ynsat logo

Alert Types

2 min read

Iterative MAD (iMAD)

Iterative MAD (iMAD) #

How It Works #

Iterative Multivariate Alteration Detection (iMAD) is an advanced change detection technique that categorizes changes based on their similarity, refining results iteratively. It starts by analyzing two multispectral images from different time periods, identifying areas that remain unchanged. By continuously refining these no-change areas, iMAD isolates significant alterations, making it highly effective for detecting subtle environmental shifts.

Common Use Cases #

  • Artisanal Small-Scale Mining (ASM): Identifies and tracks small-scale mining activities, differentiating them from natural terrain changes and categorizing the types of changes.
  • Deforestation & Land Degradation: Detects early-stage forest loss and land degradation
  • Urban Expansion & Infrastructure Development: Tracks new roads, construction, and urban sprawl impacting natural landscapes.
  • Disaster Impact Assessment (Floods, Wildfires, Landslides): Maps affected areas post-disaster for response planning and recovery efforts.
  • Water Body & Wetlands Monitoring: Monitors shrinking lakes, river changes, and wetland degradation due to climate and human activities.
    Class Segmentation

    Class Segmentation #

    How It Works #

    This method assigns each pixel a class probability for one of nine terrain types, using AI-based classification. The algorithm processes satellite imagery to differentiate land cover types, filtering out clouds and shadows to enhance accuracy. The detected classes include:

    • Water
    • Vegetation
    • Urban Areas
    • Snow & Ice
    • Bare Soil
    • Agriculture
    • Wetlands
    • Grasslands
    • Mixed Terrain
    • Cloud & Shadow Masking

    Common Use Cases #

    • Land Cover Monitoring: Track environmental and landscape transformations.
    • Disaster Assessment: Identify areas affected by floods, wildfires, or landslides.
    • Project Classification: Useful for specific projects such as mine sites.
    Insight-Based

    Insight-Based #

    How It Works #

    Insight-based alerts monitor specific environmental indicators over time. Users can set predefined conditions to track variations in elements such as water levels, vegetation changes, and land conditions. When a threshold is exceeded for a specific insight (such as NDVI, or cholophyll levels in water bodies – MAGO index), an alert is triggered, allowing for immediate action. This approach is particularly useful for monitoring sensitive locations and operational risks.

    Common Use Cases #

    • Environmental Monitoring: Detect unexpected changes in vegetation or water bodies.
    • Compliance & Safety: Automate alerts for regulatory thresholds and environmental risks.

    Updated on February 6, 2025