Distinction of true or fake blood based on photoacoustic spectroscopy combined with artificial intelligence algorithms
In order to accurately and non-destructively identify the true or fake of blood, the photoacoustic spectroscopy technique was used in this paper. Meanwhile, a kind of custom-built photoacoustic spectroscopy detection system was established. In this system, a 532nm pumped OPO pulsed laser was used as the excitation source, and a focused ultrasonic detector with central echo frequency of 2.5MHz was used to capture the photoacoustic signals of the blood samples. In experiments, five kinds of different blood samples, i.e, three kinds of animal blood, and two kinds of fake blood (props blood and red ink) was used as the experimental blood samples. The sample groups were 125, the train samples were 100 groups, the test samples were 25 groups. The photoacoustic signals and peak-to-peak values of all blood samples were obtained. To distinct accurately the blood, two different algorithms, i.e,, PCA-KNN, and BP-GA were used. The photoacoustic peak-to-peak values were used as the input data. For PCA-KNN, the distinction correct rate of five kinds of blood is 96%, which is larger than that of the KNN (88%). For the BP-GA, the distinction correct rate of five kinds of blood is 100%. Therefore, the photoacoustic spectroscopy combined with artificial intelligence algorithms have the significant values in the distinction can classification of blood