Skip to content
E3TAM

Casting-pore masks and regional area fraction

Solution concept

Casting pores can mix with glare and natural texture; a defect box alone does not describe their total visible area or local clustering.

Artificial intelligenceresearch-backed conceptraw image / analysis separationpixel / instance mask
Information delivered

Attach the visible-pore mask and area fraction to the surface record; move an out-of-limit coupon to the review nest. Internal porosity is not assessed.

Casting-pore masks and regional area fractionILLUSTRATIVE SIMULATION
Casting-pore masks and regional area fraction
  1. 01 Position
  2. 02 Raw image
  3. 03 Mask generation
  4. 04 Region calculation
  5. 05 Decision
—

Register the ROI; occluded areas are excluded from visible-area calculations.

Target-mask area
—
Connected regions / objects
—
Decision metric
—

Scenes are illustrative simulations; they are not customer references or performance guarantees.

Solution approach

Pixel or instance masks are derived from the raw image; a coded legend separates background and target regions. Area fractions use only the visible ROI. Separate pore pixels from the main surface and show connected pore count and visible ROI fraction separately. Depth is not inferred from the mask.

Process steps

  1. 01

    Position

    Register the ROI; occluded areas are excluded from visible-area calculations.

  2. 02

    Raw image

    Keep the controlled raw image in a separate panel.

  3. 03

    Mask generation

    Coded colours show target classes or object identities.

  4. 04

    Region calculation

    Derive area fraction, connected regions and the decision metric from the mask.

  5. 05

    Decision

    Hold an out-of-limit sample for review with its local region marked.

Vision system design

Let us assess your application together

We test your sample in our laboratory and report on resolution, optics and lighting selection.

+90 212 275 2280 – 81
Esentepe / İstanbul · İkitelli OSB Laboratory
Request Feasibility