Abstract
We analyze some of the visual properties of the HSV (Hue, Saturation and Value) color space and develop an image segmentation technique using the results of our analysis. In our method, features are extracted either by choosing the hue or the intensity as the dominant property based on the saturation value of a pixel. We perform content-based image retrieval by object-level matching of segmented images. A freely usable web-enabled application has been developed for demonstrating our work and for performing user queries. © Springer-Verlag 2004.
Cite
CITATION STYLE
Vadivel, A., Mohan, M., Sural, S., & Majumdar, A. K. (2004). Segmentation using saturation thresholding and its application in content-based retrieval of images. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3211, 33–40. https://doi.org/10.1007/978-3-540-30125-7_5
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
