Discrete cosine neural-network fuzzy noise reduction method for nuclear detection data

Abstract

The invention belongs to the field of nuclear detection processing in nuclear technique exploration, in particular to a discrete cosine neural-network fuzzy noise reduction method for nuclear detection data, and aims to effectively reduce noise in gamma spectrum data. The method includes the steps: subjecting spatial data formed by nuclear detection signal data to discrete cosine transform to obtain frequency spectrum data in a discrete cosine domain; subjecting the nuclear detection data to filtering and inverse discrete cosine transform in the discrete cosine domain to obtain primary noise reduction result; establishing a neural network, inputting a sample by using the nuclear detection signal data as nodes in the neural network, and estimating parameters of a membership function in the neutral network by means of least squares back-propagation to obtain final noise reduction result output by the neural network. The method is capable of well recognizing noise signals, network model parameters are regulated and optimized according to noise signal features, the noise in the signal data is reduced effectively, and signal-to-noise ratio of the nuclear detection data is increased greatly.

Description

technical field [0001] The invention belongs to the field of nuclear detection data processing in nuclear technology exploration, and in particular relates to a discrete cosine neural network fuzzy noise reduction method for nuclear detection data. Background technique [0002] Nuclear detection methods, mainly including gamma measurement and gamma energy spectrum measurement, radon and its progeny measurement, X-ray fluorescence measurement, etc., are mainly used to determine the intensity of radioactive rays emitted by the measurement object or to detect the content, activity or concentration of nuclides . It is used to search for uranium mineral resources, oil and gas resources or conduct nuclear environmental assessment. Noise reduction is one of the important contents in nuclear detection data processing. The actual data obtained by nuclear detection contains noise signals. The noise signals generated by nuclear detection are mainly due to the statistical noise generat...

AI Technical Summary

Problems solved by technology

Noise reduction is one of the important contents in nuclear detection data processing. The actual data obtained by nuclear detection contains noise signals. The noise signals generated by nuclear detection are mainly due to the statistical noise generated by the decay of radionuclides and the noise caused by human environmental factors. , at present, signal filtering is a common noise reduction method, the most basic method is smoothing filtering, this method is easy to remove useful signals after processing, the effect is not satisfactory

In the noise reduction processing of nuclear detection data, smoothing filter and high-pass filter and low-pass filter in Fourier transform are mostly used. When the difference between the noise signal and the real signal is obvious, it is easy to filter the noise by using these methods. When there is a serious overlap between real signals, the ideal effect cannot be obtained by using conventional filtering methods, which affects the data interpretation and evaluation of the measurement results

At present, the application of cosine transform at home and abroad is limited to the fields of image compression, gravity, magnetism, and seismic data processing, and no noise reduction processing applied to nuclear detection data has been seen.

Images