Limits...
An active contour model for the segmentation of images with intensity inhomogeneities and bias field estimation.

Huang C, Zeng L - PLoS ONE (2015)

Bottom Line: The proposed model first appeared as a two-phase model and then extended to a multi-phase one.The experimental results demonstrate the advantages of our model in terms of accuracy and insensitivity to the location of the initial contours.In particular, our method has been applied to various synthetic and real images with desirable results.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Optoelectronic Technology and System of the Education Ministry of China, Chongqing University, Chongqing, 400044, China; Engineering Research Center of Industrial Computed Tomography Nondestructive Testing of the Education Ministry of China, Chongqing University, Chongqing, 400044, China.

ABSTRACT
Intensity inhomogeneity causes many difficulties in image segmentation and the understanding of magnetic resonance (MR) images. Bias correction is an important method for addressing the intensity inhomogeneity of MR images before quantitative analysis. In this paper, a modified model is developed for segmenting images with intensity inhomogeneity and estimating the bias field simultaneously. In the modified model, a clustering criterion energy function is defined by considering the difference between the measured image and estimated image in local region. By using this difference in local region, the modified method can obtain accurate segmentation results and an accurate estimation of the bias field. The energy function is incorporated into a level set formulation with a level set regularization term, and the energy minimization is conducted by a level set evolution process. The proposed model first appeared as a two-phase model and then extended to a multi-phase one. The experimental results demonstrate the advantages of our model in terms of accuracy and insensitivity to the location of the initial contours. In particular, our method has been applied to various synthetic and real images with desirable results.

No MeSH data available.


Comparisons of the segmentation results for a synthetic image with intensity inhomogeneity between Li’s model, LSACM and our model.(a), (e) The original image with red initial contours. (b), (f) The results of Li’s model. (c), (g) The results of LSACM. (d), (h) The results of our model.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4383562&req=5

pone.0120399.g006: Comparisons of the segmentation results for a synthetic image with intensity inhomogeneity between Li’s model, LSACM and our model.(a), (e) The original image with red initial contours. (b), (f) The results of Li’s model. (c), (g) The results of LSACM. (d), (h) The results of our model.

Mentions: Fig. 6 shows the segmentation results for a synthetic image (the image size is 79×75) with higher intensity inhomogeneity obtained with Li’s method, LSACM (code can be downloaded at [45]) and our model. In this experiment, we chose β = 0.007×255×255. The image contains three objects with high light on the left, and the light also causes the boundary to be fuzzy in the lower region of the star-shaped object. For Li’s method, the estimation of the true image J (piecewise constants) may not be accurate in the fuzzy boundary region, and thus, the estimation of the bias field may also not be accurate. The segmentation results obtained with Li’s method, LSACM and the proposed method for different initial contours are shown in columns 2, 3 and 4, respectively. Li’s method fails to segment the object boundary or estimate the fuzzy bias field with high light even when the initial contour across three objects. While LSACM can obtain the right segment result for the initial contour shown in Fig. 6 (a), but for the initial contour as Fig. 6 (e), LSACM get the undesired results even the iterate number over than 2000. The experimental results of our method are more accurate than those obtained with Li’s method and more robustness than LSACM. In Figs. 6(d) and 6(h), the final contours of our method can converge to the correct boundaries precisely. The bias field estimation and corrected image of Li’s model, LSACM and our model are shown in Fig. 7. In the highlight region of the image, the image is not well corrected by Li’s method, LSACM and our method have the similar bias field estimation based on the right segmentation.


An active contour model for the segmentation of images with intensity inhomogeneities and bias field estimation.

Huang C, Zeng L - PLoS ONE (2015)

Comparisons of the segmentation results for a synthetic image with intensity inhomogeneity between Li’s model, LSACM and our model.(a), (e) The original image with red initial contours. (b), (f) The results of Li’s model. (c), (g) The results of LSACM. (d), (h) The results of our model.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4383562&req=5

pone.0120399.g006: Comparisons of the segmentation results for a synthetic image with intensity inhomogeneity between Li’s model, LSACM and our model.(a), (e) The original image with red initial contours. (b), (f) The results of Li’s model. (c), (g) The results of LSACM. (d), (h) The results of our model.
Mentions: Fig. 6 shows the segmentation results for a synthetic image (the image size is 79×75) with higher intensity inhomogeneity obtained with Li’s method, LSACM (code can be downloaded at [45]) and our model. In this experiment, we chose β = 0.007×255×255. The image contains three objects with high light on the left, and the light also causes the boundary to be fuzzy in the lower region of the star-shaped object. For Li’s method, the estimation of the true image J (piecewise constants) may not be accurate in the fuzzy boundary region, and thus, the estimation of the bias field may also not be accurate. The segmentation results obtained with Li’s method, LSACM and the proposed method for different initial contours are shown in columns 2, 3 and 4, respectively. Li’s method fails to segment the object boundary or estimate the fuzzy bias field with high light even when the initial contour across three objects. While LSACM can obtain the right segment result for the initial contour shown in Fig. 6 (a), but for the initial contour as Fig. 6 (e), LSACM get the undesired results even the iterate number over than 2000. The experimental results of our method are more accurate than those obtained with Li’s method and more robustness than LSACM. In Figs. 6(d) and 6(h), the final contours of our method can converge to the correct boundaries precisely. The bias field estimation and corrected image of Li’s model, LSACM and our model are shown in Fig. 7. In the highlight region of the image, the image is not well corrected by Li’s method, LSACM and our method have the similar bias field estimation based on the right segmentation.

Bottom Line: The proposed model first appeared as a two-phase model and then extended to a multi-phase one.The experimental results demonstrate the advantages of our model in terms of accuracy and insensitivity to the location of the initial contours.In particular, our method has been applied to various synthetic and real images with desirable results.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Optoelectronic Technology and System of the Education Ministry of China, Chongqing University, Chongqing, 400044, China; Engineering Research Center of Industrial Computed Tomography Nondestructive Testing of the Education Ministry of China, Chongqing University, Chongqing, 400044, China.

ABSTRACT
Intensity inhomogeneity causes many difficulties in image segmentation and the understanding of magnetic resonance (MR) images. Bias correction is an important method for addressing the intensity inhomogeneity of MR images before quantitative analysis. In this paper, a modified model is developed for segmenting images with intensity inhomogeneity and estimating the bias field simultaneously. In the modified model, a clustering criterion energy function is defined by considering the difference between the measured image and estimated image in local region. By using this difference in local region, the modified method can obtain accurate segmentation results and an accurate estimation of the bias field. The energy function is incorporated into a level set formulation with a level set regularization term, and the energy minimization is conducted by a level set evolution process. The proposed model first appeared as a two-phase model and then extended to a multi-phase one. The experimental results demonstrate the advantages of our model in terms of accuracy and insensitivity to the location of the initial contours. In particular, our method has been applied to various synthetic and real images with desirable results.

No MeSH data available.