Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

General overview of the region based technique.

More »

Fig 1 Expand

Fig 2.

From left to right; Original image with bias field, Bias field corrected image.

The green rectangle indicates the sliding window which is moving vertically from the top to the bottom of the image.

More »

Fig 2 Expand

Fig 3.

Histogram of the image.

(a) Histogram of the real data. Tbf is a threshold that segments the background from the foreground. (b) Histogram of the same data as in (a) but after thresholding the background/foreground pixels.

More »

Fig 3 Expand

Fig 4.

Illustration of 8-connected neighborhood or second order neighborhood of the interested pixel (x,y).

More »

Fig 4 Expand

Fig 5.

Finding two peaks of sorted histogram.

More »

Fig 5 Expand

Fig 6.

Illustration of clustering the sub-regions into three tissue classes.

More »

Fig 6 Expand

Fig 7.

Overall assessment of the proposed method.

More »

Fig 7 Expand

Fig 8.

The first row from left to right indicates original images with 0% noise and 20% bias field, segmentation result and ground truth.

The second row from left to right displays original images with 3% noise and 20% bias field, segmentation result and ground truth and the third row demonstrate original image with 9% noise and 40% bias field, segmentation result and ground truth.

More »

Fig 8 Expand

Fig 9.

From left to right; The first column show original images from the simulated data, second column represents normalized images, the third column demonstrates the results of segmented WM and GM obtained by the proposed method (3% noise and 20% RF level), The last column indicates the ground truth.

More »

Fig 9 Expand

Table 1.

Classification results on Brainweb T1-weighted MRI data in different levels of bias field and noise (pn).

More »

Table 1 Expand

Table 2.

The AOM for WM segmentation of the simulated databases with 3% and 9% noise and 0% and 40% bias field.

More »

Table 2 Expand

Table 3.

The AOM for GM segmentation of the simulated database with 3% and 9% noise and 0% and 40% bias field.

More »

Table 3 Expand

Fig 10.

From left to right; first column presents the brain image from the IBSR dataset, second column displays segmentation results of the proposed method and third column presents the expert segmented images.

More »

Fig 10 Expand

Fig 11.

Comparison of WM and GM segmentation results using proposed technique, Yue method and Ts-kmeans.

The vertical axis displays the AOM and the horizontal one represents the volume number.

More »

Fig 11 Expand