Login Register



Auto-login on future visits

Forgot your password?

Newsletter October 2012

Learn from local texture

a Rock sorting example of local texture feature extraction

Tutorial

Building classifiers on local texture information is fundamental for many applications, such as quality control, sorting of natural products or medical diagnostics. In this short tutorial you can see how easy it is to extract texture features from local image regions and train a classifier on it.
Watch this 4 minute video to find out.

perClass course: Machine Learning for R&D Specialists.

Announcement

training course on industrial machine learning

Want to learn how to build the best classifier for your problem?
Join us on 3-7 December 2012 in Delft, The Netherlands.

This intensive training course provides you with the practical methodology to develop your own solutions with perClass. Back to your office, you solve your challenge and demonstrate a working classifier to your colleagues! Read testimonials

Learn more register now
Free Tickets for Vision 2012

Visit us at Vision trade fair

Announcement

We will present our latest developments on 6-8 November 2012 at the 25th International trade fair for Machine Vision in Stuttgart. This year we participate in the Medical Discovery Tour. We will be happy to meet you at our booth 1-F84. Come by to see for yourself how easy it is to design classifiers for industrial applications.
Would you like a FREE admission ticket?
Just request one by email.
Looking forward to meet you at Vision!

Optimize PCA minimizing the classification error

Feature highlight

Principal Component Analysis (PCA) is a useful dimensionality reduction technique. Traditionally, we need to set the output dimensionality or fraction of preserved variance as a parameter. But how do we choose this parameter?
Often, we perform PCA in order to build a classifier in the resulting subspace. Therefore, we end up tuning PCA dimensionality based on our classifier. This process is now greatly simplified by perClass sdpca command. It automatically selects the output dimensionality so that the error of our classifier is minimized. Read more in our documentation.

perClass in use: Latest research publications

Pictures from latest publications using perClass

From the blog

We are happy to share with you some of the recent publications made by researchers using perClass:

  • Detection of mental states with brain-computer interface (BCI), studied at Utrecht Medical Center in the group of prof.Nick Ramsey. Read the article in the Journal of Clinical Psychology.
  • Automatic segmentation procedure in CT scans for preoperative planning of the hip replacement, studied by Leiden Medical Center and Delft University of Technology. Read the article in the International Journal of Computer Assisted Radiology and Surgery.
  • Research and applications of perClass software, formerly called PRSD Studio. Read the article in the Hans Journal of Data Mining (in Chinese).