|
|
|
| Patterned Fabric Image Retrieval Using Color and Space Features |
|
|
|
|
Abstract Image retrieval has been an active research topic in the last
decade. As one of the promising approaches, color histogram based
image retrieval has been attracted by many researchers. However, it
is sensitive to noisy interference and lost any spatial information.
To overcome drawbacks, the algorithm combined weighted color
histogram with image segmentation is proposed for patterned fabric
images in this paper. Firstly, the patterned fabric images are
preprocessed using image enhancement. Secondly, color and spatial
information features are respectively extracted by weighted color
histogram and image segmentation method, which are transformed into
feature vectors. Finally, the similarity between objective image and
image database is calculated with Euclidean distance. The retrieval
results are displayed in decreasing order of similarity.
Experimental results, including comparisons with color histogram
algorithm and weighted color histogram algorithm, demonstrate the
effectiveness of proposal for patterned fabric images.
|
|
|
|
|
|
| Cite this article: |
|
Junfeng Jing,Qi Li,Pengfei Li, et al. Patterned Fabric Image Retrieval Using Color and Space Features[J]. Journal of Fiber Bioengineering and Informatics, 2015, 8(3): 603-614.
|
|
|
|
|
2015, Vol. 8 
