Basic Image Features

Basic Image Features (BIFs) are a significant contribution by this group. Computation of BIFs consists in classifying each pixel of an image into one of seven categories depending on local structures and symmetries, creating a primal sketch of the input image. BIFs are computed efficiently based on the responses to a bank of Derivative-of-Gaussian (DtG) filters.

The computation of BIFs is controlled by a scale parameter (σ, the standard deviation of the DtG filters) and a threshold parameter (ε) dictating the fraction of an image that should be considered as 'flat' (i.e. without a particular structure).

BIFs can be used to encode compact representations of images (e.g. 7-bin histograms). The seven categories can be extended to 23 by quantization of the non-rotationally symetric features. Those 23 features are termed oriented Basic Image Features, or oBIFs.


Please visit the GitHub repository for implementations and examples of BIFs, oBIFs (oriented BIFs) and oBIF histograms (binned histograms of oBIFs).


BIFs and oBIFs have been employed for a wide variety of applications. Here you will find links to relevant papers ordered by the type of applications.


Griffin L. D., and Lillholm M. (2010). Symmetry sensitivities of derivative-of-Gaussian filters. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(6), 1072-1083
Griffin L. D., Lillholm M., Crosier M., and van Sande J. (2009). Basic Image Features (BIFs) arising from local symmetry type. 2009 Scale Space and Variational Methods in Computer Vision LNCS vol. 5567:343-355
Griffin L. D. (2007). The 2nd-order local-image-structure solid. IEEE Transaction on Pattern Analysis and Machine Intelligence 29(8), 1355-1366

Texture classification

Newell A. J., Morgan R. M., Griffin L. D., Bull P. A., Marshall J. R., and Graham G. (2012). Automated texture recognition of quartz sand grains for forensic applications. Journal of Forensic Sciences, 57(5), 1285-9
Crosier M., and Griffin L. D. (2010). Using Basic Image Features for Texture Classification. International Journal of Computer Vision, 88(3), 447-460

Appearance learning and object detection

Jaccard N., Rogers T. W., and Griffin L. D. (2014). Automated detection of cars in transmission X-ray images of freight containers. 11th IEEE International Conference on Advanced Video and Signal-based Surveillance, 2nd Workshop on Vehicle Retrieval in Surveillance
Griffin L. D., Wahab M. H., and Newell A. J. (2013). Distributional learning of appearance. PloS One, 8(2), e58074

Character recognition and handwritting analysis

Newell A. J., and Griffin L. D. (2014). Writer identification using oriented Basic Image Features and the Delta encoding. Pattern Recognition, 47(6), 2255 2265
Newell A. J., and Griffin L. D. (2011). Natural Image Character Recognition Using Oriented Basic Image Features. 2011 International Conference on Digital Image Computing: Techniques and Applications, 191-196

Microscopy image segmentation and analysis

Jaccard N., Szita N., and Griffin L. D. (2014). Trainable segmentation of phase contrast microscopy images based on local Basic Image Features histograms. In Medical Image Understanding and Analysis (pp. 47-52). British Machine Vision Association.
Jaccard N., Macown R. J., Super A., Griffin L. D., Veraitch F. S., and Szita N. (2014). Automated and Online Characterization of Adherent Cell Culture Growth in a Microfabricated Bioreactor. Journal of Laboratory Automation (In press)
Jaccard N., Griffin L. D., Keser A., Macown R. J., Super A., Veraitch F. S., and Szita N. (2014). Automated method for the rapid and precise estimation of adherent cell culture characteristics from phase contrast microscopy images. Biotechnology and Bioengineering, 111(3), 504-17
Reichen M., Macown R. J., Jaccard N., Super A., Ruban L., Griffin L. D., Veraitch F. S., and Szita N. (2012). Microfabricated Modular Scale-Down Device for Regenerative Medicine Process Development. PLoS ONE, 7(12), e52246