An image retrieval with color and texture features of image sub-blocks
Versandkostenfrei!
Versandfertig in 6-10 Tagen
52,99 €
inkl. MwSt.
PAYBACK Punkte
26 °P sammeln!
Each image is partitioned into 4×6 grids of equal-sized sub-blocks. The size of the sub-block is maintained as 64x64 pixels. Further the size of the sub-block is fixed for all the images. Then the color and texture features of each sub-block are computed. A color feature descriptor Local AutoCorrelogram (LAC) which is invariant to translation and occlusion is proposed to represent the color of the sub-block. Similarly, the texture of the sub-block is extracted based on Edge Oriented Gray Tone Spatial Dependency Matrix (EOGTSDM) of an image. An image matching scheme based on Integrated Minimum...
Each image is partitioned into 4×6 grids of equal-sized sub-blocks. The size of the sub-block is maintained as 64x64 pixels. Further the size of the sub-block is fixed for all the images. Then the color and texture features of each sub-block are computed. A color feature descriptor Local AutoCorrelogram (LAC) which is invariant to translation and occlusion is proposed to represent the color of the sub-block. Similarly, the texture of the sub-block is extracted based on Edge Oriented Gray Tone Spatial Dependency Matrix (EOGTSDM) of an image. An image matching scheme based on Integrated Minimum Cost Sub-block Matching (IMCSM) principle is used to compare the query and the target image, which in turn reduces the cost of finding the integrated matching distance. The adjacency matrix of a bipartite graph is formed using the sub-blocks of query and target image, which is used for matching the images. To further improve the quality of retrieval, a Relevance Feedback approach based on a feature re-weighting scheme is used to improve the retrieval accuracy. The experimental results show that this method has improved retrieval precision and recall.
Andere Kunden interessierten sich für
