Skip to content

Quick Start

Get started with Hyperseed in 5 minutes! This guide walks you through your first analysis.

Prerequisites

Before starting, ensure you have:

  • Installed Hyperseed (pip install hyperseed)
  • Hyperspectral data in ENVI format with white/dark references

Your First Analysis

Step 1: Prepare Your Data

Ensure your dataset follows this structure:

dataset/sample_001/
├── capture/
│   ├── data.raw              # Main hyperspectral data
│   ├── data.hdr              # ENVI header file
│   ├── WHITEREF_data.raw     # White reference
│   ├── WHITEREF_data.hdr
│   ├── DARKREF_data.raw      # Dark reference
│   └── DARKREF_data.hdr

Data Structure

See Data Preparation for detailed information about expected data structure.

Step 2: Run Basic Analysis

Run your first analysis with default settings:

hyperseed analyze dataset/sample_001 --output results.csv

This command: - ✅ Loads the hyperspectral data - ✅ Applies automatic calibration - ✅ Segments individual seeds - ✅ Extracts spectral signatures - ✅ Saves results to results.csv

Step 3: View Results

The analysis generates a CSV file with spectral data:

seed_id,index,centroid_y,centroid_x,area,eccentricity,solidity,band_1000nm,band_1005nm,...
1,0,234.5,156.2,435,0.34,0.92,0.234,0.237,...
2,1,345.6,234.1,421,0.28,0.94,0.229,0.232,...

Each row represents one seed with: - Spatial information: centroid coordinates - Morphological properties: area, eccentricity, solidity - Spectral data: reflectance values for each wavelength band

Adding Visualizations

To generate plots along with the CSV data:

hyperseed analyze dataset/sample_001 \
    --output results.csv \
    --export-plots

This creates four visualization files:

  1. sample_001_distribution.png: Spatial and area distribution
  2. sample_001_segmentation.png: Numbered seeds with boundaries
  3. sample_001_spectra.png: Individual spectral curves
  4. sample_001_spectra_statistics.png: Statistical analysis

Recommended First Run

Always use --export-plots on your first analysis to visually verify the segmentation quality.

For optimal results with seed analysis:

hyperseed analyze dataset/sample_001 \
    --output results.csv \
    --min-pixels 50 \
    --preprocess minimal \
    --export-plots

Parameters explained:

  • --min-pixels 50: Filter out objects smaller than 50 pixels (reduces noise)
  • --preprocess minimal: Light preprocessing optimized for segmentation
  • --export-plots: Generate visualization plots

Batch Processing

To process multiple datasets:

# Process all datasets in the directory
hyperseed batch dataset/ \
    --output-dir results/ \
    --min-pixels 50

This will: - Process all subdirectories in dataset/ - Save results to results/ directory - Apply consistent settings to all datasets

Common Options

Here are the most commonly used options:

Option Description Example
--output Output CSV file path results.csv
--export-plots Generate visualization plots (flag)
--min-pixels Minimum seed size in pixels 50
--preprocess Preprocessing method minimal, standard, advanced, none
--no-outlier-removal Disable automatic outlier removal (flag)
--config Use custom configuration file config.yaml

What's Next?

Customize Your Analysis

Learn how to create custom configurations:

# Generate a configuration template
hyperseed config --output my_config.yaml --preset minimal

# Use your custom configuration
hyperseed analyze dataset/sample_001 --config my_config.yaml

See Configuration Guide →

Advanced Workflows

Explore advanced features:

Command-Line Reference

For complete command documentation:

Example: Complete Workflow

Here's a complete example workflow:

# 1. Analyze a single dataset with visualizations
hyperseed analyze dataset/sample_001 \
    --output results/sample_001.csv \
    --min-pixels 50 \
    --preprocess minimal \
    --export-plots

# 2. Review the plots and adjust if needed

# 3. Process all datasets with the same settings
hyperseed batch dataset/ \
    --output-dir results/ \
    --min-pixels 50 \
    --preprocess minimal

# 4. Results are now in results/ directory
ls results/
# sample_001.csv  sample_002.csv  sample_003.csv  ...

Getting Help

Need assistance?

# Get general help
hyperseed --help

# Get help for specific command
hyperseed analyze --help
hyperseed batch --help

Or check the Troubleshooting Guide for common issues.

Congratulations! You've completed your first Hyperseed analysis. Continue to the Configuration Guide to learn about customization options.