Soil physical property classification recognition method and device based on geological radar

A technology of geological radar and identification method, applied in the field of detection, which can solve the problems of non-stationarity and randomness, difficulty in correctly interpreting appearance features, and time-varying echo signals.

CN103941254AInactive Publication Date: 2014-07-23CHINA SHENHUA ENERGY CO LTD +2
3 Cites 29 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2014-07-23
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a soil physical property classification recognition method based on geological radar. The method includes the data preprocessing step, the soil physical property information extraction step, the neural network training step and the soil physical property classification recognition step, wherein in the data preprocessing step, digital signals acquired by the geological radar are de-noised and filtered; in the soil physical property information extraction step, characteristic data for representing soil physical properties are extracted from data after the data preprocessing step; in the neural network training step, the characteristic data are used for training a neural network as a data set for training the neural network, and a mapping result corresponding to the characteristic data is obtained; in the soil physical property classification recognition step, the characteristic data to be recognized for representing the soil physical properties are input the trained neural network, and the classification of the soil physical properties is recognized according to the mapping result of the neural network. The invention further discloses a soil physical property classification recognition device based on the geological radar. Through the soil physical property classification recognition method based on the geological radar, the physical properties of soil of a region to be detected can be fast and accurately recognized based on radar detection data.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the field of detection technology, in particular to a method and device for identifying soil physical properties based on geological radar. Background technique

[0002] Soil physical properties refer to soil physical properties, such as soil water content, soil texture, soil porosity, etc. Taking soil water content as an example, soil can be regarded as different categories according to the value of soil water content. Mastering the types of soil physical properties is beneficial to environmental protection, engineering planning and construction, etc.

[0003] Geological radar technology is a science of underground target detection by transmitting high-frequency pulsed electromagnetic waves (frequency range 106-109Hz). The basic principle is: the transmitter transmits pulsed electromagnetic wave signals through the transmitting antenna. When detecting a target, a reflected signal is generated. The direct signal and reflecte...

Examples

Embodiment Construction

[0030] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and examples.

[0031] figure 1 It is a flow chart of the soil physical property category identification method based on geological radar in the present invention, including:

[0032] S101, a data preprocessing step, performing denoising processing and filtering processing on the digital signal collected by the ground radar;

[0033] S102, the soil physical property information extraction step, extracting characteristic data representing soil physical properties from the data that has undergone the data preprocessing step;

[0034] S103, the neural network training step, using the feature data as a data set for training the neural network, training the neural network, and obtaining a mapping result corresponding to the feature data;

[0035] S104, the step of identifying the soil physical property category, inputting the characteristic data rep...