|
|
|
|
K. Tanabe, H. Uesaka, K. Aikawa, and T. Amano
[Chem. and Software, Vol.17, pp.127-130, 1995]
We examined whether the neural network system applied to the identification of plastic wastes is of practical performance. Near-infrared spectra were measured for several standard plastics, and the neural network of the three-layered back-propagation type was trained by using those spectral data. Test was carried out by using unknown sample spectra. As a result, it was found that the neural network system applied to the identification of plastic wastes is of practical performance.