A SEMI-SUPERVISED SUPPORT VECTOR MACHINE FOR TEXTURE SEGMENTATION (MA-P1)

Author(s) : Saeid Sanei (King's College London, UK) Tracey Lee (National University of Singapore, Singapore)

Abstract : A novel semi-supervised support vector machines (S3VM) algorithm is developed for segmentation of texture images. The method classifies the uniform-texture regions from the regions of boundaries. The various-order statistics of the textures within a sliding two-dimensional window are measured. K-mean algorithm is used to approximately label the two clusters in the overall image. However, only those clusters, which are very close to the class centres, are labeled. Therefore at this stage we have both the training and the working sets. A non-linear S3VM is then developed to exploit both sets to classify the regions. It is demonstrated that the algorithm is robust and the misclassification error is negligible. However, there may be a minor misplacement of the edges.

Menu