Skip to main content
. 2026 Jun 16;12(6):263. doi: 10.3390/jimaging12060263
Algorithm 1 FPA-YeastNet for Light-Scattering Classification

  • Require: Light-scattering image Is

  • Ensure: Predicted yeast category y

  • 1:

    Begin

  • 2:

    Resize input image to 224 × 224

  • 3:

    Extract shallow features using Stem layer:

  • 4:

       F0 ← Stem(Is)

  • 5:

    Apply Frequency Pattern Attention (FPA):

  • 6:

       F1 ← FPA(F0)

  • 7:

    Pass features through stacked convolution blocks:

  • 8:

       F2 ← ConvBlock(F1)

  • 9:

    Repeat for 8 convolution blocks

  • 10:

    Perform global average pooling:

  • 11:

     z ← AvgPool(F2)

  • 12:

    Predict yeast category using classifier:

  • 13:

     y ← Classifier(z)

  • 14:

    Return y

  • 15:

    End