Ultrasonic Technology Ceramic Identification Based On Weighted Euclidean Distance

Ultrasonic nondestructive testing technology is fast,accurate,non-polluting,non-destructive,and has a wide range of applications.Since the technology was developed,it has been widely used in many fields.In order to overcome the limitations of the traditional method of ceramic identification,this paper proposes an ultrasonic identification method based on weighted Euclidean distance.The three batches of exactly the same size square ceramic box and four identically sized round dishes produced in the same batch are used as the experimental samples for the two groups of experiments.According to the section scanning electron microscope photographs,the ultrasonic wavelength and its internal particle size are analyzed.Relationship between.The signals of 10 MHz and 7.5 MHz ultrasonic waves inside the two samples were extracted respectively.Select the time-domain characteristic backscatter signal of the square dish raw signal as its feature label,calculate the weighted Euclidean distance between the backscattered signal of each sample and the backscattered signal of the sample to be identified,and the pre-stored self-identified sample itself.The weighted Euclidean distances are compared.The db6 wavelet transform is used to calculate the wavelet energy characteristics of each wavelet component of the original signal in the circular dish.Two wavelet components with a sum of energy greater than 95%are selected as the feature labels of the circular plate,and the energy characteristics of the wavelet components of each sample are calculated.The weighted Euclidean distance between the energy features of the wavelet components of the sample to be identified is compared with the Weighted Euclidean distances of the samples to be identified themselves.The calculation results of the two experiments show that both the characteristic backscattering signal of the time domain signal and the wavelet energy characteristic of the frequency domain feature of the ceramic sample can reflect the internal characteristics of the ceramic sample and can be used as the anti-counterfeiting of the ceramic sample.Feature tag.It also shows that the weighted Euclidean distances between different samples are different.By calculating the weighted Euclidean distance between the sample signal characteristics,ceramic samples can be quickly and accurately identified.