An Irregular Shaped Region of Interest Based Intelligent Image Compression Using Direction Adaptive Filter Banks
N. Udaya Kumar1, E. V. Krishna Rao2, M. Madhavi Latha3, K. Padma Vasavi4
1N. Udaya Kumar,  ECE Dept, SRKR Engineering College, Bhimavaram, India.
2E. V. Krishna Rao,  ECE Dept, KL University, VIjayawada, India.
3M. Mahavai Latha,  ECE Dept, JNTUH, Hyderabad, India.
Manuscript received on January 25, 2014. | Revised Manuscript received on February 11, 2014. | Manuscript published on February 28, 2014. | PP: 43-48  | Volume-3, Issue-3, February 2014. | Retrieval Number:  C2542023314/2013©BEIESP

Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract:  In order to transmit real time images over band limited channels it is very much essential to consider an intelligent transmission system. An irregular shaped Region of Interest (ROI) based Intelligent Compression using Direction Adaptive Filter Banks is presented in this paper. As most of the real time images have irregular boundaries, Direction Adaptive filter banks are used to extract the boundaries of irregular shaped regions. From the edge maps thus obtained, the ROI is extracted by region growing segmentation. The ROI is encoded using a lossless arithmetic encoder and the background is encoded using the most popular SPIHT encoder. At the receiving end, the decoded data from the ROI and the background is added to obtain the final decoded image. The proposed technique is compared against the SPIHT algorithm. The comparisons are made with respect to image fidelity, Peak Signal to Noise Ratio (PSNR) in dB and the bit error rate. The results obtained from the proposed algorithm for image compression are found to be better than SPIHT with respect to qualitative and quantitative analyses.
Keywords: Irregular shape, Region of Interest, Direction Adaptive Filter Bank, SPIHT.