may be it would be useful for better understanding of the project/concept. Ok, here it is:
The project is divided into phases/parts
The Training software and Live software are two major parts
Using the Training software, blind person's brain learns to percieve images from sound.
Every pixel of the image is converted in particular sound, one pixel at a time.
The way it is done is exactly like the way we read text (written pages).
one word at a time.. from left to right... and one line at a time... from top to bottom
Similarly we start from leftmost pixel on the top row and progress towards rightmost pixel on the topmost row, then the line below it.. and so on untill we reach the right most pixel in the
bottomost row.
As we get used to creating mental images using the sounds from auxel software, we go on making images from tiny to small... and medium to big and bigger ...
also add more colors... add more levels of luminance and more shades of colors...
(by the way our computer's icon size image is their big size image)
now when they reach bigger level, then they can go LIVE
Now, LIVE means... image is changing every few seconds...
( and eventually even many time every second)
These images comes from the camera directly (instead of the stored images)
--------------
As of for now, the prototype we are ready with: is an applet.. which runs on PC
We are working to make a stand alone executable file for it too.
The Mobile (phone) Phase has training software as well as LIVE software in it. we have to yet begin working on it. (Waiting for the right people to join in the team!)
The Device Phase will have only live version. This will be independent circuitry and will not need PC or mobile phone. Eventually this circutry will be made inside a microprocessor chip so that it can be hidden inside a special goggle.
Which would mean that, after learning to create mental images from sound, just by wearing this special goggle, blind people would be able to do almost everything that sighted person does in day to day life!
------------
by the way..
we are effectively just by passing the normal channel for image to reach the brain
(via eyes --> optic nerves --> part of the brain that naturally processes images)
Auxel sends image to brain via digital camera -- > microprocessor --> sound --> ears --> auditory nerves --> brain (which now learns to process the image)
****
I am including the comments in earlier post ( cuz most people do not bother to see the comments)
Anjana:I am slightly confused. Ok i got it that the person can easily draw an image from the soud that the camera would produce.
But how will it help him in knowing what's that object.
What should be planned for the long run is...that artificial intelligence be built into the camera so that it can directly tell people as what the object is.
But an excellent effort and best wishes!
Omkar Nisal: Thank you for your comment and question Anjana . We love questions!
To help you understand how this will help them identify objects.. let me first give you an analogy...
A new born, sighted baby's brain learns to extract meaning from the signals [image data] coming from eyes over the period of months and years..
and ONLY then it know that so and so is an object
similarly..
these visually challenged people's brains are now learning to extract meaning from the signals [image data] coming from digital camera (via ears). It might take many months of training for them to interpret different objects in an image. you see.. these people have just got a vision for the first time in life!
would you expect a week old baby to start identifying objects!????
ofcourse not!! learning takes its own time!
moreover, sighted babies are born with full range of eyesight,
where as these people have to gradually go on increasing the
1> size of picture,
2> number of colors,
3> levels of luminance,
4> and then saturation/chroma of the image.
5> then the complexity of the image.
6> and the speed of playing one picture
Its a long journey. We just have taken first few steps!
Fortunate thing is that, unlike the sighted babies,people we are dealing with can at least understand what we are trying to do and what we are trying to communicate.
Baby has to learn who is who, sounds and language, images and objects, and even human behaviour! Thats why we could learn this language faster than a baby would learn to speak from the day of birth.
I love to give baby's example because every one know baby takes time to grow and learn and thus better relate to this learning similar to baby's learning.
---
About artifical intelligence:
we already had carefully considered and thoughtfully decided not to go with AI built in this device.
1> we might need to embedd super computers in the goggles.. well if not supercomputers.. at least mainframes. so practicality of this happening on portable device.. at least as of for today is quite remote. Things change fast.. may be soon we could do that.
2> We are closely studying behaviourial patterns and likes/dislikes of these challenged people. we have found out although you and me(sighted/not disabled people) love automation and things being done for us.. these people hate to depend on anyone or anything. Independence is very important deciding factor for their self esteem and life. They do not want to be 'told' this is the color.. they want to 'feel' for themselves the colors..
Same with the objects.
I always try to put myself in their shoes, I always ask myself:"Would I love this if I was blind?".
You can ask yourself too!
FOr a moment lets leave practicality aside..
lets assume we manage to get AI to identify all..
ALL of the trillion objects in the world...
Would you really like AI tell you.. "ITs a beautiful orange colored sunset"
or "This a #FFGG0H shade or cyan or 4th level dark red colored beautiful dress"
"This is white color Taj Mahal" or "This is Aishwarya Rai/Tom Cruise"
GOSH! If I was blind.. I would rather invest 1 year of my life to master this language and then get back to my normal life as I am doing now!
Would you use AI trying to interpret images??
or rather use the portable, super computer that every one carries every where over ones shoulders??
The choice is of-course individual's!
What would you choose!?
(I suggest sealing your eyes for couple of days if not couple of weeks, before deciding!)
:-)