A method for recognizing disease types in GPR images based on shape features

Abstract

A method for identifying that disease type of GPR image based on shape feature comprises the following steps: step 1: for the GPR image with disease reflection wave, carrying out F. K offset to obtainthe GPR image matrix NI; 2, normalizing that GPR image matrix NI to obtain a normalized matrix NI_H; 3, binarizing that normalized matrix NI_H accord to the maximum inter-class variance threshold method to obtain a binary matrix NI_HB; 4, carrying out image segmentation on that binary matrix NI_HB according to the maximum connecte area of the image, and after segmentation, only the maximum connected area of the image is reserved; 5, carrying out image pattern recognition according to that ratio of the depth in the vertical direction to the width in the horizontal direction of the maximum connecte area after image segmentation, and judging the disease type of the GPR image. The invention realizes the prominence of the shape feature of the disease, is advantageous in extracting the diseasefeature, facilitates the disease interpretation process, saves the cost, accords with the pursuit goal of the automation of the disease identification of the ground penetrating radar industry, and hasgreat practical significance.

Description

technical field [0001] The invention relates to a method for identifying disease types of GPR images based on shapes, and belongs to the technical field of disease identification. Background technique [0002] In the interpretation of ground-penetrating radar (GPR) signal expressway subgrade disease, 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 invention provides a new solution based on the automation of GPR data disease detection and classification. Patent CN104698503A uses offset correction and Kirchhoff wave equation offset method to proc...

AI Technical Summary

Problems solved by technology

This method is computationally intensive and consumes a lot of manpower and time

Patent CN105403883A uses the amplitude component to find the region of interest, and performs edge extraction and target hyperbola positioning for the region of interest. This method is suitable for searching tubular targets, and using manual methods to find the target body is time-consuming and manpower-consuming

Patent CN1595195A uses RBF neural network to automatically identify the target body category of radar data. The premise is to analyze and extract the characteristics of the target body, and a large amount of sample data is required, but it is usually difficult to obtain a large amount of sample data in the actual implementation process.

Images