Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

NSCT Based Local Enhancement for Active Contour Based Image Segmentation Application


Affiliations
1 Department of Electronics Engineering, Sardar Vallbhbhai National Institute of Technology, Surat, Gujarat, India
     

   Subscribe/Renew Journal


Because of cross-disciplinary nature, Active Contour modeling techniques have been utilized extensively for the image segmentation. In traditional active contour based segmentation techniques based on level set methods, the energy functions are defined based on the intensity gradient. This makes them highly sensitive to the situation where the underlying image content is characterized by image nonhomogeneities due to illumination and contrast condition. This is the most difficult problem to make them as fully automatic image segmentation techniques. This paper introduces one of the approaches based on image enhancement to this problem. The enhanced image is obtained using NonSubsampled Contourlet Transform, which improves the edges strengths in the direction where the illumination is not proper and then active contour model based on level set technique is utilized to segment the object. Experiment results demonstrate that proposed method can be utilized along with existing active contour model based segmentation method under situation characterized by intensity non-homogeneity to make them fully automatic.

Keywords

Segmentation, Nonsubsampled Contourlet Transform (NSCT) and Active Contour Model (AVM).
Subscription Login to verify subscription
User
Notifications
Font Size

Abstract Views: 245

PDF Views: 0




  • NSCT Based Local Enhancement for Active Contour Based Image Segmentation Application

Abstract Views: 245  |  PDF Views: 0

Authors

Hiren Mewada
Department of Electronics Engineering, Sardar Vallbhbhai National Institute of Technology, Surat, Gujarat, India
Suprava Patnaik
Department of Electronics Engineering, Sardar Vallbhbhai National Institute of Technology, Surat, Gujarat, India

Abstract


Because of cross-disciplinary nature, Active Contour modeling techniques have been utilized extensively for the image segmentation. In traditional active contour based segmentation techniques based on level set methods, the energy functions are defined based on the intensity gradient. This makes them highly sensitive to the situation where the underlying image content is characterized by image nonhomogeneities due to illumination and contrast condition. This is the most difficult problem to make them as fully automatic image segmentation techniques. This paper introduces one of the approaches based on image enhancement to this problem. The enhanced image is obtained using NonSubsampled Contourlet Transform, which improves the edges strengths in the direction where the illumination is not proper and then active contour model based on level set technique is utilized to segment the object. Experiment results demonstrate that proposed method can be utilized along with existing active contour model based segmentation method under situation characterized by intensity non-homogeneity to make them fully automatic.

Keywords


Segmentation, Nonsubsampled Contourlet Transform (NSCT) and Active Contour Model (AVM).