In this paper, we evaluate the performance of graph-based semi supervised learning (SSL) for the clas- sification of images, by using the structural similarity index measure (SSIM) to build the adjacency matrix of the graph. Performance evaluation was carried out with the TID2013 database. The results support the conclusion that SSIM can be efficiently used with graph-based SSL to retrieve images that are similar.
|Title of host publication||Unknown Host Publication|
|Publisher||Irish Pattern Recognition and Classification Society|
|Number of pages||2|
|Publication status||Published - 26 Aug 2016|
|Event||Irish Machine Vision and Image Processing conference - NUI Galway, Ireland|
Duration: 26 Aug 2016 → …
|Conference||Irish Machine Vision and Image Processing conference|
|Period||26/08/16 → …|
- Image processing
- Semi-supervised learning
- Structural Similarity Measure.
Cecotti, H., & Gardiner, B. (2016). Classification of images using semi-supervised learning and structural similarity measure. In Unknown Host Publication (pp. 67-68). Irish Pattern Recognition and Classification Society. http://uir.ulster.ac.uk/35734/2/IMVIP%202016%20Acceptance%20-%20Classification%20of%20images%20%20SSM.pdf