Loading ...
Sorry, an error occurred while loading the content.
 

Question: Human detection

Expand Messages
  • darthjder
    Does anyone have any ideas on what kind of component to use in an IR heat sensing application, particularly for detecting human body heat? So far I ve got
    Message 1 of 7 , Jan 31, 2006
      Does anyone have any ideas on what kind of component to use in an IR
      heat sensing application, particularly for detecting human body heat?

      So far I've got circuits that are sensitive mostly to IR light than
      heat. (Phototransistor-based: trying to use Radioshack parts only) And
      it looks like RS has canceled thermistors.

      The goal is to make anything that detects people (A stalker-bot, since
      a robot following a person can be seen as both hostile or friendly;
      That, and the idea of replicating the Mechanical Hound's behavior
      sounds cool), even while the robot is running. E-field detectors I
      know have been tried, but antennae are a bit of a hassle.

      Just want to see what else is reliable besides sound: It's hard to be
      "sonitropic", since one has to be sensitive (So the robot can find
      differences in both ears) while not being vulnerable to any noise in
      the circuit.

      Idea: Use an opamp neuron network, then use the BEAMsonics front-end
      to create a signal the varies the threshold of the opamp. Advantages
      to this are the availability of parts (Opamps are available at
      Radioshack) and the fact it speeds up neuron networks instead of
      slowing them down. (When a loud sound is made, it would be best for
      the robot to behave in an animal-like manner and jump, not lock up and
      hibernate.)

      Any better ideas? Preferably ones that don't require the human being
      to be lit up in a radiant flame just so the robot can seek the light
      of the firey glow on his/her head?
    • jeffhoron
      How about hacking one of these: http://www.goldmine-elec-products.com/prodinfo.asp? number=G4567&variation=&aitem=17&mitem=24 ... IR ... heat? ... And ...
      Message 2 of 7 , Feb 2, 2006
        How about hacking one of these:

        http://www.goldmine-elec-products.com/prodinfo.asp?
        number=G4567&variation=&aitem=17&mitem=24


        --- In beam@yahoogroups.com, "darthjder" <b6ol@...> wrote:
        >
        > Does anyone have any ideas on what kind of component to use in an
        IR
        > heat sensing application, particularly for detecting human body
        heat?
        >
        > So far I've got circuits that are sensitive mostly to IR light than
        > heat. (Phototransistor-based: trying to use Radioshack parts only)
        And
        > it looks like RS has canceled thermistors.
        >
        > The goal is to make anything that detects people (A stalker-bot,
        since
        > a robot following a person can be seen as both hostile or friendly;
        > That, and the idea of replicating the Mechanical Hound's behavior
        > sounds cool), even while the robot is running. E-field detectors I
        > know have been tried, but antennae are a bit of a hassle.
        >
        > Just want to see what else is reliable besides sound: It's hard to
        be
        > "sonitropic", since one has to be sensitive (So the robot can find
        > differences in both ears) while not being vulnerable to any noise
        in
        > the circuit.
        >
        > Idea: Use an opamp neuron network, then use the BEAMsonics front-
        end
        > to create a signal the varies the threshold of the opamp.
        Advantages
        > to this are the availability of parts (Opamps are available at
        > Radioshack) and the fact it speeds up neuron networks instead of
        > slowing them down. (When a loud sound is made, it would be best for
        > the robot to behave in an animal-like manner and jump, not lock up
        and
        > hibernate.)
        >
        > Any better ideas? Preferably ones that don't require the human
        being
        > to be lit up in a radiant flame just so the robot can seek the
        light
        > of the firey glow on his/her head?
        >
      • Boštjan Jerko
        This is an IR and not body heat detector. B. ... -- www.japina.si
        Message 3 of 7 , Feb 2, 2006
          This is an IR and not body heat detector.

          B.

          On 2/3/06, jeffhoron <jeffhoron@...> wrote:
          How about hacking one of these:

          http://www.goldmine-elec-products.com/prodinfo.asp?
          number=G4567&variation=&aitem=17&mitem=24


          --- In beam@yahoogroups.com, "darthjder" <b6ol@...> wrote:
          >
          > Does anyone have any ideas on what kind of component to use in an
          IR
          > heat sensing application, particularly for detecting human body
          heat?
          >
          > So far I've got circuits that are sensitive mostly to IR light than
          > heat. (Phototransistor-based: trying to use Radioshack parts only)
          And
          > it looks like RS has canceled thermistors.
          >
          > The goal is to make anything that detects people (A stalker-bot,
          since
          > a robot following a person can be seen as both hostile or friendly;
          > That, and the idea of replicating the Mechanical Hound's behavior
          > sounds cool), even while the robot is running. E-field detectors I
          > know have been tried, but antennae are a bit of a hassle.
          >
          > Just want to see what else is reliable besides sound: It's hard to
          be
          > "sonitropic", since one has to be sensitive (So the robot can find
          > differences in both ears) while not being vulnerable to any noise
          in
          > the circuit.
          >
          > Idea: Use an opamp neuron network, then use the BEAMsonics front-
          end
          > to create a signal the varies the threshold of the opamp.
          Advantages
          > to this are the availability of parts (Opamps are available at
          > Radioshack) and the fact it speeds up neuron networks instead of
          > slowing them down. (When a loud sound is made, it would be best for
          > the robot to behave in an animal-like manner and jump, not lock up
          and
          > hibernate.)
          >
          > Any better ideas? Preferably ones that don't require the human
          being
          > to be lit up in a radiant flame just so the robot can seek the
          light
          > of the firey glow on his/her head?
          >






          SPONSORED LINKS
          Computer science distance education Computer science course Computer science school
          Computer science degree Computer science and education Computer science education


          YAHOO! GROUPS LINKS






          --
          www.japina.si
        • jeffhoron
          Objects emit infrared radiation in proportion to temperature. See: http://imagers.gsfc.nasa.gov/ems/infrared.html Scroll down to How can we see using the
          Message 4 of 7 , Feb 3, 2006
            Objects emit infrared radiation in proportion to temperature. See:

            http://imagers.gsfc.nasa.gov/ems/infrared.html

            Scroll down to "How can we "see" using the Infrared?"

            The item I linked to is intended to detect people. I suspect that
            two could be used to (crudely) follow a person. Or if someone was
            feeling bold they might take out the infrared sensors or purchase
            some other very sensitive IR photosensors and use them in a
            differentiating circuit.



            --- In beam@yahoogroups.com, Boštjan Jerko <bojerko@...> wrote:
            >
            > This is an IR and not body heat detector.
            >
            > B.
            >
            > On 2/3/06, jeffhoron <jeffhoron@...> wrote:
            > >
            > > How about hacking one of these:
            > >
            > > http://www.goldmine-elec-products.com/prodinfo.asp?
            > > number=G4567&variation=&aitem=17&mitem=24
            > >
            > >
            > > --- In beam@yahoogroups.com, "darthjder" <b6ol@> wrote:
            > > >
            > > > Does anyone have any ideas on what kind of component to use in
            an
            > > IR
            > > > heat sensing application, particularly for detecting human body
            > > heat?
            > > >
            > > > So far I've got circuits that are sensitive mostly to IR light
            than
            > > > heat. (Phototransistor-based: trying to use Radioshack parts
            only)
            > > And
            > > > it looks like RS has canceled thermistors.
            > > >
            > > > The goal is to make anything that detects people (A stalker-
            bot,
            > > since
            > > > a robot following a person can be seen as both hostile or
            friendly;
            > > > That, and the idea of replicating the Mechanical Hound's
            behavior
            > > > sounds cool), even while the robot is running. E-field
            detectors I
            > > > know have been tried, but antennae are a bit of a hassle.
            > > >
            > > > Just want to see what else is reliable besides sound: It's
            hard to
            > > be
            > > > "sonitropic", since one has to be sensitive (So the robot can
            find
            > > > differences in both ears) while not being vulnerable to any
            noise
            > > in
            > > > the circuit.
            > > >
            > > > Idea: Use an opamp neuron network, then use the BEAMsonics
            front-
            > > end
            > > > to create a signal the varies the threshold of the opamp.
            > > Advantages
            > > > to this are the availability of parts (Opamps are available at
            > > > Radioshack) and the fact it speeds up neuron networks instead
            of
            > > > slowing them down. (When a loud sound is made, it would be
            best for
            > > > the robot to behave in an animal-like manner and jump, not
            lock up
            > > and
            > > > hibernate.)
            > > >
            > > > Any better ideas? Preferably ones that don't require the human
            > > being
            > > > to be lit up in a radiant flame just so the robot can seek the
            > > light
            > > > of the firey glow on his/her head?
            > > >
            > >
            > >
            > >
            > >
            > >
            > >
            > > SPONSORED LINKS
            > > Computer science distance
            education<http://groups.yahoo.com/gads?
            t=ms&k=Computer+science+distance+education&w1=Computer+science+distan
            ce+education&w2=Computer+science+course&w3=Computer+science+school&w4
            =Computer+science+degree&w5=Computer+science+and+education&w6=Compute
            r+science+education&c=6&s=196&.sig=DRCAZLYpj1AU-i4I_CfVVg> Computer
            > > science course<http://groups.yahoo.com/gads?
            t=ms&k=Computer+science+course&w1=Computer+science+distance+education
            &w2=Computer+science+course&w3=Computer+science+school&w4=Computer+sc
            ience+degree&w5=Computer+science+and+education&w6=Computer+science+ed
            ucation&c=6&s=196&.sig=4N5M-JPpQGIm9UoThjyK2w> Computer
            > > science school<http://groups.yahoo.com/gads?
            t=ms&k=Computer+science+school&w1=Computer+science+distance+education
            &w2=Computer+science+course&w3=Computer+science+school&w4=Computer+sc
            ience+degree&w5=Computer+science+and+education&w6=Computer+science+ed
            ucation&c=6&s=196&.sig=iITDS_V1poIqaH-o4VX2mA> Computer
            > > science degree<http://groups.yahoo.com/gads?
            t=ms&k=Computer+science+degree&w1=Computer+science+distance+education
            &w2=Computer+science+course&w3=Computer+science+school&w4=Computer+sc
            ience+degree&w5=Computer+science+and+education&w6=Computer+science+ed
            ucation&c=6&s=196&.sig=xATNVTlTBzpDU37HWUfkqw> Computer
            > > science and education<http://groups.yahoo.com/gads?
            t=ms&k=Computer+science+and+education&w1=Computer+science+distance+ed
            ucation&w2=Computer+science+course&w3=Computer+science+school&w4=Comp
            uter+science+degree&w5=Computer+science+and+education&w6=Computer+sci
            ence+education&c=6&s=196&.sig=ixqGYPX6VVk2-XSwgwoNZA> Computer
            > > science education<http://groups.yahoo.com/gads?
            t=ms&k=Computer+science+education&w1=Computer+science+distance+educat
            ion&w2=Computer+science+course&w3=Computer+science+school&w4=Computer
            +science+degree&w5=Computer+science+and+education&w6=Computer+science
            +education&c=6&s=196&.sig=moaPxFJxH8aCToZI4Qhlcw>
            > > ------------------------------
            > > YAHOO! GROUPS LINKS
            > >
            > >
            > > - Visit your group "beam
            <http://groups.yahoo.com/group/beam>" on
            > > the web.
            > >
            > > - To unsubscribe from this group, send an email to:
            > > beam-unsubscribe@yahoogroups.com<beam-
            unsubscribe@yahoogroups.com?subject=Unsubscribe>
            > >
            > > - Your use of Yahoo! Groups is subject to the Yahoo! Terms of
            > > Service <http://docs.yahoo.com/info/terms/>.
            > >
            > >
            > > ------------------------------
            > >
            >
            >
            >
            > --
            > www.japina.si
            >
          • WilliamChops Westfield
            ... Hmm. contact-less thermometers are becoming relatively affordable, and might provide a much more directional sensor than the passive IR detectors like the
            Message 5 of 7 , Feb 3, 2006
              On Feb 3, 2006, at 7:32 PM, jeffhoron wrote:

              > Objects emit infrared radiation in proportion to temperature.

              Hmm. contact-less thermometers are becoming relatively affordable,
              and might provide a much more directional sensor than the passive
              IR detectors like the one you reference (which are pretty much
              designed NOT to be directional.)

              http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=91778

              BillW
            • jeffhoron
              Interesting idea. That s the lowest-cost non-contact thermometer I ve seen with a decent (8:1) distance to spot (measured area size) ratio. I suppose it
              Message 6 of 7 , Feb 4, 2006
                Interesting idea. That's the lowest-cost non-contact thermometer
                I've seen with a decent (8:1) distance to spot (measured area size)
                ratio. I suppose it depends on Darthjder's intended application, so
                Darthjder please tell us if we're headed down the path you wanted to
                go. Two PIR sensors could provide left/right/straight ahead info
                similar to a simple photovore for $11. Maybe some blinders would
                help. I could also imagine one non-contact thermometer being used
                to sweep for a person in a sort of head / rider configuration or
                multiple units being used to zero in on things in the right
                temperature range, but that could get expensive. As a bonus, the
                targeting laser(s) could really freak people out, and being class II
                they are relatively safe.




                --- In beam@yahoogroups.com, William "Chops" Westfield <westfw@...>
                wrote:
                >
                > On Feb 3, 2006, at 7:32 PM, jeffhoron wrote:
                >
                > > Objects emit infrared radiation in proportion to temperature.
                >
                > Hmm. contact-less thermometers are becoming relatively affordable,
                > and might provide a much more directional sensor than the passive
                > IR detectors like the one you reference (which are pretty much
                > designed NOT to be directional.)
                >
                > http://www.harborfreight.com/cpi/ctaf/displayitem.taf?
                Itemnumber=91778
                >
                > BillW
                >
              • darthjder
                Thanks guys! Woah, these are good ideas. Using PIR sensors in a photovore... Actually, since we re looking for humans here, we can lessen the robots confusion
                Message 7 of 7 , Feb 5, 2006
                  Thanks guys! Woah, these are good ideas.

                  Using PIR sensors in a photovore... Actually, since we're looking for
                  humans here, we can lessen the robots confusion (deciding whether to
                  target a candle or a human) by measuring sudden changes in heat rather
                  than heat itself. (The candle stays still most of the time. The human
                  being's heat moves around alot.) This also makes the robot harder to
                  fool: If a human to run past a large heat source, the robot wouldn't
                  go to the heat source, but to the running human. The basics of the
                  BEAMsonics circuit and Robinson's Hider hearing cicuit provide a good
                  example for detecting changes in readings.

                  The mention of a laser puts an idea into one's head: What if the robot
                  puts out a screen of 4 beams in front. 4 corresponding recievers check
                  if any of the laser beams begins to vary in reflection (That is, the
                  human freaks out and starts moving, which changes the reading on the
                  reciever). The robot then charges towards the human based on this
                  information.

                  A bit pricey, though. It would be cheaper to use mirrors surrounding
                  an IR LED to create the beam. The mirrors could be implemented with a
                  reflective metal tube and a reflective metal cone put on top of the
                  LED. A similar system, using two beams on each side of a robot's head,
                  could be used to create fleeing behavior for grazing robots in an RJP.
                  Or, like Jeffhoron suggests, we can mount the beam-sweep system on a
                  head! Heh.

                  All this talk about human tracking makes it seem like we should have
                  seen a human-tracking head by now... Especially from the guys who did
                  military-type work with Mark Tilden.

                  --- In beam@yahoogroups.com, "jeffhoron" <jeffhoron@...> wrote:
                  >
                  > Interesting idea. That's the lowest-cost non-contact thermometer
                  > I've seen with a decent (8:1) distance to spot (measured area size)
                  > ratio. I suppose it depends on Darthjder's intended application, so
                  > Darthjder please tell us if we're headed down the path you wanted to
                  > go. Two PIR sensors could provide left/right/straight ahead info
                  > similar to a simple photovore for $11. Maybe some blinders would
                  > help. I could also imagine one non-contact thermometer being used
                  > to sweep for a person in a sort of head / rider configuration or
                  > multiple units being used to zero in on things in the right
                  > temperature range, but that could get expensive. As a bonus, the
                  > targeting laser(s) could really freak people out, and being class II
                  > they are relatively safe.
                  >
                  >
                  >
                  >
                  > --- In beam@yahoogroups.com, William "Chops" Westfield <westfw@>
                  > wrote:
                  > >
                  > > On Feb 3, 2006, at 7:32 PM, jeffhoron wrote:
                  > >
                  > > > Objects emit infrared radiation in proportion to temperature.
                  > >
                  > > Hmm. contact-less thermometers are becoming relatively affordable,
                  > > and might provide a much more directional sensor than the passive
                  > > IR detectors like the one you reference (which are pretty much
                  > > designed NOT to be directional.)
                  > >
                  > > http://www.harborfreight.com/cpi/ctaf/displayitem.taf?
                  > Itemnumber=91778
                  > >
                  > > BillW
                  > >
                  >
                Your message has been successfully submitted and would be delivered to recipients shortly.