Nondestructive inspection of melon´s sugar content based on impedance characteristics

The objective of the study is to establish a model between the impedance characteristics and the sugar content of internal quality index of melons. The equivalent series resistor and equivalent series capacitor of melon are measured over the frequency from 1 KHz to 100 KHz by a LCR meter and an airtight shielding case. Then the impedance is calculated. Through principal component analysis (PCA), four principal components are selected to model for back propagation neural network (BPNN) optimized by genetic algorithm (GA). Comparing this method with BPNN and partial least squares (PLS), it is obviously showed that BPNN optimized by GA model is reliable and practicable. The inspecting results are assessed by correlation coefficient R=0.8413, and the root mean squares error of prediction RMSEP=0.762. A method is proposed to detect melon´s sweetness.