Adaptive Digital Image Sequence Compression Stored by Fixed Cameras Base on Sparse Representation and Dictionary Learning

In this paper, we propose an adaptive Digital image sequence compression stored by fixed cameras via dictionary learning. This method transforms images over sparsely tailored, over-complete dictionaries learned directly from image samples rather than a fixed one, and thus can approximate an image with fewer coefficients. In this research for compression of each frame of the image sequence by our proposed method, we used two different dictionary learning algorithms (RLS-DLA and K-SVD) to compare the operation each of them. Dictionaries are learned in DCT domain and wavelet domain. The results show that the RLS-DLA has better performance than K-SVD. Also the performances of wavelet domain have better results.

Adaptive Digital Image Sequence Compression Stored by Fixed Cameras Base on Sparse Representation and Dictionary Learning Keywords:

image sequence compression , sparse representation , dictionary learning