Monday, 25 April 2016 16:06
Ranklets: Orientation selective non-parametric features for image analysis
Ranklets: Orientation selective non-parametric features for image analysis
Dr. Fabrizio Smeraldi (Queen Mary, University of London)
Department Of Engineering, Aula Magna, 04 May 2016, 4pm
Dr. Fabrizio Smeraldi (Queen Mary, University of London)
Department Of Engineering, Aula Magna, 04 May 2016, 4pm
Published in
Research
Tuesday, 07 April 2015 11:34
CTMR15
The Workshop on Color and Texture in Material Recognition will take place in Genoa, Italy, in conjunction with the 18th International Conference on Image Analysis and Processing (ICIAP), Sept. 7-11, 2015. The workshop will cover different areas of image processing including colour and texture analysis, material perception, remote sensing, medical imaging and industrial inspection. Deadline for paper submission: June 3, 2015.
Published in
Research
Tuesday, 02 September 2014 09:55
Improving co-occurrence features through digital circles and DFT
Grey-level co-occurrence matrices (GLCM) have been on the scene for almost forty years and continue to be widely used today. We developed a method to improve accuracy and robustness against rotation of GLCM features for image classification. Co-occurrences are computed through digital circles as an alternative to the standard four directions. We use discrete Fourier transform normalization to convert rotation dependent features into rotation invariant ones. We tested or method on four different datasets of natural and synthetic images. Experimental results show that our approach is more accurate and robust against rotation than the standard GLCM features.
Source:
F Bianconi and A. Fernández, Rotation invariant co-occurrence features based on digital circles and discrete Fourier transform, Pattern Recognition Letters, 48:34-41,2014. Special issue 'Celebrating the life and work of Maria Petrou'.
Code and data available here.
Code and data available here.
Published in
Research
