Lan Y.,Nanyang Normal University |
Lan Y.,Beijing E COM Technology Co. |
Zhang X.,Nanyang Normal University |
Zhang X.,Beijing E COM Technology Co. |
And 6 more authors.
Information Technology Journal | Year: 2013
With the advantages of being painless, safe and easy-to-use, wireless capsule endoscopy has become a hot research topic in clinical medicine. As tens of thousands of images are generated during an examination, it is impractical for manual image checking. Utilizing computer image processing can greatly enhance image quality, decrease diagnosis time and improve diagnosis accuracy. One of the critical requirements of the employed image processing methods is to extract the lesion areas. Although, many image segmentation methods have been proposed, accurate image segmentation is also a challenging problem which is not completely solved yet. At present, those segmentation approaches can be divided into two main categories: Region based and edge based. It is often difficult to obtain satisfactory results when using only one of these methods in the segmentation of complex pictures. Therefore, in this study, a new image segmentation method using both region and edge information is proposed to combine the two types of methods or information to achieve accurate segmentation results. Experimental results demonstrate the efficacy of the proposed method. © 2013 Asian Network for Scientific Information. Source