Reducing the use of pesticides by early visual detection of diseases in precision agriculture is important. Because of the color similarity between potato-plant diseases, narrow band hyper-spectral imaging is required. Payload constraints on unmanned aerial vehicles require reduc- tion of spectral bands. Therefore, we present a methodology for per-patch classification combined with hyper-spectral band selection. In controlled experiments performed on a set of individual leaves, we measure the performance of five classifiers and three dimensionality-reduction methods with three patch sizes. With the best-performing classifier an error rate of 1.5% is achieved for distinguishing two important potato-plant diseases.
MULTIFILE
Three empirical models were used to fit the formation of acrylamide in crisps of three different cold-sweetened potato genotypes, fried under the same experimental conditions. Statistical methods were used to compare the performance of the models, with the "Logistic-Exponential" model performing the best. The obtained model parameters for the formation of acrylamide showed improvement in precision compared to an earlier study, the precision of the parameter estimates for the degradation of acrylamide was still problematic. Nevertheless, the predictive capacity of the "Logistic-Exponential" model was tested, as this model showed a strong correlation between parameter a and the reducing sugar content of the raw potato. The predictions from this model for the formation of acrylamide in potato crisps were close to earlier reported experimental values. Therefore, the use of the "Logistic-Exponential" model as a tool to predict acrylamide in potato crisps seems promising and should be developed further.
The formation of acrylamide in potato crisps was fitted by empirical mathematical models. Potato slices were fried under the same experimental conditions for different times. Besides the content of precursors in the raw potato slices, acrylamide and water content in the potato crisps were quantified after predetermined times (2-6 min). The temperature developments in the surrounding oil and outer cell layer of the potato slices were monitored, giving more insight in the frying process and making future comparisons between studies possible. The pattern found for the formation of acrylamide, which was similar to earlier studies, was fitted to three empirical models. Statistical methods were used to compare the performance of the models, with the "Logistic-Exponential" and "Empirical" model performing equally well. The obtained model parameters were in the range of earlier reported studies, although this comparison is not unequivocal as the experimental conditions differed between studies. The precision of parameter estimates was problematic; this should be improved by better experimental design. Nevertheless, the approach of this study will make it possible to truly compare acrylamide formation patterns and model parameters in the future, with the ability to develop a tool to predict acrylamide formation in potato crisps.
