GPRPy - Ground Penetrating Radar Processing
Quick Reference
import gprpy.gprpy as gp
import matplotlib.pyplot as plt
# Load and display
data = gp.gprpyProfile()
data.importdata('profile.DZT')
data.showProfile()
plt.show()
# Access data
print(f"Traces: {data.data.shape[1]}")
print(f"Samples: {data.data.shape[0]}")
print(f"Time range: {data.twtt.max():.1f} ns")
Supported Formats
| Format | Manufacturer |
|---|
| .DZT | GSSI |
| .DT1 | Sensors & Software |
| .GPR | MALA |
| .rd3/.rad | MALA |
| .sgy | SEG-Y |
Essential Operations
Basic Processing
data = gp.gprpyProfile()
data.importdata('profile.DZT')
data.dewow(window=10) # Remove low-frequency drift
data.remMeanTrace(ntraces=50) # Remove background ringing
data.tpowGain(power=1.5) # Time-power gain
data.agcGain(window=25) # Automatic gain control
data.showProfile()
Apply Filters
data.bandpassFilter(minfreq=100, maxfreq=800) # MHz
data.lowpassFilter(maxfreq=500)
data.highpassFilter(minfreq=50)
Time-to-Depth Conversion
velocity = 0.1 # m/ns (typical for dry sand)
data.setVelocity(velocity)
data.showProfile(yrng=[0, 5]) # Top 5 meters
Topographic Correction
data.topoCorrect(topofile='topography.txt', velocity=0.1)
# File format: x_position, elevation
Export Results
data.exportFig('processed.png', dpi=300)
data.exportSEGY('processed.sgy')
data.exportASCII('processed.txt')
Velocity Analysis (CMP)
cmp = gp.gprpyCMP()
cmp.importdata('cmp_survey.DZT')
cmp.showCMP()
cmp.semblance(vmin=0.05, vmax=0.15, vstep=0.01)
cmp.showSemblance()
Processing Parameters
| Parameter | Typical Value | Description |
|---|
| Dewow window | 5-20 ns | Low-frequency removal window |
| Gain power | 1.0-2.0 | Time-power gain exponent |
| AGC window | 10-50 ns | Automatic gain window |
| Bandpass | 100-800 MHz | Frequency filter range |
Material Velocities
| Material | Velocity (m/ns) |
|---|
| Air | 0.30 |
| Dry sand | 0.10-0.15 |
| Wet sand | 0.06-0.08 |
| Dry soil | 0.08-0.12 |
| Wet soil | 0.05-0.08 |
| Limestone | 0.10-0.12 |
| Granite | 0.10-0.13 |
| Water | 0.033 |
| Ice | 0.16-0.17 |
When to Use vs Alternatives
| Tool | Best For | Limitations |
|---|
| gprpy | Python-based GPR processing, scripted workflows, open-source | Limited advanced migration algorithms |
| GPRMax | Forward modelling and simulation of GPR responses | Simulation only, not for data processing |
| REFLEXW | Full commercial processing suite, advanced migration | Commercial license required |
| Custom scipy | Custom signal processing, research algorithms | Must build everything from scratch |
Use gprpy when you need open-source GPR processing in Python, batch processing of survey lines, or integration with other Python geoscience tools.
Consider alternatives when you need forward modelling of GPR responses (use GPRMax), advanced migration or commercial-grade processing (use REFLEXW), or highly custom signal processing algorithms (use scipy directly).
Common Workflows
Process raw GPR profile for interpretation
- Import raw data with
gp.gprpyProfile() and importdata() - Apply dewow filter to remove low-frequency drift
- Remove mean trace to eliminate background ringing
- Apply time-power gain or AGC for depth equalization
- Apply bandpass filter to remove noise
- Determine velocity from CMP analysis or material tables
- Convert time axis to depth with
setVelocity() - Apply topographic correction if survey has elevation changes
- Export processed profile as image and/or SEG-Y
References
Scripts