October 2011

You are currently browsing the monthly archive for October 2011.

The first step in data mining images is to create a distance measure for two images.  In the intro to data mining images, we called this distance measure the “black box.”  This post will cover how to create distance measures based on time series analysis.  This technique is great for comparing objects with a constant, rigid shape.  For example, it will work well on classifying images of skulls, but not on images of people.  Skulls always have the same shape, whereas a person might be walking, standing, sitting, or curled into a ball.  By the end of this post, you should understand how to compare these hominid skulls from UC Riverside1 using radial scanning and dynamic time warping.

Read the rest of this entry »

Footnotes
  1. Eamonn Keogh, Li Wei, Xiaopeng Xi, Sang-Hee Lee and Michail Vlachos  ”LB_Keogh Supports Exact Indexing of Shapes under Rotation Invariance with Arbitrary Representations and Distance Measures.” VLDB 2006. (PDF) []

Image processing is one of those things people are still much better at than computers.  Take this set of cats:

Just at a glance, you can easily tell the difference between the cartoon animals and the photographs.  You can tell that the hearts in the top left probably don’t belong, and that Odie is tackling Garfield in the top right.  The human brain does this really well on small datasets.

Read the rest of this entry »

The church needs to laugh more.  It’s good for the soul, especially when you laugh at yourself.  This post collects stand-up sketches that “bash” the church, but are still very funny.

Read the rest of this entry »

Alan Turing believed psychic mind-readers could undermine the Turing Test!  Yes, you read that right; and no, I’m not joking.  He spends about a page of “Computing Machinery and Intelligence” (CMI) discussing this problem.  CMI is the paper that first introduced the Turing Test; it was published in the journal Mind in 1950.

Read the rest of this entry »