Image Enhancement Scenario

Warning

This section is under construction! Some information might be missing.

I Improving image segmentation with image enhancement methods

  1. Browser

    • Load image YR01_t40_crop.inr.gz

  2. Segmentation

    • algo seededWatershedSegmentation

    • h_min -> 2

    • background -> 1

    • volume_threshold -> ~1000

    • Apply

    • Show segmentation defects (hole on the surface)

  3. Preprocess

    • algo boundaryEdgeEnhancement

    • Edge Intensity: max

    • Threshold: 15

    • on image

  4. Segmentation

    • algo seededWatershedSegmentation

    • h_min -> 2

    • background -> 1

    • volume_threshold -> ~1000

    • Apply

    • Show segmentation defects

II Improving image segmentation With imageEnhancement package:

  1. Preprocess

    • algo anisotropic3dImageEnhancement

    • iterations -> 15

    • tensor -> 10

  2. Binarization

    • algo lsmContour

    • stop criterion: 0.005

    • smoothing OFF

    • thresholds 0-8

  3. Preprocess

    • algo edgeEnhancementBinaryimage

    • on ani image without binary mask

    • drop binary mask and apply to show difference

  4. Segmentation

    • algo seededWatershedSegmentation

    • h_min -> 2

    • background -> 1

    • volume_threshold -> ~1000

    • Apply

    • Show segmentation enhancement

Videos

First Scenario Second Scenario