Image classification by a two-dimensional hidden Markov model

Jia Li, Amir Najmi, Robert M. Gray

Research output: Contribution to journalArticlepeer-review

243 Scopus citations


For block-based classification, an image is divided into blocks, and a feature vector is formed for each block by grouping statistics extracted from the block. Conventional block-based classification algorithms decide the class of a block by examining only the feature vector of this block and ignoring context information. In order to improve classification by context, an algorithm is proposed that models images by two dimensional (2-D) hidden Markov models (HMM's). The HMM considers feature vectors statistically dependent through an underlying state process assumed to be a Markov mesh, which has transition probabilities conditioned on the states of neighboring blocks from both horizontal and vertical directions. Thus, the dependency in two dimensions is reflected simultaneously. The HMM parameters are estimated by the EM algorithm. To classify an image, the classes with maximum a posteriori probability are searched jointly for all the blocks. Applications of the HMM algorithm to document and aerial image segmentation show that the algorithm outperforms CARTTM, LVQ, and Bayes VQ.

Original languageEnglish (US)
Pages (from-to)517-533
Number of pages17
JournalIEEE Transactions on Signal Processing
Issue number2
StatePublished - 2000

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering


Dive into the research topics of 'Image classification by a two-dimensional hidden Markov model'. Together they form a unique fingerprint.

Cite this