Category attribute fused deep network underground target identification method and system

Abstract

The invention discloses a category attribute fused deep network underground target identification method and system. The method comprises the steps that the received ground penetrating radar echo signals of multiple underground target objects are preprocessed, and multiple target echo signals respectively representing the multiple target objects are extracted; a sample data set is established according to the multiple target signals, and the sample data set includes the multiple target signals and multiple category information to which the target signals belong; a deep network classifier performs deep network training iterative calculation on the data in the sample data set and searches parameters enabling the cost function value of the sample data set to be the minimum cost function value; and classification and identification are performed by using a softmax regression mode in the process of deep network training iterative calculation so as to determine the multiple target objects. Compared with existing identification methods, identification rate of the target objects can be effectively enhanced when the sample size is insufficient or number of times of training iteration is few.

Description

technical field [0001] The invention relates to the technical field of ground-penetrating radar, in particular to a deep network underground target recognition method and system that integrates category attributes. Background technique [0002] Ground penetrating radar (GPR) is a non-destructive detection technology for shallow strata emerging in the 1980s that uses the principle of electromagnetic wave reflection to detect underground targets and geological structures and distributions. It infers the structure of the medium and the position and shape of the target according to the nature of the propagation path, intensity and waveform of the electromagnetic wave propagating in the medium, which changes with the electromagnetic properties and geometric structure of the medium. It has become an effective tool for shallow underground detection. means. Due to the variety and complexity of underground material components, the dielectric loss causes greater attenuation of electr...

AI Technical Summary

Problems solved by technology

[0016] The present invention provides a deep network underground target recognition method and system that integrates category attributes to solve existing technical problems, and can effectively improve the recognition rate of GPR targets when the sample size is insufficient or the number of training iterations is small

Images