A multiresolution hybrid neuro-Markovian image modeling and segmentation

Piotr Wilinski,Basel Solaiman,Alain Hillion,Witold Czarnecki

Image Processing, 1996. Proceedings., International Conference（1996）

引用 2|浏览6

暂无评分

摘要

A new textured image model is proposed. This model is described by means of a neuro-Markovian hybrid approach using a Kohonen map and a hidden Markov model (HMM). Each state of the HMM describes one resolution level in the image. The change of the state corresponds to the change of image analysis resolution level. The HMM observation space is composed of clusters which are estimated using a Kohonen map. The second role of the Kohonen algorithm is to achieve a segmentation which is done in parallel with the one proceeded by a Viterbi algorithm. The results given by the both algorithms present some complementarity

查看译文

关键词

hidden Markov models,image classification,image resolution,image segmentation,image texture,parameter estimation,self-organising feature maps,HMM observation space,Kohonen map,clusters,contextual classification,hidden Markov model,image segmentation,multiresolution image analysis,neuro-Markovian hybrid approach,textured image model

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要