Maturity prediction of intact bell peppers by sensor fusion

Abstract The objective of this study is to present a fusion of non-destructive sensor outputs and a fusion of destructive reference parameters. The sensors used in the present work were spectrophotometers in the VIS–NIR and SWIR spectral range, hyperspectral imaging in the visible range, relaxation and ultrasonic tests, and colour measurement. As reference parameters, the following were used: total soluble solids, dry matter, osmotic potential, ascorbic acid, total chlorophylls, carotenoids content, coefficient of elasticity measured in compression and rupture mode. The fusion procedure was based on the combination of sensor outputs and the combination of reference parameters. Linear and non-linear regression methods were applied for model establishment. Multi-sensor models were found to be better than the single sensor models based on the significantly lower root mean square errors of cross validation values for all tested cultivars and all reference parameters. By the reference parameter fusion a new combined quality index was developed in order to evaluate the global quality of the produce. With the new combined quality index not only the comprehensive quality of the produce could be predicted but also its maturity stage, which can serve as a basis for better decision of harvest schedule; as the new index correlate with the physical property change during growth.