Abstract:

An egg is excited by a light mechanical impact on different locations of the eggshell, then the cracks are deteted by dynamical frequency response using flexible piezoelectric film sensors. Analyse show that the dominant resonance frequency can be observed and the frequency value is lower in the intact eggs. While for cracked eggs, there are several peak frequencies, the magnitudes of these frequencies are approximate, and the frequency value is higher. In normalization average of frequency domain, 10 or 20 magnitudes of interval frequency, 10 magnitudes in turn and frequencies by magnitude in turn are extracted as eigenvalues. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are used to investigate whether the dynamical frequency response is able to distinguish intact egg from cracked egg. Results show that the frequency response could distinguish the intact egg from the cracked egg using LDA and PCA based on magnitudes in turn and frequencies by magnitude in turn, and higher distinguish rate can be obtained based on frequencies by magnitude in turn. Also higher distinguish rate can be obtained using LDA or based on 10 eigenvalues.

Key words: food engineering, egg crack, crack detection, principal component analysis,PCA, linear discriminant analysis,LDA