CNF:Weighted Mesh To Image: Difference between revisions
Jump to navigation
Jump to search
Spyridon97 (talk | contribs) |
Spyridon97 (talk | contribs) |
||
| Line 29: | Line 29: | ||
=== KNN, number of points = 4, power = 1.5 === | === KNN, number of points = 4, power = 1.5 === | ||
<gallery mode="packed" heights= | <gallery mode="packed" heights=400px> | ||
File:NT-image.png | File:NT-image.png | ||
File:NT-a_1-n_4-p_1.5.png | File:NT-a_1-n_4-p_1.5.png | ||
| Line 40: | Line 40: | ||
=== Cell Points, number of points = 4, power = 1.5 === | === Cell Points, number of points = 4, power = 1.5 === | ||
<gallery mode="packed" heights= | <gallery mode="packed" heights=400px> | ||
File:NT-image.png | File:NT-image.png | ||
NT-a_2-n_4-p_1.5.png | NT-a_2-n_4-p_1.5.png | ||
| Line 51: | Line 51: | ||
=== Intersection of KNN and Cell points, number of points = 5, power = 1.5 === | === Intersection of KNN and Cell points, number of points = 5, power = 1.5 === | ||
<gallery mode="packed" heights= | <gallery mode="packed" heights=400px> | ||
File:NT-image.png | File:NT-image.png | ||
File:NT-a_3-n_5-p_1.5.png | File:NT-a_3-n_5-p_1.5.png | ||
| Line 62: | Line 62: | ||
=== Cell points with probing weights, number of points = 4, power = 1.5 === | === Cell points with probing weights, number of points = 4, power = 1.5 === | ||
<gallery mode="packed" heights= | <gallery mode="packed" heights=400px> | ||
File:NT-image.png | File:NT-image.png | ||
File:NT-a_4-n_4-p_1.5.png | File:NT-a_4-n_4-p_1.5.png | ||
Revision as of 11:49, 10 June 2020
Repository
Code
Summer 2020
Create a new image using the weights of:
- k-nearest-neighbors points as values in a weighted-mean, were weight[i] = 1 / distance(pixel-point, neighbor-point[i]) ^ power.
- cell points as values in a weighted-mean, were weight[i] = 1 / distance(pixel-point, cell-point[i]) ^ power.
- Note: If the pixel point is not inside a cell, then the algorithm `1` will be used.
- the intersection set of k-nearest-neighbors points and cell points as values in a weighted-mean, were weight[i] = 1 / distance(pixel-point, intersection-point[i]) ^ power.
- Note 1: If the Intersection set is empty, then the algorithm `2` will be used.
- Note 2: If the pixel point is not inside a cell, then the algorithm `1` will be used.
- cell points as values in a weighted-mean, were weight[i] = probe-value[i] based on the location of each pixel point inside a cell. Probing values are result of FindCell function.
- Note: If the pixel point is not inside a cell, then the algorithm `1` will be used.
Comparison Metrics:
- Mean Weighted pixel value difference
- Average Hausdorff Distance (itk filter)
- Similarity Index (itk filter)
Examples
Summer 2020
Left: Mesh, Middle: Algorithm Result. Right: Image
KNN, number of points = 4, power = 1.5
- Mean Weighted pixel value difference: 0.000395461
- Average Hausdorff Distance: 0.030885
- Similarity Index: 0.986463
Cell Points, number of points = 4, power = 1.5
- Mean Weighted pixel value difference: 0.0002258
- Average Hausdorff Distance: 0.030885
- Similarity Index: 0.986463
Intersection of KNN and Cell points, number of points = 5, power = 1.5
- Mean Weighted pixel value difference: 0.000273839
- Average Hausdorff Distance: 0.030885
- Similarity Index: 0.986463
Cell points with probing weights, number of points = 4, power = 1.5
- Mean Weighted pixel value difference: 9.17551e-05
- Average Hausdorff Distance: 0.030885
- Similarity Index: 0.986463