Extraction of most effective wavelength of egg spectra using genetic algorithm and classifying using regression equations

The potential of Vis-IR (400–1100 nm) transmittance method to assess the internal quality (freshness) of intact chicken egg during storage at a temperature of 307 oC and 25% 4 relative humidity was investigated. One hundred chicken egg samples were used for measuring its freshness and spectral collection during egg storage times (up to 30 days). Two correlation models between Haugh unit (HU) and the yolk coefficient (YC) versus the storage time were developed with the correlation coefficients of 0.86, 0.96, respectively. The latter equations showed in that aforementioned conditions egg quality decreased dramatically. Furthermore, the method has developed to acquire the wavelength features based on genetic algorithm (GA). It can solve the problem of the effective information extraction from the high-dimensional data matrix. Regression equations were developed by raw and according to the wavelength features were acquired, respectively basing on different preprocessing (SNV, MSC, FFT, as well as 1st and 2nd derivations). The result indicated that regression equations based on GA and preprocessing SNV and MSC could classify egg’s spectra in three groups while by 1st derivation preprocessing, each daily spectrum could allocate each of them into separated groups.