Evaluation of shape classification techniques based on the signature of the blob

Pedro Gil-Jiménez E-mail The Corresponding Author, Hilario Gómez-Moreno, Javier Acevedo-Rodríguez, Roberto J. López-Sastre, Saturnino Maldonado-Bascón

Dpto. de Teoría de la señal y Comunicaciones. Universidad de Alcalá, 28805 Alcalá de Henares (Madrid), Spain

Signal Processing
Volume 92, Issue 1, January 2012, Pages 63-75

doi:10.1016/j.sigpro.2011.06.007 | 

In this site, you can download all the images, as well as the results described in the paper. We also included a demo video, and some examples of the algorithms applied to the traffic sign recognition problem.


Video example

In this paper, we report a study of different preprocessing and classification techniques that can be applied to shape classification using the signature of the blob, or its FFT, as the main feature. Eight well-known classification methods were tested and compared.
The results obtained show that, for shapes with a small to medium amount of distortion, all the methods obtained an almost 100% success probability. However, as distortion increased, those not based on the FFT performed better than the other algorithms, at the expense of a small increase in computational time.
The samples employed for training and testing purposes were not hand-selected, but were generated by an application developed as part of this study. This application simulates the main distortions that can be produced by a real camera, including shifts, scalings, rotations, affine transformations and noise. We demonstrate that the use of these synthetic images for the training process, instead of manually selected ones, had proven to perform well with real images.
A study of the false positive problem is also included, showing that, with the use of SVMs and careful selection of the training set, a large number of false positives can be discarded in the detection step.



(1) Test
21 Directories

(2) Results
8 Directories



Res Comparison





(3) Traffic Signs
15 Directories



Put your comments here
Jul 29, 2018:
Jul 22, 2018:
Jul 03, 2018:
Jun 03, 2018:
May 27, 2018:
May 05, 2018:
Apr 14, 2018:
Mar 23, 2018:
Mar 10, 2018:
Feb 17, 2018:
Feb 03, 2018:
Jan 19, 2018:
Jan 04, 2018:
Dec 20, 2017:
Dec 20, 2017:
Dec 05, 2017:
Nov 19, 2017:
Oct 31, 2017:
Oct 20, 2017:
Oct 03, 2017:
Sep 18, 2017:
Sep 03, 2017:
Aug 20, 2017:
Aug 05, 2017:
Jul 22, 2017:
Jul 05, 2017:
Jul 05, 2017:
Dec 12, 2015:
...Your comments here...stop
Apr 27, 2012:
hello how are you i was ask you in classificatin matlab.please help me.i read whole of paper but i dont understand.
what do i enter in x and y and nbclass in program of [xsup,w,b,nbsv,pos,obj]=svmmulticlassoneagainstall(x,y,nbclass,c,epsilon,kernel,kerneloption,verbose,warmstart);
Feb 07, 2012:
Good results and good web

  • Models. Images corresponding to the reference shapes used to classify in MSE, CC and EMD algorithms.
  • Training set. Images corresponding to the shapes used to train SVM and KNN classifiers.
  • Test set. Images used to obtain the results.