A mobile phone detection method based on wavelet feature vector extraction

Abstract

The invention discloses a mobile phone detection method based on wavelet feature vector extraction, which includes collecting detection signals, passing the object to be tested through the detection gate, and sampling the detection signals. The sampling frequency is 30HZ, and the signal duration is about 2s. After A / After D conversion, it is sent to the computer; using wavelet analysis, the collected detection signal is decomposed by DB3 wavelet, and 8 signal components are obtained, and the signal components are reconstructed to obtain signal feature quantities of different frequency bands, and then the signal feature quantities of each frequency band are extracted; The principal component analysis method is used to reduce the dimension of the feature vector, and the signal feature quantity is transmitted to the application program; the trained BP network outputs the analysis result. The invention uses the DB3 wavelet to process the acquisition signals of the mobile phone detection magnetic array, improves the SNR of the acquisition array, performs wavelet decomposition on the mobile phone detection magnetic array data to extract feature vectors, and identifies mobile phones and non-mobile phones, greatly improving the accuracy of mobile phone detection.

Description

technical field [0001] The invention relates to the field of mobile phone detection, in particular to a mobile phone detection method based on wavelet feature vector extraction. Background technique [0002] With the continuous improvement of confidentiality requirements, it is strictly forbidden to carry mobile phones, recording pens, video recorders and other electronic products on many occasions. There is an urgent need for a device that can detect such electronic products. There are three main detection technologies at present, one is the traditional coil technology, which detects mobile phones through eddy current detection technology; the other is magnetic sensor technology, which detects electronic products such as mobile phones by detecting changes in the geomagnetic field; the other is based on the nonlinearity of electron tubes. Node detection technology, which detects mobile phones by analyzing nonlinear node high-order harmonics. [0003] The threshold of nonlin...

AI Technical Summary

Problems solved by technology

[0005] The main barrier of the current technology is that it is difficult to distinguish non-mobile phone ferromagnetic items such as mobile phones and keys, and there is a high false alarm rate. Improved positive and false positive rates at the same time

Images