K-cluster based reconstruction for Compressive Sensing

In this paper, we extend the existing CS by including the prior knowledge of K-cluster valued intensities available for an image. In order to reduce the measurement numbers, we then propose in this paper K-cluster based reconstruction approach for Compressive Sensing (CS), by incorporating the if-means algorithm in recovery algorithm to model the prior of K-cluster valued intensities for digital images. Finally, the performance of conventional CS and if-cluster based CS is evaluated using some natural images and background subtraction images.