1. The first API I tried is called the Monitor Configuration of the Win32 API but did not work for my laptop (some kind of I2C transmission error occured)
2. The Backlight Control Interface using IOCTL was the first approach that worked for me, but getting the current brightness is not supported on newer Windows versions
3. So I ended up using WMI which is available for .NET and works surprisingly quite well.

Further usage information and a downloadable binary can be found on my Projects page
If you encounter bugs or have any suggestions please do not hesitate and give some feedback.

I am trying to build a multi-touch screen with the DSI technique, explained here but instead of using a projector and some special projection surface I will be using the recently disassembled LCD.

So why not using a projector? The answer is quite simple: I don't own one But there are some other facts that lead me to try it with an LCD:

• Projectors need quite a lot of distance from the projection surface but I am not intending to build a multi-touch table.
• Even HD projectors which are quite expensive do have very limited resolution in relation to the size of the projection surface. So small text can hardly be read on such a table.
• I wanted to build something smaller so one person is able to reach the whole screen from one location.

There are basically two issues with the LCD approach:

1. The camera is located behind the LC-panel thus has to track blobs through the panel which normally is covered by the backlight (with a white opaque layer) and some metal. This directly leads to the second problem.
2. The backlight has to be replaced with one that is itself transparent.

My first approach was to use Plexiglass Endlighten for both backlight and IR illumination but I was not able to detect any blobs with the Plexiglass behind the LC-panel. So I bought another Plexiglass Endlighten which is located in front of the LC-panel for blob tracking purposes only.

I took some picture to test whether IR light is blocked on dark regions on the LCD image as visible light is. The results were quite surprising, the visibility of IR light is not affected by the picture shown by the LC-panel at all as can be seen below.

Display with backlight on.

Backlight and IR lighting on

Only IR lighting on.

Both backlight and IR lighting on. Bandpass filter on top of the camera.

My next step will be to build a prototype and experiment with different types of LEDs for a brighter backlight.

After the successful testing in my last post, I wanted to try the whole thing on with an LCD. The problem with that was how to get one. Luckily my brother has an old Sony SDM-S71R with 17" which is not needed anymore. It's not really big, but for a first test it will do.

Disassembling the whole thing was quite easy, only the first part, getting the plastic casing off, was a bit tricky. If you are unsure how to do it, search online for 'Sony SDM-S71R Service Manual'. After removing some plastic and metal casing two circuit boards will be revealed. Both of them shown below:

Power supply of the Sony LCD

This one is the power supply for both the electronics controlling the LCD and the four fluorescent tubes for the backlight. Since I do not want to use the original backlight I do not need the high voltage part. That made me look for some alternative to the power supply. The pinout of the connection to the lcd controller is described directly on the board and is shown below:

Wiring of the connector from the LCD power supply to the controller

The purple, grey, white, and black cables on the right need not be connected. So there are only the +12V, +5V, and ground wires left. These voltages are also used in computers, so the choice was easy: I used the power supply of an old PC.

Next is the controller board:

Nothing exciting here aside from the many wires on the right that go to the display itself.

The display consists of the LC layer on top of a glass plate and the backlight with some plastic films between them. Again some pictures of the parts ordered top-down:

LC Layer with controller

Diffuser with some other plastic films on top of a plastic panel and a white paper

Fluorescence tubes that shine from the side into the plastic panel

These are the most important parts of the LCD. If you have any questions about the disassembly, just ask.

So much for the LCD disassembly. I've already done some testing with it in combination with a self-made backlight, some IR LEDs, and a bandpass filter for the camera. But this will be the topic of the next post,...

So how did I run the test? A few days ago I ordered a sample of some Plexiglass EndLighten, which is available here. It's a special glass that shines diffuse when light shines from the side into the sheet. The final setup was simple: I connected 8 LEDs in row with a 1 ohm resistor and powered the setting with 12 Volts, the LEDs were attached to the Plexiglass and and everything was filmed with the cam.

A Screenshot with blobs detected on my hands is below, the other blobs in the surface come from a screw clamp.

stay tune for some more news in the near future

The paper can be downloaded here.

More on that can be found on wikipedia.

I have finally managed to correct this issue. I had to find a replacement for the filter. It had to be of glass or a similar material with the same width as the filter. Quite tough to find a piece of glass or plastic with a depth of about a millimeter. Suprisingly the cap of a CD case did the job as a filter replacement. I first sawed a piece off and afterwards filed it into shape.

The room was quite dark but the results are  better than without the replacement, I think:

PS3 Eye Capture with IR filter replacement

Now the second part of this posting, blob detection. Blob detection is about finding bright or dark spots in an image. I used some kind of global flood fill algorithm to find the blobs. Sadly I was not able to programm this part in the pixel shader so I implemented an algorithm on the CPU. On my laptop CPU power is fairly restricted so it took me a few attempts to do blob detection at 60 fps. The algorithm is a scanline algorithm which searches for horizontal lines of bright pixels. After the complete image has been scanned, one or more connected lines become a blob. The blob center is the barycenter of the pixels, at the moment without regard of their brightness. I have to test this algorithm for robustness, but can already give you a first impression:

A first test of blob detection. The red dots are centers of detected blobs.

Hope I could help anyone with the camera focus problem. Maybe I release the source code of the project when it is out of experimental stage, or to say it straight: When it's a bit less messy.

As you can see, the picture is quite blurry which has a negative impact on corner detection. That is because of the missing IR filter, now I am unable to set the focus right, maybe I can put a piece of glass into the lens to correct the error.

Have fun testing it and don't forget to give some feedback

I know this site surely needs some customization and content so stay tuned for some news about projects that I am working on, my studies and other things I'm spending time on.