Edge Detection

Detect sharp spatial transitions such as object boundaries, roads, field edges, structural lines, or coastal boundaries from a selected raster band.

Remote Sensing CoreEnd-User DocumentationTheory + Formula + Parameters

1. What This Feature Does

Detect sharp spatial transitions such as object boundaries, roads, field edges, structural lines, or coastal boundaries from a selected raster band.

Cloud workflow: choose data → configure the scientific method → run → review the resulting layer/report. No programming is required.

2. Recommended Workflow

Select the input band. ↓ Choose an edge detector. ↓ Adjust smoothing/kernel/threshold controls if exposed for that method. ↓ Interpret strong edge responses or binary edges.

3. Theory, Methods & Equations

Canny

Multi-stage detector combining Gaussian smoothing, image gradients, non-maximum suppression, double thresholds, and hysteresis.

G = √(G_x² + G_y²)

θ = atan2(G_y,G_x)

Best used when: clean, thin boundaries are needed and threshold control is important.

Sobel

Uses 3×3 derivative kernels to estimate horizontal and vertical gradients.

G = √(G_x² + G_y²)

Best used when: a robust general gradient is needed.

Scharr

Gradient operator with improved rotational symmetry compared with a standard 3×3 Sobel kernel.

G = √(G_x² + G_y²)

Best used when: small-kernel gradient accuracy is important.

Prewitt

Simple first-derivative gradient operator using uniform 3×3 directional weights.

G = √(G_x² + G_y²)

Best used when: a simple directional edge measure is sufficient.

Roberts

Compact 2×2 diagonal gradient operator.

G ≈ √(G₁² + G₂²)

Best used when: very local/high-frequency edge detection is desired.

Laplacian

Second spatial derivative; highlights rapid intensity changes in all directions.

∇²I = ∂²I/∂x² + ∂²I/∂y²

Best used when: second-derivative edge response is appropriate.

Laplacian of Gaussian (LoG)

Smooths noise with a Gaussian then applies the Laplacian.

LoG = ∇²(G_σ * I)

Best used when: Laplacian sensitivity to noise needs smoothing.

Morphological Gradient

Difference between dilation and erosion; emphasizes object boundaries.

G_morph = dilation(I) − erosion(I)

Best used when: shape-oriented boundaries are useful, especially on discrete/structured imagery.

4. Input Data

Input	Type	Requirement	Description
Input Raster	raster	Required	Raster used for edge detection.

5. Parameters Available in the Application

Parameter	Default	Choices / Range	How to Use It
Input Band	—	—	Raster band used for edge detection. The frontend loads valid bands from the Raster Info application.
Edge Algorithm	canny	Canny, Sobel, Scharr, Prewitt, Roberts, Laplacian, Laplacian of Gaussian (LoG), Morphological Gradient
Gaussian Sigma	1.0	0.0 to 20.0	Optional noise smoothing before edge detection. Canny and LoG use this value directly.
Canny Low Threshold	0.1	0.0 to 1.0	Shown for: canny
Canny High Threshold	0.3	0.0 to 1.0	Shown for: canny
Morphological Kernel Size	3	3 to 31	Shown for: morphological_gradient
Normalize Edge Strength	On	—	Normalize continuous edge-strength outputs to 0–1. Canny is always written as a binary 0/1 raster.

6. Output & Interpretation

The result should be interpreted according to the selected method and the physical meaning of the input data. Preserve source units, coordinate reference information, NoData meaning, acquisition date, and preprocessing level when comparing results.

Scientific interpretation: an algorithm can produce a numerically valid result even when the input data are unsuitable. Always check masks, units, sensor characteristics, spatial resolution and reference data.

7. Best Practices

Use analysis-ready inputs and remove invalid/cloud/noise artifacts that are not part of the target phenomenon.
Choose parameters from the physical scale of the data, not only from visual appearance.
Keep categorical and continuous rasters conceptually separate when selecting interpolation/resampling methods.
Compare the result with the source image and independent reference information.
Document the settings used when results will be compared across dates, sensors or study areas.

← Raster Spectral & Texture Analysis Shoreline Extraction →