BRAIN DUMP

– RAHUL SANGOLE

Statistical Image Processing of Vane Wear

Objective: Investigate usage of statistical image processing to quantify turbo vane wear

Technology Used: I’m attempting this using purely open-source, free technology available on the internet:

  1. ImageMagick – For command line based auto processing of images [http://www.imagemagick.org/]
  2. R – For statistical computing of image data [http://www.r-project.org/]
  3. Arduino – Open source microcontroller for camera control, indexing table etc [http://arduino.cc/]

Test Method:

  • Nozzle vane module gets placed on some sort of an indexing table
  • The camera mount is designed such that the camera points perpendicular to the leading edge of a vane
  • The indexing will allow the table to rotate, so that we can capture pictures of all 14 vanes
  • The camera captures pictures of the leading edge of each vane, stores it on the computer
  • R scripts run ImageMagick internally to isolate the edges of interest and perform statistical calculations

Work Done So Far:

The Setup:

  • Created a platform that can be indexed to each vane; It’s leveled and mounted on an tripod head.
  • Using a 7D with a 105mm Nikkor Macro lens, set to Manual image adjustments & Manual Focus
  • Lighting – Tried a couple of techniques. The analysis and result images shown below is for incident light from a window. The code and the setup pictures show a backlight technique using a red card to filter red light only. This creates a silhouette image of the vane with the ‘non-vane’ portion in red… makes it easy to detect the edge! Extract the Red portion of the RYG information and apply the Canny edge detection algorithm.

Original image of a turbo vane, returned from the field, the Canny Edge detection algorithm, and final extracted edge:

Extracted Data:
Y Axis – Distance from base of vane to the top edge
X Axis expanded to show deviations from the nominal vane edge (which will be a straight line).

Rplot

Working code on Github : https://github.com/rsangole/vanewear

There’s a lot more to write about the workings of the code and what I’ve learnt. But, this is all I’ll write for now.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Information

This entry was posted on May 23, 2015 by in Analytics and tagged , , .

Goodreads

Blogs I Follow

Follow BRAIN DUMP on WordPress.com
The Blog of Author Tim Ferriss

Tim Ferriss's 4-Hour Workweek and Lifestyle Design Blog

Celestial events for viewing pleasure

- and what up there in the sky

a paper bird

Un pajaro de papel en el pecho / Dice que el tiempo de los besos no ha llegado

Dan Ariely

My Irrational Life

TED Blog

The TED Blog shares interesting news about TED, TED Talks video, the TED Prize and more.

DigitalJunky

A journey through information technologies

f1metrics

Mathematical and statistical insights into Formula 1

Hackaday

Fresh hacks every day

Speaking of Robots

Not Your Father's Space Sprockets

%d bloggers like this: