Title :
A hybrid method for image partitioning in low bit rate segmented image coding
Author :
Christopoulos, V.A. ; Pratikakis, I.E. ; Cornelis, J.
Author_Institution :
Vrije Univ., Brussels, Belgium
Abstract :
A new technique for extracting, perceptually ranking and selecting meaningful image details is proposed. The selected details are combined with the segments describing the main objects, produced after roughly segmenting the image. The hybrid scheme, which combines coarse segmentation and refinement by selected image details, leads to better visual image quality than using direct segmentation, especially in the case of low bit rate coding
Keywords :
feature extraction; image coding; image segmentation; coarse segmentation; feature extraction; hybrid method; image detail selection; image partitioning; image ranking; image refinement; low bit rate segmented image coding; segmentation algorithm; visual image quality; Bit rate; Chromium; Data mining; Image coding; Image quality; Image reconstruction; Image segmentation; Merging; Silicon carbide; Transform coding;
Conference_Titel :
Digital Signal Processing Proceedings, 1997. DSP 97., 1997 13th International Conference on
Conference_Location :
Santorini
Print_ISBN :
0-7803-4137-6
DOI :
10.1109/ICDSP.1997.628486
