Key.Aero Network
Register Free

Page 2 of 4 FirstFirst 1234 LastLast
Results 31 to 60 of 92

Thread: Stealth fighter vs stealth ship

  1. #31
    Join Date
    Oct 2010
    Location
    Australia
    Posts
    2,421
    Quote Originally Posted by SpudmanWP View Post
    You're missing the point. I was responding to Rii's comment about the ship being nearly the same temp as the surrounding water. These pictures clearly show that not to be the case for EOTS and EODAS.
    I never suggested that aircraft couldn't detect and identify ships using IR sensors. What I suggested (or rather implicitly hypothesised) was that if the aircraft can identify the ship using IR, then the ship can identify the aircraft using IR also. Even if that aircraft is a wunderbar F-35.

    I'm not an expert on IR systems (or anything else, duh) but I'm aware that temperature difference between the target and background plays a significant role in their functioning, and there is a much larger temperature delta for the aircraft than for the ship, in part because the air is much colder at altitude, in part because the ship experiences negligible friction heating, and in part because the ship's primary sources of heat are embedded within and dissipated throughout a large hull. And of course the sensor on the ship can be significantly larger (and be backed by more processing) than the sensor on the aircraft. If you are aware of additional, mitigating, or countervailing factors of relevance, feel free to share. I am here to learn.
    Last edited by Rii; 23rd January 2013 at 17:28.
    "We stare into the abyss we've created and wonder why it stares back at us." -- Bernard Chazelle

  2. #32
    Join Date
    May 2012
    Posts
    643
    Quote Originally Posted by seahawk View Post
    Tracking is not searching. And making something invisible to IR sensors is not enough, we are entering the time of weapons using more than one sensor. For example matching IR and imaging (UV or normal light) is nearly trivial today.
    IRST stand for infrared search and track so i think it can automatic serching as well , btw i haven't hear about optic sensor on fighter that use UV , can u show me some caused it sound quite strange

  3. #33
    Join Date
    May 2012
    Posts
    643
    Quote Originally Posted by Belethor View Post
    Correct me if i'm wrong but isn't Pirates FOV somewhat obscured by the location its positioned on the airframe? To my eye it looks far from ideal for scanning for ground/maritime targets.

    no my point is IRST and FLIR system on aircraft automatic search and track for targets rather than having the pilots finding the target himself , it work a little bit like radar rather than just binoculars

    example : the DAS can finds the target itself rather than being cue by other sensor
    http://www.youtube.com/watch?v=IZrvAFRhQZc
    http://www.youtube.com/watch?v=qF29GBSpRF4
    Last edited by moon_light; 23rd January 2013 at 18:15.

  4. #34
    Join Date
    Dec 2010
    Location
    UK
    Posts
    1,237
    IMO, answering such a question request to take into account the climate (warm water or cold water, heavy or calm weather), location deep (sea or coastal waters), type of stealth vessel (frigates, corvettes...), electronics and sensors used, skills of the crew, type of missile used (I doubt than some older missiles can have a real chance to target a low obsersable aircraft especially if they can't benefit from constant updates from the ship).

    In the end I don't think there is a simple answer. The exocets used during the Falklands war as well as naval aviation during WWII might teach us one thing or two about the advantage of air power, but I wouldn't want to be a F35 making a run against a T45...
    “Nothing is impossible, the word itself says 'I'm possible'!”

  5. #35
    Join Date
    Dec 2010
    Location
    UK
    Posts
    1,237
    Quote Originally Posted by moon_light View Post
    no my point is IRST and FLIR system on aircraft automatic search and track for targets rather than having the pilots finding the target himself , it work a little bit like radar rather than just telescope
    So far these systems have been mostly developped for AtA applications. I'm not aware of the Typhoon being currently operable for sea assault... and tbh I doubt such a system would be useful unless coupled to the radar...

    EDIT: Also there is a big difference between tracking a SpaceX's Falcon 9 launch vehicle (quite big and bright lol) and tracking a Aster 30 or a stealth corvette...
    Last edited by Mildave; 23rd January 2013 at 18:18.
    “Nothing is impossible, the word itself says 'I'm possible'!”

  6. #36
    Join Date
    May 2012
    Posts
    643
    Quote Originally Posted by Mildave View Post
    So far these systems have been mostly developped for AtA applications. I'm not aware of the Typhoon being currently operable for sea assault... and tbh I doubt such a system would be useful unless coupled to the radar...

    EDIT: Also there is a big difference between tracking a SpaceX's Falcon 9 launch vehicle (quite big and bright lol) and tracking a Aster 30 or a stealth corvette...
    yeah there is a big difference between tracking a SpaceX's Falcon 9 launch vehicle and tracking a Aster 30 or a stealth corvette but my point is it can do it automatically , the different would be in range but if it can track Falcon 9 from 1300 km then even if the range against stealth corvette is only 10% of that it still quite impress

  7. #37
    Join Date
    Jan 2009
    Posts
    3,016
    @Jonsey:

    Computerized ID of a target is quite advance. Take a look at these SAR ID images where the APG-81 was able to AutoID tanks sized targets based off of seemingly small SAR returns. EOTS has similar capabilities.




    @Rii:

    As you can see by the previous EOTS image of a ship on the water, it has no problem with temp gradients and the ship provides a very high contrasting target when compared to the surrounding water.

    If an IRST can detect an inbound fighter at say 50km and a FLIR can detect a ship at 75nm... you do the math as to whom will survive the engagement.

    Remember that the Sensor does not have to be the Shooter when it comes to the F-35. The Sensor could stay up high and far away, flying a parallel course (seemingly non-aggressive) while the Shooter comes in at 100ft and launches a AShM from over the horizon and is never seen from the ship.
    "The early bird gets the worm but the second mouse gets the cheese."

  8. #38
    Join Date
    Dec 1999
    Location
    Reading, UK
    Posts
    4,539
    Spudman,

    Computerized ID of a target is quite advance. Take a look at these SAR ID images where the APG-81 was able to AutoID tanks sized targets based off of seemingly small SAR returns. EOTS has similar capabilities
    Whether its RF, IR or visible light spectrum its still an imaging sensor and, like all imaging sensors, has resolution as a function of range to target and is utterly dependent upon target aspect at higher resolution for an 'automatic' ID.

    Your imaging sensor also is still dependent upon something providing cueing to lay it on the correct bearing to target. Staying passive that means reliance on the other side doing something dumb. If the other side doesnt do something dumb, like transmit on something readily identifiable, your pilot is going to be just as well off with a good pair of binocs looking out the side of the canopy trying to follow wakes!. Mk1 Eyeball will be the best wide area sensor he's got!. It also puts him above the visual horizon to target.

    If an IRST can detect an inbound fighter at say 50km and a FLIR can detect a ship at 75km... you do the math as to whom will survive the engagement.
    The one that ID's the other first.

  9. #39
    Join Date
    Jan 2009
    Posts
    3,016
    Your imaging sensor also is still dependent upon something providing cuing
    EOTS is passive and does not need a cue to look at a target, it finds them on it's own (the "S" in IRST).

    If you want cuing, that can come from ESM and SAR.

    The one that ID's the other first.
    It's a virtual certainty that a F-35 will ID & launch against a ship LONG before the ship even knows the F-35 was there. You just can't get away from the 50km/150km FLIR range detection difference.

    There is a reason the planes hunt ships and not the other way around.
    "The early bird gets the worm but the second mouse gets the cheese."

  10. #40
    Join Date
    Dec 1999
    Location
    Los Angeles
    Posts
    4,505
    EOIRs are great for identifying targets at long distance, but require external cueing to do so because the field of view can be very, very narrow at maximum zoom. Too narrow to be of use when searching large areas.

    I've seen imagery taken from Edwards AFB airspace that identified airliners by type and carrier taxiing at LAX airport 100 km away. (Admittedly, it was during the summer when atmospheric moisture was a minimum).

  11. #41
    Join Date
    Jan 2009
    Posts
    3,016
    but require external cuing
    It may be nice to have and would speed up the process, but is not a requirement.

    Besides, cuing would likely come from the ESM as a ship is a very "leaky" EM environment.
    "The early bird gets the worm but the second mouse gets the cheese."

  12. #42
    Join Date
    Dec 1999
    Location
    Reading, UK
    Posts
    4,539
    Quote Originally Posted by SpudmanWP View Post
    EOTS is passive and does not need a cue to look at a target, it finds them on it's own (the "S" in IRST).
    No. It detects a thermal gradient on its own. That is all the S & T parts do. If you want to relate those to something of interest you need to know which particular contacts to close for higher resolution imagery.

    I've seen the Sniper XR demos...F16's identifiable from 15 miles or thereabouts nothing really special?. Can you provide some provenance for target ident through Sniper or EOTS at 150km?

    If you want cueing, that can come from ESM and SAR.
    ESM see earlier comment of 'doing something dumb' or deliberate. If you are reliant on the other side telling you where he is dont be surprised if he lies to you.

    It's a virtual certainty that a F-35 will ID & launch against a ship LONG before the ship even knows the F-35 was there. You just can't get away from the 50km/150km FLIR range detection difference.
    Nope. This goes...

    (1) plane tries to find ship by ESM and gets nowhere...
    (2) plane uses optronics and shuffles back and forth trying to pick out the ship he wants from those he doesnt...
    (3) plane goes back to base for more fuel and a spot of crew rest because it has no real persistence...you need a ship for that!
    (4) another plane comes out and repeats step two because all the boats have moved when no-one was watching.
    (5) the ship notices an aircraft flying a non-commercial profile on nav radar or IRSTS and evaluates as a potential hostile.

    There is a reason the planes hunt ships and not the other way around.
    There is also a reason why, even quite recently, they've not always been very good at it!.
    Last edited by Jonesy; 23rd January 2013 at 21:08.

  13. #43
    Join Date
    Oct 2010
    Posts
    329
    Quote Originally Posted by Jonesy View Post
    Sanem,
    Another photo to illustrate:
    This is a 1m resolution image of Manhattan from Ikonos. At 1m resolution you get a strip depth of 11km.
    don't get me wrong, I don't know much about camera resolution and all that. what I do know:

    a) your average UAV can read a license plate from high altitude
    b) camera resolutions are increasing greatly every year. today's resolution will be outdated tomorrow
    c) as SpudmanWP pointed out, while we humans may not be able to make much of this image, a modern computer programmed in the correct way can probably tell you every type within seconds, regardless of how many there might be in the picture. it's a simple of matter of dimensions, angles and ratios

    Quote Originally Posted by Jonesy View Post
    Adding automation of the type you are theorizing about suggests someone would be foolish enough to let a computer select targets for engagement. Not going to happen.
    I didn't suggest automated targetting, I suggested detection and giving this information to a human controller who then asks his commander who asks his boss if he can open fire... which is how humans do it today, you just automate the detection process, where most of the complexity and workload lies

    but as I mentioned before, a Chinese aircraft carrier heading for Taiwan may be filled with orphan children, and Mother Theresa's pleasure cruise may be carrying a nuclear bomb, so long range detection has its limits

    Quote Originally Posted by Jonesy View Post
    Its one thing to have a computer reject an image that doesnt match that shown in a high resolution terminal seeker FoV....its quite another to rely on a computer tell you which are the targets you can shoot at and which not.
    actually, that's kind of the whole selling point of the F-35: it can, and it will

    http://www.youtube.com/watch?v=9fm5vfGW5RY
    at 2:51

    Quote Originally Posted by Rii View Post
    What I suggested (or rather implicitly hypothesised) was that if the aircraft can identify the ship using IR, then the ship can identify the aircraft using IR also. Even if that aircraft is a wunderbar F-35.
    detecting an aircraft would be harder: detecting a fast flying little stealthy aircraft in an entire sky (that's what, billions of cubic meters?) should be a lot harder than detecting a huge, slow ship at a limited sea level (a million square meters by comparison?)

    but I do agree with your point, as sensors become better and better, they'll be able to detect even stealth aircraft faster and easier. maybe with superior power, maybe with alternative techniques and technologies like sound maybe?

    either way it is a fact that radar and IR stealth is already pretty much useless in dogfighting, with most modern jets having optical detection and engagement methods. the F-35 for example can engage an F-22 the second it gets within optical range, and while its AMRAAM missiles will have a hard time finding the Raptor, Israeli or Russian optical missiles will have no such problems, radar and IR stealth doesn't matter squat to such weapons

    btw once ships start installing combat-capable lasers this whole discussion become moot as ships will have the power and range to blast any munitions out of the sky before they can impact, except super-fast rail shots and maybe fat ICBMs

  14. #44
    Join Date
    Dec 2012
    Posts
    175
    Quote Originally Posted by djcross View Post
    EOIRs are great for identifying targets at long distance, but require external cueing to do so because the field of view can be very, very narrow at maximum zoom. Too narrow to be of use when searching large areas.

    I've seen imagery taken from Edwards AFB airspace that identified airliners by type and carrier taxiing at LAX airport 100 km away. (Admittedly, it was during the summer when atmospheric moisture was a minimum).
    Are you allowed to say which system?

  15. #45
    Join Date
    Dec 2012
    Posts
    175
    Quote Originally Posted by Sanem View Post
    actually, that's kind of the whole selling point of the F-35: it can, and it will

    http://www.youtube.com/watch?v=9fm5vfGW5RY
    at 2:51
    That's as much a product of the F-35's CNI suite as much as it is the actual veiwing sensors. It's not all down to the EODAS.

  16. #46
    Join Date
    May 2012
    Posts
    643
    Quote Originally Posted by Jonesy View Post
    No. It detects a thermal gradient on its own. That is all the S & T parts do. If you want to relate those to something of interest you need to know which particular contacts to close for higher resolution imagery.

    I've seen the Sniper XR demos...F16's identifiable from 15 miles or thereabouts nothing really special?. Can you provide some provenance for target ident through Sniper or EOTS at 150km?



    ESM see earlier comment of 'doing something dumb' or deliberate. If you are reliant on the other side telling you where he is dont be surprised if he lies to you.



    Nope. This goes...

    (1) plane tries to find ship by ESM and gets nowhere...
    (2) plane uses optronics and shuffles back and forth trying to pick out the ship he wants from those he doesnt...
    (3) plane goes back to base for more fuel and a spot of crew rest because it has no real persistence...you need a ship for that!
    (4) another plane comes out and repeats step two because all the boats have moved when no-one was watching.
    (5) the ship notices an aircraft flying a non-commercial profile on nav radar or IRSTS and evaluates as a potential hostile.



    There is also a reason why, even quite recently, they've not always been very good at it!.
    here the pic taken by ATFLIR ( almost identical to sniper-XR )from 70 km

    i think with this kind of resolution detect a normal ship would not be a problem
    not sure how good it is against ship with IR treatment
    btw Irstshow target on screen just like radar , to ID u can zoom in to see the picture of each targets by EO sensor
    Example the FSO system on rafale ( EOTS + DAS on F-35 do basiclly the same thing )

    the left is screen showing target being detected and track by IRST , on the right is the image of target being zoom in
    Last edited by moon_light; 23rd January 2013 at 21:51.

  17. #47
    Join Date
    Jan 2009
    Posts
    3,016
    Quote Originally Posted by Jonesy View Post
    I've seen the Sniper XR demos...F16's identifiable from 15 miles or thereabouts nothing really special?. Can you provide some provenance for target ident through Sniper or EOTS at 150km?
    Here is degraded photo of downtown Las Vegas from 36nm as seen by SniperXR (the predecessor of EOTS). See the size of the restaurant on top of the tower and how big it is (easily the size of a frigate). At twice the distance the tower will still be big enough to PID.



    Quote Originally Posted by Jonesy View Post
    ESM see earlier comment of 'doing something dumb' or deliberate. If you are reliant on the other side telling you where he is dont be surprised if he lies to you.
    You cannot put a ship into an anti-echoic chamber to chase down leaky EM as you can with the F-35. The ship also has to deal with a much harsher work environment than a fighter.

    Quote Originally Posted by Jonesy View Post
    There is also a reason why, even quite recently, they've not always been very good at it!.
    When would that have been?
    Last edited by SpudmanWP; 23rd January 2013 at 22:00.
    "The early bird gets the worm but the second mouse gets the cheese."

  18. #48
    Join Date
    Sep 2004
    Posts
    3,929
    Quote Originally Posted by hjelpekokk View Post
    Hmm, and here I though that it got 1 × 8-cell Mk41 VLS w/ 32 × RIM-162 ESSM!
    And that this means it got 32 surface to air missiles.
    You do know that each cell have 4 missils, so total 32 missils ;-)
    You forgot to mention that Ticonderoga CG has only 4 Raytheon/RCA AN/SPG-62 and Burke DDG only 3, so for a frigate FFG a pair is not bad.

  19. #49
    Join Date
    Jan 2009
    Posts
    3,016
    ESSM Blk2 will have an active seeker.
    "The early bird gets the worm but the second mouse gets the cheese."

  20. #50
    Join Date
    Dec 1999
    Location
    Reading, UK
    Posts
    4,539
    Sanem,

    As pointed out by Belethor:

    http://www.is.northropgrumman.com/pr...NI_datasht.pdf

    a) your average UAV can read a license plate from high altitude
    ...but with what field of view. As stated the Ikonos image was an 11km strip...this allowing for some semblance of area search...even then its not great. The point I'm making is that the stealth aircraft needs to search for the ship and it wont do so with the field of view even in the, relatively low res, ATFLIR image posted by moonlight.

    b) camera resolutions are increasing greatly every year. today's resolution will be outdated tomorrow
    We are presumably talking about today though.

    c) as SpudmanWP pointed out, while we humans may not be able to make much of this image, a modern computer programmed in the correct way can probably tell you every type within seconds, regardless of how many there might be in the picture. it's a simple of matter of dimensions, angles and ratios
    3 words. False. Alarm. Rate.

    Moonlight

    i think with this kind of resolution detect a normal ship would not be a problem
    The resolution isnt that spectacular there though...you'd likely distinguish a warship from a merchantman with that, I agree, but if thats 70km you're certainly not getting an ID at 150...plus, as mentioned, whats the FoV there....looks to be about 6 or 7km across?. Going to take you a while sweeping a search pattern looking at that narrow an image and, if the ship can see the plane at 50km, getting a reliable detect and ID at 70km is cutting it a bit fine if you happen to be looking in the wrong place as you close on the surface contacts!.

    btw Irstshow target on screen just like radar , to ID u can zoom in to see the picture of each targets by EO sensor
    Example the FSO system on rafale ( EOTS + DAS on F-35 do basiclly the same thing )
    Indeed. Hadnt seen the French system but its still the same thing IRST gives you a series of tracks within its FoV. Without the ability to tell from the IRST which one is the warship the pilot is obliged to look at each contact if doing so at extreme range i.e the 70km image you've shown thats time consuming. Go back to the AIS image and image how long its going to take to scrutinize a couple of hundred contacts...and THAT is in closed waters where its easier to contain your contacts...imagine how hard it gets in open water!.
    Last edited by Jonesy; 23rd January 2013 at 22:44.

  21. #51
    Join Date
    Jan 2009
    Posts
    3,016
    The F-35 would not suddenly appear in the middle of that CF. As contacts appear, they are classified and tracked in realtime due to the AESA scan rates. Data from ESM & off-board sources are then fused. Only then, if needed, does a PID need to happen. Once a PID is done, that contact can be either ignored or attacked.

    Take a look at those photos above. A ship sized contact at 150km is a piece of cake. The tower in Las Vegas is less than 200 feet across and an AB Class Destroyer is over 500 (2.5 times larger). To put it simply, an AB Destroyer would appear larger at 150km than the above restaurant does in the above photo (at 36nm).

    The F-35 will not need to ID hundreds of contacts at the same time. It only needs to ID them as they are detected. Don't forget that the F-35 is not alone and they will share the load & compare data.

    A contact rich environment can actually work to the F-35's advantage as it helps it to hide it's radar use.
    Last edited by SpudmanWP; 23rd January 2013 at 23:07.
    "The early bird gets the worm but the second mouse gets the cheese."

  22. #52
    Join Date
    Dec 1999
    Location
    Reading, UK
    Posts
    4,539
    Spudman

    When would that have been?
    Certainly well into the 90's that I know of personally!. Whether it was RAF Nimrods miss-ID'ing merchies as Argentine aircraft carriers, USN Outlaw Hunter TASM shots targetting USN RAS vessels accidentally, Russian Far East marpat being led a merry dance around the Sea of Japan etc, etc its all the same.

    Lots of specialised anti-ship planes with highly trained, professional, crews and they couldnt find coal in snow!. You'll forgive me if I dont think a one-man fighter with a few clever IR imagers is going to do a job!.

  23. #53
    Join Date
    Jan 2009
    Posts
    3,016
    You're going to compare 20 year old non-US, non-AESA, non-FLIR, non-ESM tech to the F-35???

    ROFL
    Last edited by SpudmanWP; 23rd January 2013 at 23:11.
    "The early bird gets the worm but the second mouse gets the cheese."

  24. #54
    Join Date
    Dec 1999
    Location
    Reading, UK
    Posts
    4,539
    Yes I am because I know how difficult it is to ID a ship at range and a silly little single seat fighter on its own...as per the question posed here...is not going to change that.

    When I see the identifiable images of LHA's at 150km I'll be impressed...tower buildings and palaces with no other similar structures nearby at half the touted distance are not that. The XR demo with the F-16's at 15 miles impressed me more owing to the resolution...half the distance ten times the resolution...as you'd expect...nothing revolutionary there!.

    Even then though its still not solving the wide area search question is it?. ESM's not the answer...going active from a stealth aircraft seems somewhat contradictory...sooo you are left with???

  25. #55
    Join Date
    Dec 1999
    Posts
    3,371
    It's not the sensor resolution that matters since it already reached beyond the critical limit of processing speed. Data storage speed, image processing speed/fast algorithms, and detector sensitivity (noise) are the key parameters. Take a relatively common experience to give you a hint. A fast high end DSLR still cannot shoot 30fps at full resolutions. Just the speed to store the images, let alone processing, is quite lacking. Same reason is why until recently it's 1080P even for large TVs of over 60 inches. You can argue these are "military grade", which in electronics term means even slower processors. The system may be more robust, secure, fault tolerant, and higher sensitivity. I recall a NATO tech book indicate you only need 3 pixels to ID an object as tank and 5 pixels to know type, iirc! It's not an image that a human can recognize.
    Country::US of A

  26. #56
    Join Date
    May 2012
    Posts
    643
    Quote Originally Posted by Jonesy View Post
    Sanem,

    As pointed out by Belethor:

    http://www.is.northropgrumman.com/pr...NI_datasht.pdf


    ...but with what field of view. As stated the Ikonos image was an 11km strip...this allowing for some semblance of area search...even then its not great. The point I'm making is that the stealth aircraft needs to search for the ship and it wont do so with the field of view even in the, relatively low res, ATFLIR image posted by moonlight.


    We are presumably talking about today though.


    3 words. False. Alarm. Rate.

    Moonlight



    The resolution isnt that spectacular there though...you'd likely distinguish a warship from a merchantman with that, I agree, but if thats 70km you're certainly not getting an ID at 150...plus, as mentioned, whats the FoV there....looks to be about 6 or 7km across?. Going to take you a while sweeping a search pattern looking at that narrow an image and, if the ship can see the plane at 50km, getting a reliable detect and ID at 70km is cutting it a bit fine if you happen to be looking in the wrong place as you close on the surface contacts!.



    Indeed. Hadnt seen the French system but its still the same thing IRST gives you a series of tracks within its FoV. Without the ability to tell from the IRST which one is the warship the pilot is obliged to look at each contact if doing so at extreme range i.e the 70km image you've shown thats time consuming. Go back to the AIS image and image how long its going to take to scrutinize a couple of hundred contacts...and THAT is in closed waters where its easier to contain your contacts...imagine how hard it gets in open water!.
    quite good info
    btw i have just read sth quite the same here (but about irst against fighter )

    http://www.network54.com/Forum/21183...82/Raptor+demo.

    btw i just think of an idea that may help aircraft detect the ship first EX: the aircraft may carry 1-2 decoys like MALD and launch it , then the aircraft itself fly down below radar horizon, as the MALD can fly for 920 km it can be program to fly around an area if there is any stealth ship detect the decoy and attack it by missiles ,their bearing will be detected when the missiles rised from horizon ( as the missile much hotter detect them must be much easier) and the aircraft only have to zoom it's FLIR to that bearing to find the ship
    what do u think ?
    ( i think the fighter hunting Scud missiles launched did use the same tactic)

  27. #57
    Join Date
    Feb 2010
    Posts
    73
    Quote Originally Posted by SpudmanWP View Post
    You're going to compare 20 year old non-US, non-AESA, non-FLIR, non-ESM tech to the F-35???

    ROFL
    He has a point, though. Unless you are implying the F-35 has the same ship detection capabilities than a specialized platform like a P-8 or other MPA's...

  28. #58
    Join Date
    Jan 2009
    Posts
    3,016
    I did say an asset from 20+ years ago, without FLIR, without AESA, etc....

    Obviously a modern dedicated SAW platform will likely have better sensors, processing power, and larger FLIR.
    "The early bird gets the worm but the second mouse gets the cheese."

  29. #59
    Join Date
    Oct 2010
    Location
    Australia
    Posts
    2,421
    Quote Originally Posted by Vortex View Post
    I recall a NATO tech book indicate you only need 3 pixels to ID an object as tank and 5 pixels to know type, iirc! It's not an image that a human can recognize.
    I find this difficult to believe. Object recognition is still, I believe, one of the major stumbling blocks for autonomous systems. Computers are rubbish at it compared to humans, who have evolved under the necessity of distinguishing a tiger amidst the concealing grass. Both air and (calm) sea environments are very favourable conditions for object recognition, not only because of the relatively uniform background but also because of the limited range of objects that need to be identified. Identifying a tank amidst varied terrain and terrain features, amidst an assortment of similarly sized/shaped objects, etc. is an order of magnitude more difficult.
    Last edited by Rii; 24th January 2013 at 02:39.
    "We stare into the abyss we've created and wonder why it stares back at us." -- Bernard Chazelle

  30. #60
    Join Date
    Dec 1999
    Posts
    3,371
    Ah... but signature isn't just limited to the outline of an object... that's too human in thinking. Furthermore, there's the concept of a single pixel can paint an image. Of course they can be fooled, hence Jonesy's basic premise of man in the loop before weapons away. This does answer why cueing is an issue but it's not pixel size/count limited.
    Country::US of A

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

- Part of the    Network -

KEY AERO AVIATION NEWS

MAGAZINES

AVIATION FORUM

SHOP

 

WEBSITES