Visual Field Augmentation
1st Generation Device
This device is based on the concept of the "Wearable Computer" a device that has been pioneered by Professor Steven Mann, over the last 20 plus years or more. Prof. Mann's vision was initially as I understand it to create a device that would offer a perspective for image composition truly in the Point of View of the photographer as reality unfolds.
Among the numerous papers that Prof. Mann has published his ideas on the concept of "Mediated Reality" have also been inspirational in this project. This type of building on the work of others has been the mainstay of innovation and hopefully will serve me well.
Theory of Operation
Going by the info I've read and seen on the Net, I believe that at minimum a normally sighted person has approximately 90 Degrees of vision per eye, totaling a range of vision that varies from 120 degrees (legal limit required to hold a drivers License in the Province of Ontario) to as high as 180 degrees (Pilots, professional race drivers, and people that can see their own ears). An interesting observation I've made is that the greater your peripheral vision the less you can tolerate reading in a moving vehicle with exposed stationary surroundings due to the onset of nausea or motion sickness.
As noted below, the field of an RP Patient is notably narrower than normal values. The value of 20 degrees is a demarcation point used to define the point of "Legally Blind" and "Low Vision" individuals. Note angular Vectors may not be accurately depicted as I just wanted to draw some quick Diagrams to illustrate the point.
The meat of my theory is the mathematical linearity of optical transformations... If a video camera that has a field of image capture of 80 degrees is fed to a display that has a field of 30 degrees a net gain of 2 and 2/3rds or field compression to 37.5% of original field has occurred. When viewed by an RP Patient a net gain of 33.2 degrees totaling approx 53 degrees total field in one eye.
Proof of Concept VFA1 (Visual Field Augmentation)
With all that's been said it's time to rip into this...
If you've seen any of my other projects (be sure to check them out... especially if you have RP as there is still lots that can be accomplished, though with some adaptation in work habits) you may have noted that I enjoy working with Video equipment, the Viewfinder that is disassembled on the table is from a 1990 Panasonic Video Camera.
It uses a miniature CRT (Cathode Ray Tube) and runs on 5VDC... Don't be fooled by it's small size or low voltage, the internal circuitry to energize the tube can be as high a 6,000 Volts. Unless you have a good understanding of electronics and the potential hazards associated with high voltages defer to a qualified technician.
The camera is a small "Board Cam" meaning that all associate circuitry is contained on a single circuit board, with a CCD (Charge Coupled Device) array in the center and miniature optical lens fixed over the CCD. This unit is between 3 and 5 year old technology. It was a gift from a doc that I do occasional computer repairs and config's for... to whom I'm grateful. This unit is a Black & White cam with what I'm guessing is about 280 lines of resolution.
Originally after scoping out the video input line, grounds and Voltage feed, I wired the ViewFinder directly to the camera and tapped a 5VDC line off of the cam's board. This nearly cooked the camera as an SMD (Surface Mount Device) regulator began to get real hot & fast. Within less than 20 seconds I couldn't touch it. It was the providence that protects idiots, drunks and children that had me holding the cam when I first powered it up, as it would surely have been toast within minutes.
I opted to improvise a cheap, quick & dirty regulator off the 12 VDC line that fed the camera. From what little I do know about these types of cameras, many can use 12 to 24 V AC or DC, so a certain amount of on board regulation must be incorporated. I ran the scope on the 12 VDC line prior to the camera and saw what I'd suspected the output had a pronounced ripple, typical of a cheap wall adapter.
The circuit shown was hammered together as fast as I could pull parts from a dead Computer monitor. The capacitors may not have been necessary, but certainly don't hurt. The full size 7805 can handle a current draw of 150ma with no heat sinking. After 30 minutes continuous use it is warm but can still be touched.
Pictured above is the unit assembled... I lost a few hours trying to work-out the best way to fix the camera to the viewfinder, from nylon washers and drilling tiny holes in the casing of the viewfinder to hot gluing it into place and molding a plastic bezel to cover the glue.
Eventually I acknowledged this for what it is, just a proof of concept, not something that I'd use publicly, at least not in rural Ontario (especially when getting tractor parts or picking up scrap metal). The (above) unit to the left was actually running when the image was taken. The CRT projects an image against a surface mirror mounted at a 45 degree angle to the viewing lens that has a clever slide to compensate for viewing with uncorrected lenses.
Ultimately I just used black electrical tape to mount the camera. It is sturdy and provides better electrical isolation of the circuitry rather than being exposed. Prior to mounting the camera, I nearly fried it when the back of the board contacted the regulator and the screen blanked out but came back a split second later.
Here is one of the few picture of me on the web, I was satisfied that my field compression theory was partially sound, given the opportunity to view an number of items and begin the task of assessing the viability of the technology.
It struck me that I could capture what I saw by feeding the cam directly into the composite input of the computer...
I was astounded at what I saw on the screen... The results shown to the right are a sequence of shots I took holding the unit and swinging it around my head and noting the background behind me.
Honestly my forehead is not deformed nor is my nose that long... In actual fact the image clarity on the computer is in huge orders of magnitude better than what is seen in the miniature CRT... Both my wife and I tried the unit and found that we could in fact see or approximate vision (she is sighted with 20/20 vision, my control unit (in more ways than this..)) but had a hard time recognizing each others faces...
The 1st Gen VFA unit proved to me that this could be possible. Curiously I had started this with the assumption that the Camera would be the weak point and the CRT would move onto a Gen 2 VFA device...
As it stands now the CRT will be ditched first and the Camera will move onto Gen 2. This no great loss as I have reservations about having a hi voltage device in such close proximity to my face.
The next step to Gen 2 will require a device that can be comfortably used for several days continuously. The fish-eye or wide angle perspective of the camera lens makes it difficult to co-ordinate such fine motor control as picking up a screw in one hand and screwing it accurately into a hole with the other hand.
The literature on visual perception modification has demonstrated that the visual cortex, located behind the ears in the Occipital lobe where vision occurs has the ability to adapt to perform the 3 dimensional transformation between what it expects and what is being received. Within a matter of days fine motor control is adapted... I assume with a limited field of view less neuron-activity would be evident and a reduced acclimatization period will be required.
The issue of using central acuity for peripheral perception is a minor concern, as the sensitivity of the remaining active retinal tissue is genetically not strong on edge and movement detection, but primarily a detail and high resolution detection medium.
All accounts I've read to date support complete recovery from a distorted field experience more quickly than the acclimatization period. The most extreme of these was a researcher that wore a prism of sorts that completely inverted his view to approximate standing on ones head for several days. There were no ill effects post recovery that would lead me to believe that I'm jeopardizing either my retinal tissue or my ability to revert to natural vision.
The results of trying to view the view I'm viewing on the computer yielded some logical but potentially problematic results. The CCD array on the camera must be scanned serially to produce the individual lines of video information that are in turn traced onto the CRT (either Computer or ViewFinder). The lack of synchronization between these devices allows me to see the a dark band that rolls down the screen slowly like an old TV that needs a Vertical Osc adjustment.
TV is easily avoided, and should be regardless of your vision, as it's primarily the new-age "opium of the masses"... But Computers are my livelyhood, without gainful employment All of this grinds to a halt...
Luckily I find my self in a position that my work is both personally and financially rewarding, but this visual anomaly presents a paradox, in that this experiment is intended to extend my work-life beyond the next 2 years, but this issue precludes me from performing the function that I'm employed for...
Next Steps, research and acquire enhanced display technology. resolve Banded video issue...
If you have any insight or experience with either of the above or what is to me, this uncharted area into which this project is moving please feel free to email me...
Comments or Suggestions
Disclaimer
(an unfortunate necessity)
All
Rights Reserved theworkshop.ca © February 18, 2003.