CN-115902863-B - Method, device and equipment for determining actual size of disease

CN115902863BCN 115902863 BCN115902863 BCN 115902863BCN-115902863-B

Abstract

The invention provides a method, a device and equipment for determining the actual size of a disease, wherein the method comprises the steps of obtaining an actual disease radar map of a test road based on a radar detection result of the test road, setting forward parameters based on the position and the size of the disease in the test road, obtaining a disease forward map, drawing a disease color pixel map by taking the actual disease radar map as a reference, constructing an evolution model from the actual disease radar map to the disease color pixel map through the disease forward map, and inputting the actual disease radar map of the road to be tested into the evolution model to obtain a target disease color pixel map corresponding to the road to be tested. The invention belongs to the field of road exploration, and aims to intuitively observe the actual size of invisible diseases in a road by carrying out radar detection on the road to obtain a color pixel diagram of the road diseases, so that a maintenance person can conveniently select proper maintenance measures to repair the road diseases.

Inventors

LIN XIANG
YI MINGWEI
Lv Xinpi

Assignees

中公高科养护科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20221205

Claims (8)

1. A method of determining the actual size of a disease, the method comprising: Acquiring an actual disease radar map of a test road based on a result of radar detection on the test road; Based on the forward parameter, forward modeling is carried out on the road structure and the disease position and the size in the test road, and the disease forward modeling is obtained, wherein the forward modeling parameter is consistent with the parameter for carrying out radar detection on the test road; drawing a disease color pixel map based on the disease position and the size in the test path by taking the actual disease radar map as a reference, wherein the disease position and the size in the disease color pixel map are in one-to-one correspondence with the actual disease radar map; The method comprises the steps of constructing an evolution model from an actual disease radar map to a disease color pixel map through the disease forward map, constructing a first inversion model, and obtaining a relation between the actual disease radar map and the disease forward map, wherein the first inversion model is used for outputting a disease forward simulation map corresponding to the actual disease radar map, performing deep learning on the disease forward simulation map and the disease color pixel map, constructing a second inversion model, and obtaining a relation between the disease forward map and the disease color pixel map, wherein the second inversion model is used for outputting a target disease color pixel map corresponding to the disease forward simulation map; inputting an actual disease radar map of a road to be tested into the evolution model, and obtaining a target disease color pixel map corresponding to the road to be tested, wherein the target disease color pixel map is used for describing the actual size of the disease in the road to be tested.
2. The method according to claim 1, wherein before the step of obtaining the actual disease radar spectrum of the test road based on the result of radar detection of the test road, the method further comprises: acquiring a detected disease radar map corresponding to the radar detection result; Acquiring the depth of a disease position based on the disease detection radar map; performing target processing on the disease detection radar map based on the depth of the disease position, wherein the target processing comprises at least one of direct current drift, static correction, gain, band-pass filtering and background removal; And determining the radar spectrum of the detected disease after the target processing as an actual radar spectrum of the test path.
3. The method according to claim 2, wherein the target processing of the disease-detecting radar map based on the depth of the disease location comprises: under the condition that the depth is at a first depth, performing primary signal processing on the detected disease radar spectrum, and determining the detected disease radar spectrum after the primary signal processing as an actual disease radar spectrum of the test road; Under the condition that the depth of the disease position is at a second depth, performing secondary signal processing on the primary signal processed detected disease radar spectrum, and determining the secondary signal processed detected disease radar spectrum as an actual disease radar spectrum of the test road; The primary signal processing comprises direct current drift processing, static correction processing, gain processing, band-pass filtering processing and background removing processing, and the secondary signal processing comprises the gain processing; The first depth is less than the second depth.
4. The method according to claim 1, wherein the drawing a disease color pixel map based on the disease position and size in the test road with reference to the actual disease radar map comprises: obtaining disease positions, disease widths and disease dielectric constants in the test path; acquiring the disease length in the actual disease radar map; Acquiring the dielectric constant of a road structure of the test road; and drawing the disease color pixel map based on the disease length, the disease position, the disease width, the disease dielectric constant and the road structure dielectric constant.
5. The method of claim 4, wherein the obtaining of disease length in the actual disease radar profile comprises: Acquiring a waveform diagram of the actual disease radar map; picking up first-arrival radar waves of the diseases based on the oscillogram, wherein the first-arrival radar waves of the diseases comprise a plurality of first-arrival radar waves; Judging the first arrival radar waves according to disease length indexes based on the first arrival radar waves to obtain a starting point position and an ending point position of the disease; And acquiring the disease length in the actual disease radar map based on the difference value between the starting point position and the ending point position of the disease.
6. The method of claim 5, wherein determining the plurality of first-arrival radar waves based on the plurality of first-arrival radar waves according to a disease length index to obtain a start position and an end position of the disease comprises: acquiring trough information of the first arrival radar waves, wherein the trough information is the ratio of the trough amplitude to the trough height; constructing the disease length index based on the ratio of the trough amplitude to the trough height; determining two first-arrival radar waves which are positioned at the middle end of the plurality of first-arrival radars and are the starting point first-arrival radar waves and the ending point first-arrival radar waves, wherein the trough information meets the disease length index; And acquiring the starting point position and the end point position of the disease based on the trough position of the starting point first-arrival radar wave and the trough position of the end point first-arrival radar wave.
7. An apparatus for determining the actual size of a disease, the apparatus comprising: The first acquisition module is used for acquiring an actual disease radar map of the test road based on a result of radar detection on the test road; The second acquisition module is used for setting forward parameters based on the disease position and the size in the test road and acquiring a disease forward spectrum, and comprises the steps of acquiring a road structure of the test road, forward-modeling the road structure and the disease position and the size in the test road based on the forward parameters and acquiring a disease forward spectrum, wherein the forward parameters are consistent with the parameters for radar detection of the test road; the drawing module is used for drawing a disease color pixel map based on the disease position and the size in the test path by taking the actual disease radar map as a reference, wherein the disease position and the size in the disease color pixel map are in one-to-one correspondence with the actual disease radar map; The third acquisition module is used for constructing an evolution model from the actual disease radar spectrum to the disease color pixel map through the disease forward spectrum, and comprises a first inversion model, a second inversion model, a third inversion model, a fourth inversion model and a fourth inversion model, wherein the first inversion model is used for performing deep learning on the actual disease radar spectrum and the disease forward spectrum and finding out the relation between the actual disease radar spectrum and the disease forward spectrum, the second inversion model is used for performing deep learning on the disease forward simulation spectrum and the disease color pixel map and finding out the relation between the disease forward spectrum and the disease color pixel map, and the second inversion model is used for outputting a target disease color pixel map corresponding to the disease forward simulation spectrum; The fourth acquisition module is used for inputting an actual disease radar map of the road to be tested into the evolution model, and acquiring a target disease color pixel map corresponding to the road to be tested, wherein the target disease color pixel map is used for describing the actual size of the disease in the road to be tested.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory, characterized in that the processor executes the computer program to implement the method of determining the actual size of a disease as claimed in any one of claims 1 to 6.

Description

Method, device and equipment for determining actual size of disease Technical Field The invention belongs to the field of road exploration, and particularly relates to a method, a device and equipment for determining the actual size of diseases. Background The ground penetrating radar can emit and receive microwave band high-frequency broadband electromagnetic waves. Because electromagnetic waves can reflect at the interface of the underground medium, the spatial position of an underground target can be obtained by analyzing the waveform characteristics of the electromagnetic waves reflected by the interface of the underground medium to form characteristic information such as materials, the road disease is collected by using the ground penetrating radar in a common mode, but the road disease image actually collected by using the ground penetrating radar is influenced by objective factors such as road working conditions, the operation of a collector, limited sample size and the like, for example, the traditional collection mode is to firstly collect data of the position which possibly generates the road disease, and finally excavate and verify the road, so that the actual environment inside the road is difficult to research and clear, and the specific connection between the image and the disease cannot be established. Disclosure of Invention In view of the above, embodiments of the present invention provide a method, apparatus and device for determining the actual size of a disease, so as to overcome or at least partially solve the above-mentioned problems. In a first aspect of the embodiment of the present invention, there is provided a method for determining an actual size of a disease, the method comprising: Acquiring an actual disease radar map of a test road based on a result of radar detection on the test road; based on the disease position and the size in the test path, forward modeling parameters are set, and a disease forward modeling map is obtained; drawing a disease color pixel map based on the disease position and the size in the test path by taking the actual disease radar map as a reference, wherein the disease position and the size in the disease color pixel map are in one-to-one correspondence with the actual disease radar map; constructing an evolution model from the actual disease radar spectrum to the disease color pixel map through the disease forward spectrum; inputting an actual disease radar map of a road to be tested into the evolution model, and obtaining a target disease color pixel map corresponding to the road to be tested, wherein the target disease color pixel map is used for describing the actual size of the disease in the road to be tested. Further, the method for constructing an evolution model from the actual disease radar spectrum to the disease color pixel map through the disease forward spectrum comprises the following steps: Performing deep learning on the actual disease radar spectrum and the disease forward spectrum to construct a first inversion model, wherein the first inversion model is used for outputting a disease forward simulation spectrum corresponding to the actual disease radar spectrum; Performing deep learning on the disease forward modeling spectrum and the disease color pixel map to construct a second inversion model, wherein the second inversion model is used for outputting the target disease color pixel map corresponding to the disease forward modeling spectrum; and acquiring the evolution model based on the first inversion model and the second inversion model. Further, the method sets forward parameters based on the disease position and size in the test path, and obtains a disease forward spectrum, the method comprising: acquiring a road structure of the test road; forward modeling is carried out on the road structure, the disease position and the size in the test road based on the forward modeling parameters, and a disease forward modeling map is obtained; the forward parameter is consistent with the parameter for carrying out radar detection on the test path. Further, before the actual disease radar spectrum of the test path is obtained based on the result of radar detection on the test path, the method further comprises: acquiring a detected disease radar map corresponding to the radar detection result; Acquiring the depth of a disease position based on the disease detection radar map; performing target processing on the disease detection radar map based on the depth of the disease position, wherein the target processing comprises at least one of direct current drift, static correction, gain, band-pass filtering and background removal; And determining the radar spectrum of the detected disease after the target processing as an actual radar spectrum of the test path. Further, the target processing of the disease detection radar map based on the depth of the disease position includes: under the condition that the depth is at a first depth, performing