30-05-2023 дата публикации
Номер: CN116183579A
Принадлежит:
The invention discloses a rapid identification method for drugs in hair based on a regional convolutional neural network, and the method comprises the following steps: S1, carrying out SERS detection on the drugs in the hair, and obtaining spectral data; s2, airPLS baseline correction is carried out on the spectral data, and then ridge extraction is carried out to remove noise; s3, carrying out dimension reduction processing on the spectral data processed in the S2 through reverse feature elimination; s4, building a convolutional neural network model, and inputting the processed data as a training set into the neural network model for training; and S5, identifying the hair sample by using the trained convolutional neural network model. The method solves the problems of complex system, small spectral difference of similar substances and low identification accuracy in a conventional SERS drug identification method, has the characteristics of rapidness, accuracy and high sensitivity, and can ...
Подробнее