کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
407333 | 678137 | 2012 | 19 صفحه PDF | دانلود رایگان |
![عکس صفحه اول مقاله: Data field-based mechanism for three-dimensional thresholding Data field-based mechanism for three-dimensional thresholding](/preview/png/407333.png)
Introducing more information to improve the segmentation quality was regarded as an effective way, such as three-dimensional Otsu thresholding. However, it should be led to be very time consuming for real-time applications, and the Otsu criterion is questionable in some cases, for example, nondestructive testing. In the paper, a novel mechanism based on data field, originated from physical fields, is proposed for three-dimensional thresholding. Without any explicit criterions, an optimal threshold vector is produced using the self-adaptive evolution of data particles in the data field. And the proposed method has low time complexity. Experimental results, compared with the state-of-art algorithms and the related methods, suggest that the new proposal is efficient and effective.
Journal: Neurocomputing - Volume 97, 15 November 2012, Pages 278–296