Interpolating full-resolution color images from color-filterarray (CFA) samples is Image demosaicing. Bayer pattern is popular among various CFA patterns and demosaicing with this Bayer pattern produces high resolution color images. This paper presents a new demosaicing approach for spatially sampled image data perceived through a color filter array, and thereby exploiting the correlation of color components for subsampled image reconstruction. The above method is compatible with wavelet-domain denoising before demosaicing. It is also a general framework to apply existing image denoising algorithms to color filter array data. Compression is through Huffman coding and application of biorthogonal wavelets with the results proving that the proposed method is satisfactory in comparison with other techniques available in the literature.