Results of Synthetic System Boundary Estimation

Figure 9.31 shows in the FCM classification system all the steps for the left and right lumen detection, identification, and boundary estimation process in the synthetic images. We look at large noise protocol as an example below with noise level a2 = 500. In the first row the left image shows the synthetically generated image. In the first row the right image shows the image after it has been smoothed by the Perona-Malik smoothing function. In the second row the left image shows the classified image after the image has gone through the FCM

1

1

1

1

1

2

3

4

1

1

1

1

5

6

7

8

1

1

1

1

1

10

11

13

14

15

Assigning unique IDs

Binary Image

Assigning unique IDs

1

1

3

3

1

1

1

1

1

1

3

3

1

1

1

1

1

1

1

1

1

1

1

1

1

1

Label Propagation: left to right and top to bottom

Label Propagation: right to left and bottom to top

Label Propagation: left to right and top to bottom

Label Propagation: right to left and bottom to top

Figure 9.28: Region identification: ID Propagation. The top left image is a binary image with a value of 1 assigned to each of the white pixels. Each of the white pixels are assigned a unique value in the top right image. The left to right and top to bottom label propagation propagates the labels of value 1 and 3, and the result is the bottom left image. Then, the right to left and bottom to top label propagation propagates the label value of 1 to the pixels having a label value of 3. The result is the bottom right image, in which the connected white pixels have all the same label values of 1.

Figure 9.29: Region identification: CCA. The input image has the lumens detected, but they are all the same color. Connected component analysis (CCA) identifies the lumens by labeling each with a different color.

Gray Scale Image

Ideal Boundary Image

Dilate Ideal Boundary and Convert to Red Color

Fusion of 3 images

Estimated Boundary Image

Dilate Ideal Boundary and Convert to Red Color

Red Boundary Image

Color Overlay Image

Figure 9.30: Color overlay block. The first input is a gray scale image. The second input is the ideal boundary image. This image is dilated and converted to a red color, resulting in a red ideal boundary image. The third input is the estimated boundary image. This image is dilated and converted to a green color, resulting in green estimated boundary image. These three images are fused to produce a color overlay image.

classification system. In the second row the right image shows the binarization of the image after selecting only the core class for binarization (K = 1). In the third row the left image shows the binarization of the image after selecting the core class and the edge classes for binarization (K > 1). In the third row the right image shows the image (K = 1) after the labeling of CCA. In the fourth row the left image shows the image (K > 1) after the labeling of CCA. In the fourth row the right image shows the image (K = 1) after the labeling of assign ID. In the fifth row the left image shows the image (K > 1) after the labeling of assign ID. In the fifth row the right image shows the computer-estimated boundary of the image (K = 1) using the region-to-boundary algorithm. In the sixth row the left image shows the computer-estimated boundary of the image (K > 1) using the region-to-boundary algorithm. In the sixth row the right image shows the original image on which is overlayed the ideal ground truth boundary, the artifacted boundary (K = 1), and the corrected boundary (K > 1).

Figure 9.32 shows in the MRF classification system all the steps for the left and right lumen detection, identification, and boundary estimation process in the synthetic images. We look at large noise protocol as an example below with

0 0

Post a comment