Research Paper | Computer Science & Engineering | India | Volume 4 Issue 7, July 2015
Improved Hidden Markov Random Field and its Expectation Maximization Algorithm for Image Segmentation
Pooja Kalra | Amit Sharma 
Abstract: Image segmentation is used to understand images and extract information or objects from them. Unsupervised image segmentation is an incomplete data problem as the number of class labels and model parameters are unknown. In this paper, we have analyzed HMRF that defines a probability measure on the set of all possible labels and select the most likely one for unsupervised image segmentation. As HMRF model parameters are unknown, to handle this problem Expectation Maximization algorithm is used. The modulated intensity along with edge map, gray level pixel of intensities and initial labels from K-means information is provided to HMRF-EM algorithm for segmentation. The results obtained by proposed Improved HMRF-EM algorithm are compared with the HMRF-EM algorithm. on the basis of PSNR and Improved HMRF-EM will result in better segmentation quality.
Keywords: Image segmentation, HMRF, Expectation Maximization, RMSC, PSNR
Edition: Volume 4 Issue 7, July 2015,
Pages: 311 - 315
How to Cite this Article?
Pooja Kalra, Amit Sharma, "Improved Hidden Markov Random Field and its Expectation Maximization Algorithm for Image Segmentation", International Journal of Science and Research (IJSR), Volume 4 Issue 7, July 2015, pp. 311-315, https://www.ijsr.net/get_abstract.php?paper_id=SUB156207
How to Share this Article?
Similar Articles with Keyword 'Image segmentation'
Image Segmentation using Biogeography based Optimization and its Comparison with K Means Clustering
Survey on Various Image Segmentation Techniques