Research on Wood Species Recognition Method Based on Hyperspectral Image Texture Features

Hyperspectral imaging technology originated from remote sensing measurement, which combines two-dimensional imaging and spectral technology organically. It can obtain spatial and spectral information of the object of study at the same time. However, it has just begun to solve the problem of wood species recognition in China. In this paper, the hyperspectral data of wood species sample are obtained in the range of 400-2500 nm. PCA is used to reduce the dimension of the original hyperspectral image data, and the principal component image sequence is obtained. The wood texture feature parameters are extracted from the principal component image sequence. The recognition rate of PNN neural network classifier is 96.5%. Experiment shows that this method is suitable for solving the problem of wood species recognition.