Skip to content

Visualization

After processing your InSAR data, you have two powerful visualization options: an interactive GIS map viewer and detailed time-series plots. This guide covers both tools and their capabilities.

Interactive GIS Map Viewer

The gis_map_viewer.py script provides a web-based interface for exploring your change detection results.

Prerequisites

Ensure these files are present in your project directory:

  • change_detection_results.csv (required)
  • binary_detection.tif (optional - GeoTIFF overlay)
  • Basisdata_..._FGDB/ folders (optional - historical data)

Launching the Viewer

Bash
1
python gis_map_viewer.py

This starts a local web server and opens the map in your default browser.

Map Features

Interactive Elements

  • Click Points: Click any detected change point for detailed information
  • Layer Toggle: Show/hide different data layers
  • Zoom Controls: Navigate to areas of interest
  • Legend: Understand color coding and symbols

Information Panels

Each point displays: - Change magnitude and direction - Detection confidence - Temporal information - Coherence statistics - Original data source

Layer Options

  • Change Points: Primary detection results
  • GeoTIFF Overlay: Binary detection visualization
  • Historical Data: Background reference layers
  • Base Maps: Satellite imagery and street maps

Finding Significant Changes: - Use layer filters to show only high-confidence detections - Look for clustering patterns in urban areas - Check coastal and industrial regions

Exploring Details: - Click points for popup information - Use zoom to examine spatial patterns - Compare different time periods if available

Example Visualization

Here's an example of the detailed time-series analysis you can generate:

Example Change Detection Plot

Figure: Time-series analysis showing significant ground motion change detected at point 20201 on Track 139. The plot displays displacement over time with clear indication of the change point detection.

This visualization demonstrates: - Temporal progression of ground displacement measurements - Change point detection algorithm results
- Statistical confidence intervals and thresholds - Before/after analysis of the detected change

Time-Series Plots

The insar-visualizer.py script generates detailed static plots for specific points.

Basic Usage

Bash
1
python insar-visualizer.py --results change_detection_results.csv --num-points 5 --selection mixed

Parameters

--results

Path to the change detection results file.

Bash
1
--results /path/to/change_detection_results.csv

--num-points

Number of points to visualize for each category.

Bash
1
--num-points 10

--selection

Which points to visualize based on change magnitude:

  • highest: Largest change magnitude
  • lowest: Smallest change magnitude
  • middle: Medium change magnitude
  • mixed: Combination of all categories
  • all: All significant points
Bash
1
--selection highest

--output-dir

Directory for saved plots (default: visualizations/).

Bash
1
--output-dir /path/to/plots

Plot Types Generated

Individual Point Plots

Each plot shows: - Time series: InSAR measurements over time - Change point: Detected change moment - Trend lines: Before and after velocities - Confidence intervals: Data uncertainty - Coherence overlay: Data quality indicators

Summary Statistics

  • Change distribution: Histogram of detected changes
  • Spatial patterns: Geographic distribution
  • Temporal patterns: When changes occurred
  • Quality metrics: Coherence and confidence statistics

Examples

High-Impact Changes

Bash
1
2
3
4
5
python insar-visualizer.py \
    --results change_detection_results.csv \
    --num-points 3 \
    --selection highest \
    --output-dir high_impact_plots

Comprehensive Overview

Bash
1
2
3
4
5
python insar-visualizer.py \
    --results change_detection_results.csv \
    --num-points 20 \
    --selection all \
    --output-dir comprehensive_analysis

Quality Assessment

Bash
1
2
3
4
5
python insar-visualizer.py \
    --results change_detection_results.csv \
    --num-points 5 \
    --selection mixed \
    --output-dir quality_check

Interpreting Visualizations

Map Viewer Insights

Spatial Patterns: - Clustered points may indicate localized processes - Linear patterns might suggest infrastructure effects - Isolated points could be individual events

Magnitude Distribution: - Large changes (red points) need immediate attention - Medium changes (yellow) require investigation - Small changes (green) may be monitoring priorities

Time-Series Analysis

Change Detection Quality: - Sharp transitions: Clear change events - Gradual transitions: Slow-onset processes - Noisy data: Lower confidence detections - Multiple changes: Complex temporal behavior

Velocity Analysis: - Acceleration: Increasing subsidence/uplift rates - Deceleration: Slowing ground motion - Reversal: Change from subsidence to uplift (or vice versa)

Advanced Visualization

Custom Analysis

For specialized analysis, you can:

  1. Filter Results: Select specific confidence or magnitude ranges
  2. Temporal Subsets: Focus on particular time periods
  3. Spatial Regions: Analyze specific geographic areas
  4. Quality Thresholds: Use only high-coherence data

Integration with GIS

Export results for use in professional GIS software:

Python
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
import pandas as pd
import geopandas as gpd

# Load results
df = pd.read_csv('change_detection_results.csv')

# Convert to GeoDataFrame
gdf = gpd.GeoDataFrame(
    df, 
    geometry=gpd.points_from_xy(df.easting, df.northing),
    crs='EPSG:3035'  # Using EPSG:3035 for accurate European analysis
)

# Export to shapefile
gdf.to_file('change_points.shp')

Troubleshooting

Map Viewer Issues

Browser won't open: - Manually navigate to http://localhost:5000 - Check firewall settings - Try a different browser

Missing data layers: - Verify required files are in the correct directory - Check file permissions - Confirm file formats are correct

Slow performance: - Use fewer data points with filtering - Close other browser tabs - Restart the viewer

Plot Generation Issues

No plots generated: - Verify results file exists and has data - Check output directory permissions - Ensure sufficient disk space

Poor plot quality: - Increase figure resolution in code - Use different output formats (PNG, PDF) - Adjust plot styling parameters

Memory errors: - Process fewer points at once - Use --selection to limit data - Close other applications

Best Practices

Workflow Integration

  1. Start with Map Viewer: Get overview of spatial patterns
  2. Identify Key Points: Focus on high-magnitude changes
  3. Generate Time Series: Analyze temporal behavior
  4. Document findings: Save plots and screenshots
  5. Validate Results: Cross-check with known events

Quality Control

  • Always examine high-confidence detections first
  • Compare multiple visualization methods
  • Look for consistency in spatial and temporal patterns
  • Validate against external information when available

Next Steps