Code covered by the BSD License  

Highlights from
Trainable COSFIRE filters for keypoint detection and pattern recognition

Be the first to rate this file! 33 Downloads (last 30 days) File Size: 2.74 MB File ID: #37395
image thumbnail

Trainable COSFIRE filters for keypoint detection and pattern recognition



05 Jul 2012 (Updated )

A COSFIRE filter detects features or patterns of interest, e.g. traffic signs in complex scenes.

| Watch this File

File Information

A COSFIRE filter is automatically configured to be selective for a local contour pattern specified by a single example. The configuration comprises selecting given channels of a bank of Gabor filters and determining certain blur and shift parameters. Gabor filters are, however, not intrinsic to the method and any other orientation-selective filters can be used. A COSFIRE filter response is computed as the weighted geometric mean of the blurred and shifted responses of the selected Gabor filters. It shares similar properties with some shape-selective neurons in visual cortex, which provided inspiration for this work.

In our publication, which is given below, we demonstrated the effectiveness of the proposed filters in three applications: the detection of retinal vascular bifurcations (DRIVE data set: 98.50% recall, 96.09% precision), the recognition of handwritten digits (MNIST data set: 99.48% correct classification),
and the detection and recognition of traffic signs in complex scenes (100% recall and precision).

COSFIRE filters are conceptually simple and easy to implement. They are versatile keypoint detectors and are highly effective in practical computer vision applications.

You are kindly invited to use this Matlab implementation and cite the following article:

George Azzopardi and Nicolai Petkov, "Trainable COSFIRE filters for keypoint detection and pattern recognition", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35(2), pp. 490-503, 2013.

Paper [pdf]:


This file inspired Trainable Cosfire Filters For Vessel Delineation With Application To Retinal Images.

Required Products Image Processing Toolbox
Signal Processing Toolbox
MATLAB release MATLAB 7.7 (R2008b)
Tags for This File   Please login to tag files.
Please login to add a comment or rating.
11 Mar 2013

Added file dilate.mexw64 which is needed for 64-bit machines.

17 Jun 2013

1. Added the compilation of a required c-function.
2. Added examples to detect traffic signs in complex scenes.

17 Jun 2013

Removed a pre-compiled version of a c-function.

17 Jun 2013

Added a hyperlink in the Description to download the paper.

18 Jun 2013

Corrected the url to download the paper

24 Jun 2013

Fixed a spelling mistake in the description

03 Jan 2014

Changed link to pdf file

Contact us