Invariant encoding schemes for visual recognition
Many encoding schemes, such as the Scale Invariant Feature
Transform (SIFT) and Histograms of Oriented Gradients (HOG), make use of
templates of histograms to enable a loose encoding of the spatial
position of basic features such as oriented gradients. Whilst
such schemes have been successfully applied, the use of a template may
limit the potential as it forces the histograms to conform to a rigid
spatial arrangement. In this work we look at developing novel schemes
making use of histograms, without the need for
a template, which offer good levels of performance in visual
To do this, we look at the way the basic feature type changes
across scale at individual locations. This gives rise to the notion of
column features, which capture this change across scale by concatenating
feature types at a given scale separation. As
well as applying this idea to oriented gradients, we make wide use of
Basic Image Features (BIFs) and oriented Basic Image Features (oBIFs)
which encode local symmetry information. This resulted in a range of
We then tested these schemes on problems of current interest in
three application areas. First, the recognition of characters taken from
natural images, where our system outperformed existing methods. For the
second area we selected a texture problem, involving
the discrimination of quartz grains using surface texture, where the
system achieved near perfect performance on the ﬁrst task, and a level
of performance comparable to an expert human on the second. In the third
area, writer identiﬁcation, the system achieved
a perfect score and outperformed other methods when tested using the
Arabic handwriting dataset as part of the ICDAR 2011 Competition.
Next employment after CoMPLEX:
Research Fellow at UCL, Department of Computer Science
- A. J. Newell, L. D. Grifﬁn, R. M. Morgan, and P. A. Bull.
Texture-based estimation of physical characteristics of sand grains. In
Proceedings of the International Conference on Digital Image Computing: Techniques and Applications (DICTA) Sydney, pages 504–509, 2010.
- A. J. Newell and L.D. Grifﬁn. Multiscale histogram of oriented gradient descriptors for robust character recognition. In Proceedings of the International Conference on Document Analysis and Recognition (ICDAR), 2011.
- A. D. F. Clarke, F. Halley, A. J. Newell, L. D. Grifﬁn, and M. J. Chantler. Perceptual similarity: A texture challenge. In Proceedings of the British Machine Vision Conference (BMVC), 2011.
- A. J. Newell and L.D. Grifﬁn. Natural image character recognition using oriented Basic Image Features. In Proceedings of the International Conference on Digital Image Computing: Techniques and Applications (DICTA), 2011.
- A. J. Newell, R. M. Morgan, L. D. Grifﬁn, P. A. Bull, J.R. Marshall, and G. Graham. Automated texture recognition of quartz sand grains for forensic applications. Journal of Forensic Sciences 2012