Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
Technology Science

Smell Camera Snapshots Scents For the Future 117

Posted by samzenpus from the a-picture-is-worth-a-thousand-smells dept.
An anonymous reader writes "Designer Amy Radcliffe has created an 'analog odor camera' that can be used to recreate a smell. From the article: 'When a smell source is placed under the device's glass cone, a pump extracts the smell via a plastic tube. After being drawn to Madeleine's main unit, the smell goes through a resin trap which absorbs the particles so molecular information can be recorded. That data is expressed in a graph-like formula, which essentially contains a fingerprint of the smell. In a special lab, that formula can then be inscribed on a bronze disk to artificially reproduce the smell. The smell can also be recreated in small vials.'"
This discussion has been archived. No new comments can be posted.

Smell Camera Snapshots Scents For the Future More

Smell Camera Snapshots Scents For the Future

Comments Filter:

  • Cheap Perfume (Score:4, Insightful)

    by Master Moose ( 1243274 ) on Sunday July 07, 2013 @08:02PM (#44212765) Homepage

    I cannot see the big scent manufacturers liking this one. .

    • Eaux De Feces

    • Re:Cheap Perfume (Score:4, Funny)

      by iggymanz ( 596061 ) on Sunday July 07, 2013 @08:06PM (#44212777)

      no worries, the reproduced scent would be approximate. this has been tried before with collection of supposed "primary scents" but with little success when judged by real human noses.

        I predict a limited market for augmenting internet porn.

      • Re: (Score:3, Interesting)

        by postbigbang ( 761081 )

        One person's scent of smashed peanut powder is another person's death. Hacking a scent generator seems both easy, and protections dubious at best.

        • Re: (Score:3)

          by Meski ( 774546 )
          Are people allergic to the smell of peanuts, produced synthetically, or the real thing?

          • Go ahead and test the histamine reactions. Stand by with your epi pen. This isn't the only possibility, either. There are any number of easily made aromatic chemical combos with crazy reactions.

            • Re: (Score:2)

              by Meski ( 774546 )
              So it's more likely we'd go forth and find strange new allergy reactions, than commonplace peanut ones. Why peanuts, anyway? Do people get reactions from other common legumes? (peas, beans)

              • Statistically, peanuts are strangely deadly. Rapid onset reaction, have an epi pen nearby or perform a ballpoint tracheal breathing way, or suffocate, more or less within 2min.

                There are other combos that can do this as well, but this is one of the fastest. Raw nicotine can kill you pretty quickly, too. The list is endless. Think of keyword chemicals that your friends at the NSA would dock you with.
                Play a bit with captive subjects.

                Death by scent-o-gram
                !

              • Re:Cheap Perfume (Score:4, Interesting)

                by iggymanz ( 596061 ) on Sunday July 07, 2013 @08:55PM (#44212995)

                similar proteins are found in two other plants, lupin (a legume) and Fenugreek (not a legume), and people with peanut allergies often react to those as well.

          • Re:Cheap Perfume (Score:4, Informative)

            by iggymanz ( 596061 ) on Sunday July 07, 2013 @08:41PM (#44212949)

            the reaction is to several proteins found in peanuts, something that "smells like peanuts" but not containing specific allergens would not trigger immune reaction.

            • Except if one protein required for the scent is the one that triggers the reaction. Not likely, but we do not know how specific the device can be. Is it able to report from a set of predefined molecules (which could include allergic ones), or is it able to detect specific molecular functions, which would be a smaller sequence of proteins with less odds of producing an allergy.
              • What's described is a smell recording device. The smell reproducing device is another machine, and much more difficult to make, methinks.
          • With the right hack, your body won't know the difference, even if it doesn't actually *smell* like peanuts.

        • One person's scent of smashed peanut powder is another person's death. Hacking a scent generator seems both easy, and protections dubious at best.

          Q: You know what's easier than hacking a scent generator in a hope that it will generate a facsimile close enough to peanuts such as to produce an allergic reaction?

          A: Throwing a handful of peanuts at the guy.

          Oh wait, you wanted it to be covert?

          A: Soak a letter in a solution of peanut butter and water. Dry. Mail letter.

          I'm not mocking peanut allergies here, I

          • Certainly.

            But a good hack is potential death (intended or otherwise) to a scent generator. We won't go to castor beans, etc.

      • Re:Cheap Perfume (Score:4, Insightful)

        by davydagger ( 2566757 ) on Sunday July 07, 2013 @11:44PM (#44213569)
        goatse and tubgirl get a whole new dimension.

        the only people who will use this are trolls
    • As always, early adoption will be driven by the porn industry.

      • na, japanese men are perfectly happy with tuna fish scented panties

        • The whole point would be to isolate the scent from the source. With this tech maybe it could be proven which panties are fraudulent and which came from real vag.
          A tuna smell usually means she needs a course of antibiotics. A tuna smell isn't that bad. It's the garbage smell that really makes a body want to hurl. Either way the answer is doxycycline. Or a better girlfriend.

          What would be far more interesting is a vag scent fingerprint with this tech. Hell, you could probably even use it for ID. Instead of log

          • Re: (Score:2)

            by Macgrrl ( 762836 )

            There's a heap of reasons why your authentication probe wouldn't work, none the least being a woman's smell varies based on her menstrual cycle. And how do guys authenticate? Free prostate examine perhaps?

            • Ah yes and probably ovulaion as well. I'd forgotten about the whole menstrual cycle thing. Been a while since I've played with kitty. After 20 years you start to forget stuff like that. Well maybe fuzzy logic? Training a neural network based on a wide range of scents?

              The scent fingerprint may not be as invariant as actual fingerprints or retinal scans, but it is still something you are (I suppose) which does seem to be unique based on my admittedly limited experience. Actually I just realized that a quick f

      • Re: (Score:3)

        by vlad30 ( 44644 )
        Tub Girl now with real aroma ! NO THANKS!!!

      • As always, early adoption will be driven by the porn industry.

        That sure worked out well for HD-DVD!

    • I cannot see the big scent manufacturers liking this one. .

      Get ready for the Digital Smellenium Act.

    • On the contrary, they will love it because now you can smell that someone was wearing their perfume. Meanwhile, people are already using science to clone perfumes more accurately than this device will manage.

    • Re: (Score:1)

      by Xicor ( 2738029 )
      actually, i would say it would make them quite happy... they would now have a much easier time prosecuting copycat perfumes

  • Moderator is drunk again (Score:4, Funny)

    by Princeofcups ( 150855 ) <john@princeofcups.com> on Sunday July 07, 2013 @08:06PM (#44212779) Homepage

    "In a special lab, that formula can then be inscribed on a bronze disk to artificially reproduce the smell." What, is this from the Church of Mormon?

    • Re:Moderator is drunk again (Score:4, Funny)

      by icebike ( 68054 ) on Sunday July 07, 2013 @08:37PM (#44212913)

      Actually it said:

      In a special lab, that formula can then be inscribed on a bronze disk to artificially reproduce the smell. The smell can also be recreated in small vials.'"

      Neither of these sound optimum. What you would want is the building blocks of those scents stored in little plastic ink-jst like cartridges, each holding half a dozen or so different molecule mixtures. It would be vitally important to size these cartridges so that the most common components would be in the smallest cells, such that it would run out first, requiring you to buy the entire cartridge well before the rest of the compounds were exhausted, You also want to be sure it isn't refillable.

      Then you can almost give the smell generator away, and make a fortune selling smell-cartridges.
      This would allow you to sell the entire patent structure to HP, and retire on your profits, and thumb your noses (figuratively and literally) at the world by releasing the first Olfactory Goatse.

      • No less than two companies had products precisely as you describe at GDC 2000. It has been some time since then...

    • No, those would be gold disks. Bronze is used in this case because it denotes third place, as in, not as good as the original and not even the next best thing. But kinda also pretty good.
    • "A freshly mowed lawn, lit charcoal and sunscreen are just a few summer smells that are as ephemeral as they are memorable." Alternatively go outside , mow the lawn ,then start a BBQ and slap on some sunscreen - i suspect that will be cheaper and more readily available. These smells hardly require complex analysis in order to recreate them . We can do that just fine. It would perhaps be more useful if this device could be put to use for example to capture the smells of fruits, or flowers that are rare or

  • New perfume (Score:5, Funny)

    by PPH ( 736903 ) on Sunday July 07, 2013 @08:08PM (#44212789)

    Chanel No. 0x05

  • Can't believe this made it past the editors (Score:5, Insightful)

    by bshell ( 848277 ) on Sunday July 07, 2013 @08:16PM (#44212825)
    This is vapourware. Perhaps pun intended. Imaginative, kind of, but such devices have been predicted since the early 1900s. Never comes to market. Why? Headspace analysis is super expensive. Even the right library of molecular signatures, which would be needed to interpret the output of the GC/Mass Spec is in the neighbourhood of tens, if not hundreds of thousands of dollars. Then you need a really good GC/Mass Spectrometer machine. Or at least a really good Gas Chromatograph. All this is very expensive, in the hundreds of thousands of dollars neighbourhood. And that's JUST TO FIND OUT WHAT MAY BE IN THE SMELL. Then you need a professional perfumer and a perfume house with thousands of fresh organic compounds available to reblend the smell. This is a million dollar project, or at the very least tens of thousands for each "photograph". Prohibitively expensive. I cannot believe a moderator let this one through at slashdot. Oh well.

    • Re: (Score:3)

      by Dracos ( 107777 )

      Well, now that the folly that is 3d TV/movies is all but dead (again), what makes you think the TV and film industries aren't working to bring the long-awaited Smell-O-Vision to market?

      • Re: (Score:2)

        by AK Marc ( 707885 )
        all but dead? It's getting harder to find a top-end TV without 3D built in. I bought a 3D TV because it was cheaper than the same model in 2D. And more 3D movies are released every year than the year before. How is that "all but dead"?

    • Re: (Score:2)

      by Meski ( 774546 )
      Like the early days of photography were. Grainy, black and white, expensive. SO you will have a relatively coarse smell camera to start with as well, but that's no reason to dismiss it.

      • Re: (Score:3)

        by bshell ( 848277 )
        The difference here is that in the early days of photography we knew that light created images, and we knew that lenses refracted light in such a way that we could capture it. We also had silver halide which changed from clear to black on exposure to light. In other words, we had a lot of technologies based on light that we understand. At the moment this simply does not exist for the sense of smell.

        • Re: (Score:2)

          by Meski ( 774546 )
          Agreed. At the moment we have sensors for a lot of these smell related things, but they're expensive (except for the extremely narrow use dedicated ones)

    • Re: (Score:2, Funny)

      by Anonymous Coward

      So you're saying that it doesn't pass the smell test?

    • Re: (Score:2)

      by icebike ( 68054 )

      Even the right library of molecular signatures, which would be needed to interpret the output of the GC/Mass Spec is in the neighbourhood of tens, if not hundreds of thousands of dollars.

      Ah but think out of the box (lab).

      You probably wouldn't need a mass spec, what with the newer chem-lab-on-a-chip that they are coming out with these days. Remember, you are not trying to identify the elements involved so as to faithfully reproduce them. You merely want something that smells like them. Coffee tasters and perfume smellers have been doing this by nose for decades.

      In other words, you don't care what elements are actually there, you only care what they smell like. Given the human sense of sm

      • Remember, you are not trying to identify the elements involved so as to faithfully reproduce them. .

        But do we know all receptors involved in smell, and are we able to reproduce them on a chip? And in what extent small receptors differ from an individual to another?

        • Re: (Score:2)

          by icebike ( 68054 )

          Receptors counts probably don't differ that much, but individuals might have different sensitivities.

          As for your first question, Google found this in under a second:

          There are a large number of different odor receptors, with as many as 1,000 in the mammalian genome which represents approximately 3% of the genes in the genome. However not all of these potential odor receptor genes are expressed and functional. According to an analysis of data derived from the human genome project, humans have approximately 400 functional genes coding for olfactory receptors and the remaining 600 candidates are pseudogenes.

          ...

          The reason for the large number of different odor receptors is to provide a system for discriminating between as many different odors as possible. Even so, each odor receptor does not detect a single odor. Rather each individual odor receptor is broadly tuned to be activated by a number of similar odorant structures.

          http://en.wikipedia.org/wiki/Olfactory_receptor [wikipedia.org]

          So if you could narrow it down to 400 chemicals you might be able to fine tune it manually (by nose) to even fewer that are "good enough" matches.

          • Re:Can't believe this made it past the editors (Score:4, Interesting)

            by bshell ( 848277 ) on Sunday July 07, 2013 @09:37PM (#44213149)
            Smell receptors are very similar to immune system receptors. They are designed to use combinatorics so that they can "identify" on the order of trillions of different shaped molecules. In fact they are designed to "detect" molecules that have never even been invented. The mathematics of smell combinatorics is what makes it (currently) an intractable problem. Also, I believe we only know what a few of the human smell receptors detect, in terms of molecular shape and smell. It's extremely difficult to conduct experiments on smell--i.e. on how smell receptors work--in human subjects. There are countless problems and this is why smell remains a mystery. This story is not about *science*. It's about an Art Project. As long as you treat the original post as a fanciful work of art, all is fine. As soon as you assume that anything like that can be created with present day technology you are off in dreamland.

            • Re: (Score:2)

              by icebike ( 68054 )

              Combinatorics works in your favor when trying to make a smell library.

              You don't have to be that precise. You don't even have to know which receptors sence which compounds.
              Enough trials followed by less chemistry will get you there faster.
              You are probably better off relying on discriminating noses, as do perfume companies, and tell the chemists to shut up, sit down, watch and learn.

          • Genes don't necessarily map one-for-one with receptors. A few genes might differentiate many times as many kinds of sensors. How many unique numbers can you represent with 400 bits?

            • Re: (Score:2)

              by tftp ( 111690 )

              How many unique numbers can you represent with 400 bits?

              That's hardly the right question to ask. Have a look at all kinds of errors in ADCs and DACs. Want to buy a 3 GHz, 24-bit ADC? I will sell you one. It will produce codes from 0x000000 to 0xFFFFFF. Unfortunately, those are the only codes it will ever produce... this is just one example.

      • Re: (Score:3)

        by bshell ( 848277 )
        Perhaps you are not a chemist. When you say, "You merely want something that smells like them," how do you characterize that smell in the first place? Are you aware that there is no system for doing this today? Smell is not like light or sound, where there is a well understood spectrum of vibrations or frequencies and all you have to do is duplicate those frequencies. Nobody knows what smell is. If we knew that, then "smell photographs", as well as copiers, smell-phones and even "smell glasses" to correct a

        • Re: (Score:2)

          by Macgrrl ( 762836 )

          I am perpetually surprised that given taste is 90% scent, that certain foods which smell tantalising taste like crap - hot dogs and bad coffee for example.

        • Re: (Score:2)

          by sjames ( 1099 )

          The first step is working that spectrum out. The human nose can only have finite receptors. We know aliasing happens so it's not a single receptor for each smell, but more a matter of what combination of receptors is triggered.

          There's a lot of work ahead, but it's going to help to have devices such as the one in TFA to help figure it all out.

          Science actually has a very good idea how smell works in overview, it's just that there's a great deal of detail that has to be worked through.

      • Re: (Score:2)

        by Dunbal ( 464142 ) *

        I doubt, at the end of the day if humans actually have smell receptors for everything.

        Like water, for instance. Most animals have no trouble identifying and finding water even in a new environment. I don't think there are examples of humans that can do this - not by smell alone.

        • Nonsense. Animals can't smell water itself though because they're made of it. Their own body's water would be constantly flooding (pun intended) their water recepetprs.

          I can "smell water", by smelling the things that are commonly in water, like algae and I can smell if there are fish in a pond or stream, for instance. I can smell humid air, or the changes that humidity makes in the smells of other things anyway. I can also smell and taste a number of other things that many people say they can't smell. M

        • Like water, for instance. Most animals have no trouble identifying and finding water even in a new environment. I don't think there are examples of humans that can do this - not by smell alone.

          Blindfold me and take me to within 500 meters of a lake, pond, or small river. Assuming a natural environment and that I'm downwind, I'd lead you in a beeline right to the water source. It's nothing amazing, it's just being familiar with the scents associated with bodies of water. I'd wager anyone with an average s

    • However, I so want to go to H.R. and have evidence that that that huge thing that got off the elevator 50 feet from me is NOT complying with company policy about wearing strong perfumes.
      Next up, heavy smoker after every break and the dude that doesn't have to shower because his sweat doesn't smell, just ask him.

  • And you thought..... (Score:3)

    by MasseKid ( 1294554 ) on Sunday July 07, 2013 @08:38PM (#44212925)
    And you thought autoplay ads with sound were bad, just wait, autoplay ads with smells are coming. And then will come the trolling....

    • Re: (Score:2)

      by msobkow ( 48369 )

      Yeah, but it could give a whole new meaning to the ever popular fart apps for phones... :P

    • Re: (Score:2)

      by mjwx ( 966435 )

      And you thought autoplay ads with sound were bad, just wait, autoplay ads with smells are coming. And then will come the trolling....

      Now you will know I've farted in your general direction.

  • Mildly amusing, no practical purpous.
  • this is going to be real expensive at first, probably won't be readily available for 30 yrs to the common folks, but, on the other hand, this can lead to very interesting projects, from new forms of entertainment vehicle to new forms of crime scene investigation tools and perhaps even for memory therapy and other uses which I can't think of at the moment.

  • IIRC, this is the third sensation that makes the tri of the tricorder.

  • There is much known about the connection between smell and other aspects of the human experience. Photographs are more objective. Scents are more subjective. And a scent might remind someone of a personal experience that is important to them, but an objective 'viewer' would not experience the same when presented with even a 100% rendering of a scent.

    And even for the original scentographer, the smell revisited may some across very differently that the one remembered as ambient smells are filtered out. Yo

  • Google Nose [google.com] is real?

  • Time to buy stock in John Waters' production company!

  • So in just a few years we might have a way to remember dining at that fabulous Mexican restaurant literally forever.

    ;-)

  • Just as long as you don't take it to space and make me smell Uranus.

  • Needless to say, the actual article left me disappointed.

  • Inscribed on a bronze disk? Is this a research project or an art installation?

  • The summary mentions a *designer*. Do we have a working prototype (cause that would be a sensation right away) or just a design mock-up? "Here is what a smell camera could look like. We'll build one as soon as someone discovers how our mysterious "main unit" could work"

  • Interesting article—makes me think of four things:

    1) The Harold and Maude movie: she invented a method of smell playback.
    2) Scratch and Sniff technology (microencapsulation) may reach a whole new level.
    3) Are there "primary smells" like primary colors? If so, imagine people creating new smells and posting the formulas online. I imagine engineered tastes would be possible too, as it's a closely related sense.
    4) Imagine if High Times starts using that technology...

  • Ready? I'm gonna capture this on Smell-O-Vision, so hold yer farts!

  • I think "creates" doesn't mean what you think it means.

    She 'created' this "scent camera" the way I just "had sex with" Nicole Kidman.

    This "story" links to an aggregator, that links to a blog where she talks about "wouldn't this be cool?"

    She built a model of one with some tubes, a glass bell jar, and a ceramic pot, and then took a picture of it. Either the /. editors are colossally lazy or stupid that this even got posted.

  • If we're building stuff from local stores of ((grey goo)), why not just D/L plans for LSD ?

    and yes I'm joking about the feasibility

  • Looney Tunes predicted this years ago only it is 13 years late. "Smell-O-Vision Replaces Television: Carl Stalling Sez It Will Never Work!" (The Old Grey Hare 1944 Warner Bros.)

Slashdot Top Deals

You can tune a piano, but you can't tuna fish. You can tune a filesystem, but you can't tuna fish. -- from the tunefs(8) man page

Close