Feature Detection of Nanofibers Based on Nonlinear Mapping Pattern Recognition

To enhance the feature extraction capacity of nanofibers, a method of feature detection based on nonlinear mapping pattern recognition is proposed. The characteristic distribution model of nanofibers is constructed, and the spectral characteristic decomposition method is used to recognize the nonlinear mapping pattern of nanofibers at current density. The spatial spectrum beam forming processing of nanofiber features is carried out by using cluster–cluster hybrid molecular reconstruction method, and the association rule feature decomposition of nanofibers is carried out by recursive graph analysis method, and the nonlinear mapping pattern recognition of nanofiber features is realized. The classification and recognition of nanofiber features are carried out by combining the correlation attribute clustering method, and the characteristics detection optimization of nanofibers is realized. The proposed method has higher acurracy than other methods. The pattern recognition performance of nonlinear mapping is good, and the ability of accurate recognition of the crystal structure characteristics of nanofibers is better.