A novel region-based active contour approach relying on local and lobal information

Aitfares, W.; Herbulot, Ariane; Devy, Michel; Bouyakhf, El Houssine; Regragui, Fakhita

doi:10.1109/icip.2011.6115599

Cited by 4 publications

(24 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This sensitivity is due to the fact that the information is extracted locally. Thus, with the aim to enable segmenting the object in the case of presence of heterogeneity and using an inadequate AC initialization in noisy image, we proposed an approach in [2] that combines both benefits of local-based and global-based techniques using local-based selection in one side of the AC and global-based selection in its other side at each point along the contour. Indeed, depending on…”

Section: Fig1mentioning

confidence: 99%

“…where the heterogeneity appears (in ROI or in background), two cases have been investigated in our previous work [2]. Unlike the standard global region-based approaches or the local region-based approach, the approach presented in [2] shows more robustness both against noise and contour initialization thanks to the global statistical information and against heterogeneous appearances thanks to the use of local statistical information.…”

Section: Fig1mentioning

confidence: 99%

“…In order to overcome the drawbacks of each approach alone (global region-based and local region-based) and cope with the aforementioned limitations, namely, the presence of the heterogeneity (in ROI or in background) in noisy image and the use of an inadequate contour initialization, we developed an approach that combines the local-based and global-based image statistics [2] where the global and local information are both region-based.…”

Section: Introductionmentioning

confidence: 99%

“…Likewise, Sum and Cheung [13] have proposed a combined energy minimization based on the sum of a global region-based and energy based local image contrast. In our approach [2], the hybrid region-based combination depends on where the heterogeneity appears (in ROI or in background). For instance, if the object is heterogeneous compared to the background, the image statistics are extracted locally inside the AC and globally outside it using the technique Local IN-Global OUT presented in [2]; whereas, in the case of heterogeneous background, image statistics are extracted locally outside the AC and globally inside it using the technique Global IN-Local OUT presented in [2].…”

Section: Introductionmentioning

confidence: 99%

“…In our approach [2], the hybrid region-based combination depends on where the heterogeneity appears (in ROI or in background). For instance, if the object is heterogeneous compared to the background, the image statistics are extracted locally inside the AC and globally outside it using the technique Local IN-Global OUT presented in [2]; whereas, in the case of heterogeneous background, image statistics are extracted locally outside the AC and globally inside it using the technique Global IN-Local OUT presented in [2]. The approach presented in [2] was found to be more efficient than the global region-based and the local region-based methods.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A robust region-based active contour for object segmentation in heterogeneous case

Aitfares

Bouyakhf

Regragui

et al. 2014

Pattern Recognit. Image Anal.

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a new approach using a local version of the region-based active contour for object segmentation in images presenting heterogeneity in both the object of interest and the background. The local version was recently developed to deal with heterogeneous appearances by relying on extracting the local instead of global image statistics where the local extracted area is defined by using a disk, sketched at each point along the active contour, with a constant radius size inside and outside the contour. However, the use of a constant radius may prevent the active contour getting more information from its neighborhood and thus being trapped by undesired boundaries. To avoid this error segmentation, the local extracted area using our proposed approach is determined based on using two different radii to extract separately the interior and the exterior local information of the active contour. Using synthetic and real images, our approach shows an improvement in term of computation time and outperforms the conventional methods in noisy images presenting heterogeneity in both the object of interest and the background and using inadequate contour initialization.

show abstract

Section: Fig1mentioning

confidence: 99%

Section: Fig1mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%