Upload a micrograph, apply Otsu thresholding, and calculate porosity as area fraction. Client-side processing — images never leave your browser.
Try it out
Load example porosity data to see the full workflow
Upload a micrograph
Drag & drop or click to browse — any image format
When to use
Do not use for
Otsu's method assumes a bimodal histogram. If your histogram is unimodal (e.g., very low porosity < 2%), multimodal, or has a long tail, the automatic threshold may be suboptimal. Always inspect the histogram overlay and adjust manually if the binary image doesn't visually match your assessment of pore locations.
JPEG compression introduces intensity variations at block boundaries (8x8 pixel blocks) that can be misclassified as pores at certain thresholds. Use PNG or TIFF images when possible. If you must use JPEG, apply a slight Gaussian blur before thresholding (not available in this tool — preprocess in ImageJ or GIMP).
The single biggest source of error in porosity measurement is usually field-of-view selection bias, not threshold choice. Porosity can vary 2-5x across a sample cross-section. Analyze at least 5 non-overlapping fields at consistent magnification and report the mean and standard deviation.
At low magnification, small pores below the resolution limit are missed, underestimating porosity. At very high magnification, you see fewer pores per field and the sampling variance increases. Choose a magnification where the smallest pores of interest span at least 3-5 pixels.
Grayscale conversion uses ITU-R BT.601 luma weighting (0.299R + 0.587G + 0.114B). Otsu's threshold is computed by maximizing between-class variance over all 256 intensity levels. Porosity is calculated as the ratio of pore pixels to total pixels. The 95% confidence interval is derived from the binomial standard error: SE = sqrt(p(1-p)/n), where p is the area fraction and n is the total pixel count. All processing uses the browser Canvas API — no server upload.
Last validated 2026-04-05. Calculations are designed for planning and documentation support; verify procurement decisions against manufacturer specifications or institutional SOPs.
ConductScience Porosity Estimator (v1.0). ConductScience, Inc. 2026. Available at: https://conductscience.com/tools/porosity-estimator
Otsu N. A threshold selection method from gray-level histograms. IEEE Trans Syst Man Cybern. 1979;9(1):62-66. doi:10.1109/TSMC.1979.4310076
ASTM E1382-97. Standard Test Methods for Determining Average Grain Size Using Semiautomatic and Automatic Image Analysis. ASTM International. doi:10.1520/E1382-97R23
Porosity — the volume fraction of void space in a solid — is one of the most fundamental microstructural parameters in materials science. It governs mechanical properties (strength, stiffness, fatigue life), transport properties (permeability, diffusivity), and functional performance (thermal insulation, biocompatibility).
Common measurement techniques include Archimedes' method (buoyancy), mercury intrusion porosimetry, gas adsorption (BET), micro-CT, and image analysis. Image-based methods using optical or electron micrographs provide spatial information that bulk methods cannot: pore size distribution, shape, connectivity, and spatial arrangement.
The simplest image analysis approach is global thresholding — converting a grayscale micrograph to binary (pore vs. matrix) using an intensity cutoff, then counting the fraction of pore pixels. Otsu's method automates the threshold selection step.
Nobuyuki Otsu published his automatic thresholding method in 1979. The algorithm assumes the image histogram is bimodal — two overlapping distributions representing two phases (e.g., pores and matrix). It exhaustively tests every possible threshold from 0 to 255 and selects the one that maximizes the between-class variance:
where ω₀ and ω₁ are the class probabilities and μ₀ and μ₁ are the class means at threshold t.
Otsu's method works well when the histogram is clearly bimodal with similar class sizes. It can struggle with highly skewed distributions (very low or very high porosity), uneven illumination, or multimodal histograms. In these cases, manual threshold adjustment or local adaptive thresholding may be needed.
Several ASTM and ISO standards govern image-based porosity and phase fraction measurement:
For formal reporting, multiple fields of view (typically 5+) should be analyzed, and results reported as mean ± standard deviation with the number of fields, magnification, and thresholding method documented.
AI-powered image analysis for porosity, particle counting, and microstructure characterization
Browse EquipmentMeasure grain size from metallographic micrographs using the Heyn intercept method
Browse EquipmentMetallographic microscopes and imaging systems for specimen analysis
Browse QuoteGet a quote on equipment or talk to the ConductScience team.
Request quote