SeekOFix - Better software for SeekThermal camera

Windows software that will produce nice clean thermal images in contrast to original android software.

Frenky

Following Follow project

Liked Like project

Become a Hackaday.io member

Just one more thing

To make the experience fit your profile, pick a username and tell us what interests you.

OK, I'm done! Skip

Similar projects worth following

View Gallery

Description

SeekThermal is really nice thermal module with resolution 208x156, but it produces really low quality images. The huge amount of noise makes the module barely usable.

The aim of this project is to make a windows software that will produce clean images with no noise.

All pixels will be profiled for offset and thermal drift and bad pixels replaced with neighbour pixel average.

This should be sufficient to produce clean images.

Details

Steps that will be taken:

1. Take raw reading of all pixels in 5sec intervals for 1 hour while pointing to uniform temp plate at room temperature. This will show how big error each pixel has and thermal drift of each pixel though time.

2. Create a diff matrix 208x156 with values showing difference between each pixels raw value and the correct value.

3. Repeat steps 1 and 2 while pointing to heated bed with temp set to 50*C and 100*C.

3. Find all pixels that do not response to temperature change and generate a badpixel map for further usage.

4. Implement a code that compensates each pixel drift with diff matrix.

5. Replace all badpixels with special averaging of neighbour pixels values.

6. Create realtime software

Components

1 × SeekThermal (the first model) Thermal module that connects to smart phone
1 × Aluminium block As a uniform temperature plate
1 × Thermal pad Covering aluminium plate
1 × DIY OTG female micro USB to male USB cable So you can connect SeekThermal to PC

Project Logs

Collapse logs

Progress reports
Frenky • 05/09/2016 at 13:42 • 0 comments

I've made a new app for Seek:
http://www.eevblog.com/forum/thermal-imaging/seekofix-new-windows-software-for-seekthermal/
Hardware changes:
http://www.eevblog.com/forum/thermal-imaging/seek-thermal-hardware-modification-to-improve-image-quality/
Curves of pixels drifting
Frenky • 10/29/2015 at 21:33 • 0 comments

Looking at some random thermal pixels as they change values through time (from start to 15min).
All of the pixels are pointed to the same 19*C plane.
Temperature drift is really big and each pixels has curve of it own.
There is also some weird jumping of values:
Profiling noise and pixel values
Frenky • 10/29/2015 at 18:37 • 0 comments

I wrote some come to see how much raw values fluctuate and which are values when pointing to thermal plane at 19*C.
Most pixel values is between 6000 and 9000 with middle of the hill at 7500. So 7500=19*C.
Video: