Color Segmentation Based Resolution Enhancement of Depth Image and Stereoscopic Image Synthesis
Sreejith K.M.1, K. Suresh2, K. Rajeev3
1Sreejith K.M., HCL Technologies, Bangalore (Karnataka), India.
2Dr. K. Suresh, Department of Electronics and Communication Engineering, Government Engineering College, Barton Hill, Thiruvananthapuram (Kerala), India.
3K. Rajeev, Quest Global Engineering Services Private Limited, Thiruvananthapuram (Kerala), India.
Manuscript received on 15 December 2015 | Revised Manuscript received on 25 December 2015 | Manuscript Published on 30 December 2015 | PP: 7-14 | Volume-5 Issue-2, December 2015 | Retrieval Number: B4322125215/15©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In this paper a color segmentation based resolution enhancement of depth image is proposed. The resolution enhancement technique is combined with depth image based rendering (DIBR) method to generate stereo images. The major problem in transmission of stereo images is bandwidth. By using DIBR the transmission efficiency can be increased. Also if we are using low resolution depth map for transmission, the bandwidth for transmission can be further reduced. A color segmentation based interpolation procedure is used to enhance the resolution of the depth map. This depth resolution enhancement method sharpens depth image using the color information from the high resolution color image. From this resolution enhanced depth image the stereoscopic images are synthesized using DIBR method. For more textured region, we are using an edge-guided image interpolation algorithm to enhance the performance of the color segmentation based interpolation algorithm.
Keywords: DIBR, Depth Image, Enhanced Depth Map, Color Segmentation
Scope of the Article: Image Processing