This paper presents an adaptive interpolation algorithm for an image resizing based on two separate modes in terms of activity of pixel in an image. First mode corresponds to smooth region and interpolation operation is performed by the conventional n...
This paper presents an adaptive interpolation algorithm for an image resizing based on two separate modes in terms of activity of pixel in an image. First mode corresponds to smooth region and interpolation operation is performed by the conventional nearest neighbor method because this method is more computationally efficient and gives good quality result for pixels having neighboring pixels with pixel intensity in close proximity to each other. Second mode corresponds to complex region. The target pixel is computed based on Newton interpolation method. To preserve the edge details and to reduce the aliasing effect in scaled image, consecutive pixels are separated into two groups. Group with highly relative pixels is selected for effective interpolation. The simulation results show that the proposed algorithm outperforms the conventional methods even though the computational complexity is low.