A Data Interpretation and Recognition Method for 3D GPR
A ground-penetrating radar and recognition method technology, applied in the field of ground-penetrating radar, can solve problems such as low execution efficiency, deviation of processing results, and inability to work continuously, and achieve the effects of accurate classification, fast positioning, and high efficiency
Active Publication Date: 2022-03-22
深圳安德空间技术有限公司
View PDF10 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
At present, the data processing and image interpretation of 3D ground penetrating radar are mainly carried out manually, which has low execution efficiency and cannot work continuously
Inconsistent standards of different processing personnel lead to deviations in processing results
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0072] refer to figure 1 , the present invention provides a kind of technical scheme:
[0073] A data interpretation and identification method for three-dimensional ground penetrating radar includes steps S1 to S4.
[0074] In step S1, the data of each channel of the three-dimensional ground penetrating radar is read and preprocessed;
[0075] During implementation, the preferred step S1 of this scheme further includes stage 1 and stage 2:
[0076] In stage 1, the radar raw data of all 16 channels in all survey lines are read and converted to the image domain. The conversion formula is:
[0077]
[0078] where x i is the original i-th echo value, f(x i ) is the filter function, y i is converted to the corresponding value in the image domain whose gray value range is [0,b].
[0079] Through the above conversion, the normalization processing of radar data amplitude components to image pixels is completed.
[0080] In stage 2, the raw data is preprocessed by gain and fi...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to the technical field of ground penetrating radar, in particular to a data interpretation and identification method for three-dimensional ground penetrating radar. This method uses artificial intelligence based on deep learning to interpret and identify 3D ground penetrating radar data, which includes reading the data of each channel of 3D ground penetrating radar and preprocessing it; Scan and C-scan scan, combine scanned images to generate sample pictures, and classify and label the sample picture set; divide the sample picture set into training set, verification set and test set according to a specific ratio, and use InceptionV3 deep convolution The neural network performs migration training; reads the data collected by the 3D ground penetrating radar, preprocesses it and generates a combined scan image, inputs the deep convolution model hdf5 format file, outputs the confidence degree, and classifies and recognizes the target object according to the confidence degree.
Description
technical field [0001] The invention relates to the technical field of ground penetrating radar, in particular to a data interpretation and identification method for three-dimensional ground penetrating radar. Background technique [0002] Ground Penetrating Radar (Ground Penetrating Radar), also known as geological radar, detects underground media by emitting high-frequency pulsed electromagnetic waves (frequency between 1MHz-1GHz), and determines the distribution of underground media. It has simple operation, high detection accuracy, no damage, and acquisition It is the most active detection technology for engineering inspection and survey at present, and it is widely used in the detection of underground diseases of urban roads. Compared with the traditional two-dimensional radar with only one longitudinal and vertical profile waveform, the three-dimensional radar seamlessly splices radar data and position information through real-time rapid sampling, collects horizontal p...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06V10/764G06K9/62G01S13/88G06N3/04G06N3/08
CPCG01S13/885G06N3/08G06N3/045G06F18/24
Inventor 项芒
Owner 深圳安德空间技术有限公司