acquire signature eighty samples vineyard. Subsequently, grape were analysed using standard methods generate ground-truth values ripening status indicators (soluble solid content (SSC) titratable acidity (TA)). pre-process reflectance measurements input training artificial neural network models estimate two target parameters (SSC TA). developed estimation through leave-one-out cross-validation approach obtaining promising results (R2 = 0.70, RMSE 1.21 SSC; R2 0.67, 0.91