Method for identifying a disease type of a GPR image

Abstract

The invention discloses a method for identifying a disease type of a GPR (Ground-penetrating radar) image. The method comprises the following steps: step 1, normalizing pixels of a GPR image for the GPR image with disease reflected waves and obtaining a matrix NI; step 2, carrying out histogram equalization on the matrix NI to obtain a matrix NI_H; step 3, carrying out a ternary operation on the matrix NI_H, wherein a ternary threshold is set accord to the maximum inter-class variance double threshold method, and obtaining a matrix NI_HB; step 4, conducting derivation of each column of the matrix NI_HB, then making phase adjacent elements in each column of the matrix NI_HB have different signs, and finally obtaining two phase types; step 5, selecting the phase type with a large number as aGPR image disease type. The beneficial effects of the method are as follows: disease features are highlighted, extraction of the disease feature is facilitated, common diseases under a highway are highlighted, a disease interpretation process can be conveniently known, the cost is saved, and the method conforms to a pursuit goal of realizing automation of disease identification in the ground-penetrating radar industry and has a great practical significance.

Description

technical field [0001] The invention relates to a method for identifying disease types of GPR (Ground-penetrating radar, ground-penetrating radar) images, and belongs to the technical field of disease identification. Background technique [0002] In the interpretation of roadbed diseases of GPR signal expressway, the most widely used method at home and abroad is manual image interpretation. However, manual interpretation of images relies heavily on the experience of the interpreter and is highly subjective. When the amount of data is large, the manual interpretation cycle is very long and has a certain lag. How to automatically identify the type of disease and give feedback to save interpretation time has become one of the urgent needs of GPR signal road subgrade image interpretation. The present invention provides a new solution based on the automation of GPR data disease detection and classification. Other existing patents such as Patent No. CN104698503A use offset corre...

AI Technical Summary

Problems solved by technology

This method is suitable for the search of tubular targets, and the use of manual methods to find the target body is time-consuming and labor-intensive

Another example is the patent number CN1595195A, which uses RBF neural network to automatically identify the target category of radar data. The premise is to analyze and extract the characteristics of the target, and a large amount of sample data is required. However, in the actual implementation process, a large amount of sample data is usually difficult to obtain.

Images