In many image processing tasks it is useful to have a saliency map that highlights the important parts of an image. Many algorithms currently exist that generate saliency maps. In this work a new algorithm is developed that is capable of generating multiple saliency maps from a single image, highlighting different parts of the image. This is done using the Complementary Color Wavelet Transform, a tool which captures various color changes at different scales in an image. The outputs of this transform are then fed into an adapted version of a previously developed saliency map generation algorithm. Using this transform, the algorithm is able to generate four saliency maps for a given image. This algorithm is applied on the images in the CAT2000 dataset. The saliency maps generated using this algorithm are evaluated using two popular performance metrics, the similarity metric and linear correlation coefficient. In many cases, the proposed algorithm provides better results than other approaches.
|Advisor:||Fontaine, Fred L.|
|School:||The Cooper Union for the Advancement of Science and Art|
|School Location:||United States -- New York|
|Source:||MAI 82/2(E), Masters Abstracts International|
|Keywords:||Image Processing, Saliency, Wavelets|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be