I like how you are thinking outside the box a bit.
Having said that, there are a couple of issues:
1) This is really not a hard problem for a computer to solve. Detecting different image types is pretty trivial. You made it a lot easier by making the letters sit in the boxes in a regular manner.
2) A high percentage of captchas are now broken by humans, instead of robots. This is not a problem you will likely encounter in a small site, though.
(1) Is an easier problem to solve. You could do "input the letters near the cat", or "input the letters near the map of florida".
(2) Is harder, but most of the people who break captchas live in 3rd world countries, so you could possibly arbitrage cultural knowledge (see florida example above). They will catch on pretty damn quick, though.
The first example is not indicative of the actual difficulty because the letters are pink (this is noted, but still).
Beyond that, the examples use extremely hard to see letters... that completely defeats the purpose. There's no need if the challenging part for the computer is detecting satellite imagery.
There are multiple letters on some of the images, sometimes hovering between two images. Again, why? The computer is going to have no problem divvying up the images, but the ambiguity is confusing for humans.
Make the letters as easy to read as possible please. If that breaks your captcha, find a new captcha method.
In fact (and I can't believe I didn't think of this before, why have letters at all?!? The user has to scan over the images then go back to the text field and enter in characters. Why not just click to select?
Great idea but I don't think it's the solution to the captcha problem. There's a lot of visual noise for a person to analyse and I found it considerably harder than reCaptcha types. However that may be a bonus for deterring human solved captcha à la deathbycaptcha.com
It's not very quick or readable, and it's not useful for users with impaired vision.
Technically it has two weaknesses:
- non-map images are chosen from a small set. Good implementation will need equally large set of non-map (and presumably non-offensive and deceptively map-like) images.
- satellite photos don't seem very hard to identify, e.g. even using histogram as an input to a simple classification algorithm could give decent accuracy (gray/brown/dark green color scheme dominates) and tree shadows, cars and rooftops may be easy features to detect with fancier algorithms.
I was expecting "click on Tennessee" and then a region of the US - which might actually be an appropriate captcha if you assume the site users know US state geography.
This however, is difficult to read for a human and probably easy to write a bot for.
one thing i found very irritating: in one of the tests that asked to type the group of letters where the image is from satellite, i find the group "y p k a 1" but the answer i got was 'wrong, group is "k p y a 1"'. if i am not mistaken, based on the mathematical definition of the group, groups "y p k a 1" and "k p y a 1" are the same
I was expecting this to be a random lat/lng and asking you to pick (for example) the most prominent feature displayed in the map. A major city or a famous mountain or something.
An interesting approach, although one line of javascript will solve all samples beyond the first one (didn't bother writing extra code to handle your demo case):
Ok, obviously the above is mostly in jest. I always enjoy seeing new captchas, despite most having major flaws. I do have to echo other comments that mention the user difficulty issue. Using this in production would require more explanatory text than your average captcha, and it does seem to take a little longer to figure out than, say, a recaptcha image. That said, likely the bigger issue is one of image classification allowing bots to solve most of them.
the above is also directly ignoring the text immediately below on the page:
This is dumb! I already figured out how to crackit, and yes, of course I realize that the productionversion will render the MAPTCHA as a single image,not as separate elements, and it will draw from apractically limitless database of images.
I'm not saying this is a good captcha, but if your first response is seriously to not read any of the text on the page and go for the easy "hey, look how dumb this is" answer, are you really trying hard enough?
I did read that. And that is why I added the "in jest" part, followed by my serious comment. However, rereading my comment, I should have said "in jest" instead of "mostly in jest."
ah, I can re-read and see how that is meant as a joke specifically targeting the text I quoted. Gotta love the internets, where no one can hear your intonation and subtleties
This might have been my downfall too, personally it was more of a scan over the entire image and type letters as I see satellite images, didn't even notice I was doing it like that until I saw your comment. I read something a while back how "left brain hemisphere" people scan sequentially and "right hemisphere" people jump in to find things at random. I can't speak for its accuracy or its place in reality but the end result is that this thing could do with accepting values in any order.
Not sure if that'd make the whole thing less useful however..
I don't think that would make it less useful; either the bot reads it right or it doesn't, the leeway introduced by allowing any order input is miniscule, I think. So I updated it so it won't get fussy about order.
It's good that there's effort going into improving captchas but IMO the big issue is finding something that also works for users with poor or no sight.
At OpenStreetMap we would love a map-based CAPTCHA that actually contributes to our map. Right now we (like most people) use ReCAPTCHA on our wiki... which has the rather exasperating effect of contributing map data (StreetView house numbers) to our main competitor, Google.
On the upside, you can usually type a wrong number into the ReCAPTCHA. ;)
Hi everybody, designer here. I really appreciate all the feedback; this concept is still in its infancy and can obviously use quite a bit of improvement, but I'm trying to tweak it as best I can to maximize its usability.
Couple of notes:
1. The ultimate intention is to either pull images (both satellite and decoy) from the web as they are needed or create a database of images and update it regularly. Thus identifying images as real or decoy will only work on a short-term basis; there will always be new images entering the fold.
2. If people like this concept well enough to keep at it, I'm planning to automate some kind of color-matching scheme so that it's not possible to just look at the histograms of the images and identify the satellite images thusly.
3. I'm only just starting to learn about OCR, so at this moment I have no idea whether it's worth bothering to disguise letters or even whether it's worth using letters at all. I prefer having a two-tiered approach; having to ID a satellite image AND identify characters. If the bots are too good at identifying the characters and/or it's too hard for humans, that just leaves the satellite question and from what I've read that's not a high enough level of entropy.
4. I don't have anything lined up for an audio solution just yet. I'm fooling around with a few concepts, but nothing worth mentioning just yet.
5. Thanks to all those who wrote in suggestions/solutions and signed up!
Having said that, there are a couple of issues: 1) This is really not a hard problem for a computer to solve. Detecting different image types is pretty trivial. You made it a lot easier by making the letters sit in the boxes in a regular manner. 2) A high percentage of captchas are now broken by humans, instead of robots. This is not a problem you will likely encounter in a small site, though.
(1) Is an easier problem to solve. You could do "input the letters near the cat", or "input the letters near the map of florida".
(2) Is harder, but most of the people who break captchas live in 3rd world countries, so you could possibly arbitrage cultural knowledge (see florida example above). They will catch on pretty damn quick, though.
reply