The University of California, San Diego Data Mining Competition lets you test you data mining skills on a real world dataset.

The goal is to predict if a person is a potential customer. They have two data sets: a raw dataset with has some 300 variables per person and the transformed dataset has some 2400 variables per person. Using all this data about a person, you must predict if he is interested in buying things from you.

The last date for submission of your results is 30th September.

The competition home page: http://mill.ucsd.edu/

{ 0 comments }

In the last post, we had found some lines. But the numerous lines were not good enough for detecting the location of the puzzle. So we’ll do some math today and find out exactly where the puzzle is. We’ll also un-distort the puzzle so we have a perfect top-down view of the sudoku puzzle.

Merging lines

Each physical line on the image has several “mathematical” lines associated with it. This is because of its One way to fix this is to “merge” lines that are close by.

Lines detected by the hough transform

Lines detected by the Hough transform

By merging lines I mean averaging nearby lines. So lines that are within a certain distance will “fuse” together.

We’ll write another function to fuse lines [click to continue…]

{ 0 comments }

Thumbnail image for Detecting a SuDoKu puzzle in an image (Part 1)

Detecting a SuDoKu puzzle in an image (Part 1)

August 27, 2010

In this post,we ‘ll look at detecting a SuDoKu puzzle. This include all preprocessing done on the image: filtering the image to ensure we’re not affected too much by noise. Also, segmenting the image is dealt with here. I’ve used a weird segmentation approach, so you might want to have a look at that. By [...]

Read the full article →
Thumbnail image for SeaSwarm swarm robots to help clean up oil spills

SeaSwarm swarm robots to help clean up oil spills

August 26, 2010

Oil spills are simple to clean up. Only problem is the sheer volume of oil. And an even greater volume of water. It’s the perfect task for a robot! Put several such robots and you’ve got a scalable solution! And robots in large numbers means swarm robotics! SeaSwarm is one such robot. Each SeaSwarm unit [...]

Read the full article →
Thumbnail image for SuDoKu Grabber with OpenCV

SuDoKu Grabber with OpenCV

August 25, 2010

Here’s an interesting project that I’ll be taking your over the next few days. We’ll create an app that can recognize a SuDoKu puzzle from a picture taken by a camera (maybe on a phone). We’ll be going over a lot of stuff: geometric transformations, character recognition, logic, etc. This post is an overview of [...]

Read the full article →
Thumbnail image for Obstacle avoidance with the Bug-1 algorithm

Obstacle avoidance with the Bug-1 algorithm

August 24, 2010

Obstacle avoidance is an indispensable behavior in mobile robots. If a robot has no sensors, it’s a blind robot. If it has proximity sensors but still it keeps hitting whatever comes in its way, then it’s a ‘’brainless’’ robot. Imagine an ant or any insect walking on the ground. The insect is heading towards a [...]

Read the full article →
Thumbnail image for Image Convolutions in OpenCV

Image Convolutions in OpenCV

August 20, 2010

Convolutions are a very important tool for anyone interested in signal processing. Image Convolutions is a simpler method to do convolutions on images. And they have a lot of uses too. So of course, OpenCV does have a way to do image convolutions easily and super efficiently! OpenCV’s Convolution Function The C++ convolution function The [...]

Read the full article →
Thumbnail image for Timing macros in C/C++

Timing macros in C/C++

August 19, 2010

Often you need to calculate the time it takes to execute a certain set of operations. And if you’re trying to optimize your code, you’ll need this even more. So here’s a set of useful  macros that can help you do that (without any plumbing). The Timing Macros These macros use the OpenCV functions cvGetTickCount() [...]

Read the full article →
Thumbnail image for Fast connected components labeling

Fast connected components labeling

August 18, 2010

Labeling connected components in an image is a common operation. But the original algorithm proposed is slow. It works fine if the image is small. But as the image becomes larger, the algorithm slows down really fast. Recently, a few researchers find a simple solution to this. Image that previously took hundreds of seconds would [...]

Read the full article →
Thumbnail image for The Distributed Flight Array

The Distributed Flight Array

August 18, 2010

The Distributed Flight Array (DFA) is a combination of vertical take off and landing modules and modular reconfigurable robots. A single robot is completely self contained but cannot control itself in flight. So for flight, several robots dock together. Once enough robots have assembled, the entire thing takes flight for a predetermined amount of time. [...]

Read the full article →