That only shows how pixel manipulation can be achieved in VB.Net using a lib called FastPix - and yes it mentions colours and inversion etc, that does not mean that is what your application does.
The banding suggests the guesses I made on size & position of data etc are not correct. (and as already mentioned, the colour translation from the byte value to an RGBA was a complete fudge due to not having relevant information.
So you paid for development - did your specification not require that "source code becomes the property of your company"?
Whats the name of your trust? A link to a home page maybe?
Still, not allowed to share an image file? As the 5.jpg
Also, strange purchasing agreement. You say you already payed for what you have so far but no ownership of what you payed for? Seems you should make a better agreement next time
Yeah, unfortunately the agreement was formulated in such a way that it didn't cover the scope of ownership/sharing the VB source code or the programming done for the sensors.
Did you mean these actual images ? .
And a blank image without light
Steve,
I think, I could share some (reduced/sanatised) code with you.
How do I do that? Any email or something?
In practical the camera continuously receives the images and displays it in the VB application in order to give you a live feel. To replicate the same in node red and for testing/experimenting purpose I have just posted some of the images here to get an idea about what we are going to deal with.
To get the image I usually point the torch light beam on the camera and then get the image from the VB application. Then without moving the torch or the focus on the camera I then head on to our flow from node red and then click the "Take picture" inject node from my flow. This is how I got the images and base64 data just for testing purpose. If you want more such images, I will have to do the above procedure again and then send you the images and the relevant data.
Yes I understand that. I'm merely suggesting that in the absence of actually seeing the source code, then more than one sample of data (base/64pdf) and corresponding image, may help (Steve) check if he is on the right lines with decode so far.