TY - GEN
T1 - MODEL-BASED TEXTURE SEGMENTATION.
AU - Chatterjee, Shankar
AU - Chellappa, Rama
PY - 1987
Y1 - 1987
N2 - Segmentation of images containing textured regions is useful in many applications. In the current work, the authors propose a model-based texture segmentation scheme where homogeneous regions of the image are modeled by non-causal Gaussian Markov random fields. The segmentation is performed pixelwise, requiring simple computations. Assuming another Markov random field which describes transitions from one region type to another, the segmented output is updated till no further changes occur. Results of application to real textures are shown.
AB - Segmentation of images containing textured regions is useful in many applications. In the current work, the authors propose a model-based texture segmentation scheme where homogeneous regions of the image are modeled by non-causal Gaussian Markov random fields. The segmentation is performed pixelwise, requiring simple computations. Assuming another Markov random field which describes transitions from one region type to another, the segmented output is updated till no further changes occur. Results of application to real textures are shown.
UR - http://www.scopus.com/inward/record.url?scp=0023173147&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0023173147&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:0023173147
SN - 0818608161
T3 - Conference Record - Asilomar Conference on Circuits, Systems & Computers
SP - 235
EP - 239
BT - Conference Record - Asilomar Conference on Circuits, Systems & Computers
PB - IEEE
ER -