Image segmentation using situational DCT descriptors

It is of utmost importance in multimedia processing to achieve still image segmentation, i.e., to partition images into regions of coherent color and texture. We present a new image segmentation method using a special visual descriptor. For each pixel p, the discrete cosine transform (DCT) of the block centered on p together with its location in the image is employed as its content descriptor thus resulting int a long vector ν&oarr;_p, referred to as situational DCT descriptors (SDDs). A scalar quantization step is then carried out on the DCT component of SDDs to reflect the fact that the human vision system is not of uniform discrimination to details of different frequencies. Next the principal component analysis is conducted to drastically reduce the dimensionality of SDDs. The adaptive K-mean algorithm is then performed to arrive at the region assignment for each pixel. The final partitioning results are obtained after performing the post-processing step. Experimental results using this method demonstrate encouraging performances