No. If my reduced apaeratures , close coupled heater in cone and pyro thermometer does not resolve 6mm wide lettering I may have another go. Perhaps indirect bounce from a large distributed heater, using some sort of diffusing surface to bounce off.
Now that is an interesting idea, for this application , may require 2 mirrors, but feasible.
I got back to this project yesterday. I' d tried oblique lighting with NIR and modded webcam before and not seen anything. This time I could make out lettering under pattern-free areas but not convincingly under patterning. But I remembered I'd painted letters NCE V I could quite plainly see the N and V and the curve of the C but also curve in the E area. I dug out the paper I'd written down the letters and it was NCQ V So without patterning, in the real application, it may be possible, "brass-rubbing" optically. I'll also have a go with mark3 close-coupled heater cone and cut down pyro thermometer,tomorrow , on a grid of readings and repeat for different temperatures and hopefully improve on the 0.7 deg F differential. I tried the modded webcam , with no floppy disc filter,viewing a Weller magnastat soldering iron barrel. The heater area glowed bright and dulled to nearly no discernable pixels , and the thick part of the barrel stayed at some intermediary luminosity, brightening and dimming but observable in space by its conical shape at all times , with a 700 deg F magnastat in place.
On Sat, 16 May 2015 20:04:48 +0100, N_Cook wrote as underneath :
Thanks for continuing your experimentation story! C+
I've now got some convincing AVI of the N followed by a circular letter not complete as an O , could be C or G . With floppy disc material over the webcam lens , so I can work in room light and laptop screen is black with no IR torch on. With all controls on the webcam driver set to min brightness and max for contrast, saturation and mid white balance. As only 5 frames per second, sweeping the NIR LED torch gives a better chance of reading the letters then snapshots. In the real application the context of words should make distinguishing the likes of Q and C, or E and F easier. In the next hour or two I'll give the heater and pyro method another go over 1/4 inch steps horizontal and vertical at a few heater settings, but stick to 3 seconds per position, for optimum of speed of capture and one level of consistency. Could increase to 10 seconds or so if required for greater resolution later, over questionable letters.
Got nowhere on the heater cone and pyro. I even tried a rubber hot water bottle with hot water ,held against the test wall for 10 minutes. Loads of temp variation but nothing to do with black paint, just change of heat capacity of different thickness of plaster over brick presumably. So I'm wondering if even an expensive thermal imaging camera and some sort of diffuse and even heat source would work either, unless just flashed over the paper and not hanging around enough to deep heat. The NIR webcam is relying on "brass-rubbing" relief in all probability. The \ part of the N is evident in raking white light but not the rest of the N or any of the next letter or any of the other 2 letters so bit of a mystery why the NIR+webcam is picking up so much differentiation. With no patterning on the paper it would probably resolve the likes of C and G, Q and O, E and F ,P and R, B and 8 etc. I tried polarizing filters as well I've run out of ideas, what I would like to try was a long IR, 5 to 10 micron LED to direct heat energy onto a small spot, the pyro seems to be reacting to heat radiating off the close by metal heater cone rather than heat bouncing off the paper. All the parabolic reflectors I have are too big to direct a little halogen bulb or similar "point" heat source.
Put a small halogen bulb, 5-20 watts and 12 VDC like those used for under cabinet kitchen lighting, in one end of a short piece of copper tubing. The inside of new tubing is pretty shiny in the visible and should be a great reflector at 10 um, and something like 1/4 or 3/8" tubing a few inches long should give a pretty good spot that size when held against or almost touching the wall paper. Probably comparable to a small parabolic reflector but with a closer working distance. The longer the tube the more directed the beam, and the more light lost to the walls. I would start with 1.5 or
2" of tubing but you might need to try even shorter. Good part is it's cheap :-).Have you called a few of those companies that use ir cameras to map heat leakage from a house, to see if they would be willing to try a quick experiment for free in return for whatever publicity you can give when you examine the final target?
----- Regards, Carl Ijames carl.ijames aat deletethis verizon dott net
Got nowhere on the heater cone and pyro. I even tried a rubber hot water bottle with hot water ,held against the test wall for 10 minutes. Loads of temp variation but nothing to do with black paint, just change of heat capacity of different thickness of plaster over brick presumably. So I'm wondering if even an expensive thermal imaging camera and some sort of diffuse and even heat source would work either, unless just flashed over the paper and not hanging around enough to deep heat. The NIR webcam is relying on "brass-rubbing" relief in all probability. The \ part of the N is evident in raking white light but not the rest of the N or any of the next letter or any of the other 2 letters so bit of a mystery why the NIR+webcam is picking up so much differentiation. With no patterning on the paper it would probably resolve the likes of C and G, Q and O, E and F ,P and R, B and 8 etc. I tried polarizing filters as well I've run out of ideas, what I would like to try was a long IR, 5 to 10 micron LED to direct heat energy onto a small spot, the pyro seems to be reacting to heat radiating off the close by metal heater cone rather than heat bouncing off the paper. All the parabolic reflectors I have are too big to direct a little halogen bulb or similar "point" heat source.
Great minds ... I'd just removed a 6V 10W halogen bulb from its reflector, I was going to roll up some polished tin-plate, soldered on the outside.
I'll have to abandon the close coupled heater and pyro thermometer. Even overdriving the 6V bulb to 8V and 8 deg C nominal over ambient on the pyro thermometer , nothing convincing found. That 8 deg C is the reading on the scale with my cut down brass cone, probably corresponds to 15 deg C or more for the proper cone that it is properly calibrated to. Whether a FLIR device and heat source would work, is a bit questionable. But the IR filter removed web cam and 5 degree or so raking angle NIR torch, video image recorded to laptop, is worth trying.
I came across something that sounds pretty intriguing and might be within your budget, a long wave IR camera chip that responds to 7-14 um IR and delivers 80x60 pixel images at about 2 frames per second. Here is where I found it:
----- Regards, Carl Ijames carl.ijames aat deletethis verizon dott net
The off the shelf FLIR 60x80 camera is not much more in cost. Next week I'll take my modded webcam + NIR torch to the site and have a go on site. I'll also take high quality pics of the wallpaper along with architects colour swatches and ruler. If my NIR webcam does not show anything then the next pricewise plan of campaign is look into custom-made wallpaper (usually for kids bedrooms) and permission to remove and replace a patch of wallpaper. Luckily the relevant patch of wall is isolated from the rest of the room , so an absolutely colour and texture matched patch should not be noticeable.
I'm hoping, if not fully readable image, to have enough image to show something is under there, intriguing enough to get permission to remove a section of wallpaper. As it should be very localised text then knowing where it is then perhaps simply remove the patch of wallpaper and move one of the pictures to cover the spot, may even be possible. Because of the significance of the text they may wish to have it accessible, but not necessarily on show, for the odd academic visitor.
Since you first brought this subject up I have been intrigued with modding a camera in order to see NIR. I did this mod to a Canon SD
300 Powershot camera. It not only sees NIR but also near UV. I have been using the camera to look through stuff that is opaque to my eyes, in both NIR and NUV. Is it possible that NUV might go through the wallpaper? Eric
A assume it was near UV, 20 LED cluster only labelled UV , numerous materials flourescing under it. I returned to the test wallpaper patch and tried straight on and oblique with various visibly opaque or going on opaque "filters" over the webcam lens , passing the UV image, mylar, polyester , black polythene, CD in overprinted-free area of label, floppy disc material blocked the uv illuminated image. Could see not the slightest hint of any letters under the paper wiht any of them, or absent filter, but could see the wallpaper printed design in each case, but nothing in the background room light illumination.
The best value (not the lowest price) in consumer thermal IR imagers is still the FLIR E4. It's only 80x60 out of the box but anyone active on s.e.d or s.e.r can probably follow the thread over at
With the advent of the Lepton and similar low resolution sensors it's unlikely FLIR (or anyone else) will release a 320x240 imager that's software-crippled to a lower res anytime soon so this is pretty much it for bargain sensors, until maybe the next gen.
It's also possible to add a ZnSe close-up aux lens which makes it quite useable for PCB inspection. FLIR doesn't supply them, so it's ebay for the lenses and 3D printing for the holders. Works, though.
The trouble with thermal imaging cameras is I'm not sure they would work for this purpose. I've not found an image of something that is natively at the same temperature to its surroundings , hidden under something visibly opaque. I'm assuming some sort of search heat source would be required to highlight hidden black from white, but I've not found a pic of someone doing that . My coupled use of non contact pyro thermometer and external heat sourse seemed to just show differences in the thermal capacity of the underlying structure. to differentiate , it would probably require a very specific wavelength of IR, specific for picking up paint v plaster difference.
If you're lucky, the paint might be electrically conductive. (My fave Krylon #1609 is.)
Then an induction coil to heat it plus a thermal camera to see the warm spots might be the ticket.
Cheers
Phil Hobbs
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.