Image identification and segmentation according to random-field models
作者:
Highlights:
•
摘要
Some images have texture variations which are most easily characterized by random-field models. The identification of the model which evoked a given such image is described. The method involves a readily computable criterion which is a refinement of those previously used for image identification. The general method is also shown to be suitable for the more specialized problem of image segmentation, a preprocessing operation fundamental to many techniques in artificial intelligence.
论文关键词:random-field model,image identification,artificial intelligence
论文评审过程:Available online 14 August 2003.
论文官网地址:https://doi.org/10.1016/0262-8856(83)90018-5