What It Would Really Take To Sink a modern Aircraft Carrier

[Taj_91]

image recognition, not carrier detection is used for PRISM. In order for PRISM to run in automation, you have to have a way to ID a target by image recognition.

i.e. if someone send someone a photo over the internet and that someone is being monitored by PRISM, you will need to know who was in that photo to be flagged, to be autonomous, they will need a image recognition software suite to match the photo that someone send to the terrorist database, you cannot ask that someone to take a photo at the same distant and the same angle as with the one you had in the database, so a geometrical match is impossible, you must have image recognition.

Seeing the current level of Google's advancement, I think it would have failed hopelessly!
It is one thing to differentiate between cargo ship and carrier, even if it detects other carriers there is less than 20 of them. It is quite another thing to differentiate between human beings autonomously having minute differences in facial features, that too for such an important task as terrorist identification!

 

[randomradio]

@Taj_91 @jhungary @gambit

http://www.ece.neu.edu/fac-ece/elmiller/nuwiirl/pdfs/tip_templates_02.pdf
No landmarks, only the object against a cluttered background. Even the dimensions and orientation are unknown.

https://cseweb.ucsd.edu/classes/wi08/cse190-a/reports/scheung.pdf
Image recognition of the make and model of cars from arbitrary angles.

Once a carrier is picked up by a satellite, autonomous identification and tracking are easy. Extremely easy. Satellites have picked up and identified aircraft on their own, what's so special about a carrier?
https://www.irjet.net/archives/V3/i10/IRJET-V3I1012.pdf

Image recognition has long surpassed the human eye.
https://www.theguardian.com/global/...omputer-better-than-humans-recognising-images

LOL! With the amount of ships surrounding the carrier, it would take more than 24 to try to sink the carrier. And haven't even mentioned the SM-6 that can intercept during terminal phase besides the SM-3 which intercepts mid course. And in the near future there is possible BMD ship that can handle larger salvos of anti ship ballistic missiles. So imagine building about a dozen for the carrier groups.

Ballistic missiles are large enough to carry MaRVs or even MIRVs.

24 BMs could end up releasing well over 50 warheads. Maybe even 100+.

You should. Or rather -- MUST.

Whatever is your sensor, radar or photo, in order to distinguish the unique outlines of container transport vs aircraft carrier, your sensor is sensitive enough to distinguish out the details that are on the surface of the ships, even in a 2D view. But then, this is why SAR capability is enabled on the next generation of radars installing in the non-5th gen fighters.

But if you chose not to distinguish out those unique shapes of each type of ships, then you can and will be fooled.

Carrier aircraft on deck actually give away the carrier. Carrier aircraft, even their individual models, the bridge, the unique silhouette of the ship, they all give away the carrier. And there's actually nothing you can do to prevent the carrier's detection except probably blind the camera using a laser or blow up the satellite.

@jhungary
The Indian Army has specifically asked for new stealth technologies to counter the advances made in radar and optical sensors.
http://www.telegiz.com/articles/213...ouflage-makes-fighting-vehicles-invisible.htm
The army is asking industry to develop an "electronic camouflage solution" that enables AFVs to blend with their surroundings. This solution will also change the image of the hull in concert with the changing environment.

Another but more fantastic solution the army wants developed is "Quantum Stealth," a system being developed by a Canadian company where a special material makes a vehicle invisible by bending the light waves around the vehicle. The special material also suppresses infrared and thermal signatures, and also the shadow of the vehicle.

 

[The Accountant]

LOL! With the amount of ships surrounding the carrier, it would take more than 24 to try to sink the carrier. And haven't even mentioned the SM-6 that can intercept during terminal phase besides the SM-3 which intercepts mid course. And in the near future there is possible BMD ship that can handle larger salvos of anti ship ballistic missiles. So imagine building about a dozen for the carrier groups.

Actually the DF21's job is just as difficult if it cannot detect its target. You depend on satellites to detect the ships. What if they are destroyed or jammed? AWACs or drones won't help because they be detected long before getting even anywhere near the proximity of the carrier.

Agreed but when you are considering other platforms as well then kindly do consider that Chinese are on thier mainland and have lot more options than yours ... They can literally shower stand-off smart bombs, cruise missiles and torpedos to weaken the defenses one by one ... Remember we are assuming USA is an aggressive force so what we are talking about is max five battle groups, ICBMs of US and long-range bombers against Whole Chinese force ... Please don't bring in Japan and South Korea otherwise it will bring north Korea, Russia, and Iran as well so lets consider this is one on one USA vs China excluding Nukes ...

China has more than 700 fourth generation advance and more than thousand three generation interception so it is a huge force to deal with and flying from mainland means they can have much higher sortie rates ,, the supply line is much faster and virtually unlimited in comparison to CBG ....

So I would have been military planner then the game is simple for me ... I will take out air defenses of CBG group one by one bu using stand off weapons and will push US to consume as much ammunitions as it can using F7s and other third generation aircraft whereas support to F7s will be provided from powerful AWACs and radars ...

Only challenge will be from F22s and F35s and it yet to Proved Chinese capability of tracking down F22s ...

With regards to taking down Chinese satellites ,,, I think you should realize that Chinese also have same capabilities infact if you are moving in this direction than it is more harmful to US then China as a foreign attacking force US has much more dependency on satellites ,,, China can still rely on ground radars and AWACs but if it comes to destroying satellites than US will be almost blind ...

 

[Oldman1]

If a submarine can stay at a depth of more than a 100 feet and sneak up on a ACC then all it takes are small torpedoes which can be hard for cat systems to detect.

How small are we talking about?

Agreed but when you are considering other platforms as well then kindly do consider that Chinese are on thier mainland and have lot more options than yours ... They can literally shower stand-off smart bombs, cruise missiles and torpedos to weaken the defenses one by one ... Remember we are assuming USA is an aggressive force so what we are talking about is max five battle groups, ICBMs of US and long-range bombers against Whole Chinese force ... Please don't bring in Japan and South Korea otherwise it will bring north Korea, Russia, and Iran as well so lets consider this is one on one USA vs China excluding Nukes ...

You seem to have forgotten that the U.S. also has long range missiles and other stand off weapons to weaken a country even like China. Remember, the U.S. built their force to take on the Soviet Union in trying to destroy their defenses or penetrate it. Just like you saw in Gulf War 1. Japan and South Korea are important because of bases nearby that can also have affect on China's military.

China has more than 700 fourth generation advance and more than thousand three generation interception so it is a huge force to deal with and flying from mainland means they can have much higher sortie rates ,, the supply line is much faster and virtually unlimited in comparison to CBG ....

Umm thats nothing. We have thousands of 4th gen aircraft besides hundreds of 5th gen. And you seem to think that their airbases would be left intact.

So I would have been military planner then the game is simple for me ... I will take out air defenses of CBG group one by one bu using stand off weapons and will push US to consume as much ammunitions as it can using F7s and other third generation aircraft whereas support to F7s will be provided from powerful AWACs and radars ...

If I was a military planner, I would ambush the ships with submarines and mines. Also use cruise missiles and smart decoys to take out their ships to force them to expend ammo before aircraft carriers send their aircraft in.

Only challenge will be from F22s and F35s and it yet to Proved Chinese capability of tracking down F22s ...

With regards to taking down Chinese satellites ,,, I think you should realize that Chinese also have same capabilities infact if you are moving in this direction than it is more harmful to US then China as a foreign attacking force US has much more dependency on satellites ,,, China can still rely on ground radars and AWACs but if it comes to destroying satellites than US will be almost blind ...

Wouldn't worry too much about the tracking the F-22s or the F-35s. On destruction of satellites, actually this is more harmful for China because they depend on it to track the carriers for their ballistic missiles to take out the carriers. Without it, the AWACs or ground radars are useless because they will not be able to get near the carriers.

 

[jhungary]

Seeing the current level of Google's advancement, I think it would have failed hopelessly!
It is one thing to differentiate between cargo ship and carrier, even if it detects other carriers there is less than 20 of them. It is quite another thing to differentiate between human beings autonomously having minute differences in facial features, that too for such an important task as terrorist identification!

Actually, it's a lot easier to detect human face than a set object.

Human face is not smooth, and it's quite unique, your jawline and my jawline is most definitely different, and hence, a lot of unique identifier can be pull from each of the human face, where there are some unique feature of my face, and some unique feature of your face. If we uses vector matrix, we can build quite good vector matrix mapping of one's face, because the unique identifier is different in each face with regard to the different part of the face (like a mole to the side of the cheek or freckled to the side of the eyes) On the other hand, a smooth surface such as ship hull, offer no "State Change", meaning it's no use to break apart an image to analyse it simultaneously (The bridge offer the same scheme, reflection than the part of the hull, unlike Human nose offer more reflection than human jaws) Hence making identifier of smooth object harder than rough surface.

 

[Oldman1]

=


Ballistic missiles are large enough to carry MaRVs or even MIRVs.

24 BMs could end up releasing well over 50 warheads. Maybe even 100+.

Well if they increase the number of warheads, can easily respond with this. These are little bit bigger than a coffee thermos, can be stacked or packed together.

 

[jhungary]

@Taj_91 @jhungary @gambit

http://www.ece.neu.edu/fac-ece/elmiller/nuwiirl/pdfs/tip_templates_02.pdf
No landmarks, only the object against a cluttered background. Even the dimensions and orientation are unknown.

This is a paper identifying the problem associated to the image extraction and rescale, it does not actually offer a solution. Hence this is a theory, and it may not applies in the following circumstance

Read the last paragraph, extract as follow

In this work, we have examined the problem of finding a target in a noisy image. Following in the work of Abu-Naser [1], were present the problem as an image restoration problem where the object to be reconstructed is a delta function encoding the target location and the blur is a target template. Previous work showed that the restoration can be performed using a LLSE and choosing the location of maximum response [1]. However, this did not consider the problem of unknown geometric parameters, i.e., the size and rotation of the target within the image

and

This approach was not amenable to minimization,

however, due to local minima and regions of zero derivative on the likelihood surface. To overcome these difficulties, we present a new method of generating a library of target templates which range from smooth monomodal approximations to the exact target. Using this library of templates, it is possible to sequentially estimate the parameter set using standard optimization tools.

Hence it basically saying this theory may be able to put in the multi-target function, it mostly talking about how to transpose an object in term for non-geometric identification, this can already be done by today technology.

https://cseweb.ucsd.edu/classes/wi08/cse190-a/reports/scheung.pdf
Image recognition of the make and model of cars from arbitrary angles.

Same thing, this is a theory running SIFT Feature and Harris Feature, notice the image they compare are "UNIQUE" meaning a car can be of similar make but have unique marker which can draw with inline extraction (Draw two Unique Identifier (UI) into a line of matrix.)

No where it does works with non-unique object (without UI, such as same car but different branding), or dynamic object (where UI keep changing).

notice;

At 5 : Result

Implementing SIFT interest point detection with the SIFT interest point matcher was a natural choice due to its invariance over image translation, scaling, and rotation. SIFT interest points are also known to be highly distinctive and easy to extract. However, for the hundreds of interest points detected by SIFT per image that were passed to the SIFT interest point matcher, only at most approximately 30 pairs of matched interest points were found.

Denoted that interest point matching only provide 30% accuracy (30 out of more than 100 interest point). And;

A remaining problem consists of the location of matched pairs. As shown in Fig. 9, the final set of matched pairs are all located on a single restricted area of the car, even when features are detected in all regions of the car. Though we have proposed a technique to implement a database that will have the ability to accomplish MMR given an image of a car from an arbitrary angle of view, we have not reached that point of the project given the duration of the quarter.

Denoted that matching probability is small and limited.

Once a carrier is picked up by a satellite, autonomous identification and tracking are easy. Extremely easy. Satellites have picked up and identified aircraft on their own, what's so special about a carrier?
https://www.irjet.net/archives/V3/i10/IRJET-V3I1012.pdf

A. The detection of an aircraft used by such alrogithm is not 100% (not even 90% match) depending on the algorithm used and the sample supplied, it reaches from 53% to 93%.

B.) An aircraft is not as dynamic as an aircraft carrier.

As I said before.

THis, and This,

Is the same Aircraft Carrier, but the aircraft on deck will make too much "noise" to the image and would give you a negative reading. While in an aircraft, the "noise" itself is not as great as the body of the aircraft will not distorted within flight.

Image recognition has long surpassed the human eye.
https://www.theguardian.com/global/...omputer-better-than-humans-recognising-images

Again, this is a geometry recognition. We are talking about Dynamic and Non-Geometric

They system can do that because, for example, a Siamese Cat have different Unique Identifier than say, A British Short Hair.

British short hair cat (Above) have;

Wide and fat ears,
Fat Cheek
Wide Eyes
Triangular Nose,
Blunt Ear tips
Short Fore and Hind Legs
Short Tail

Siamese, on the other hand have

Double Fold ears
Close Eyes gap
Sharp face
Round nose
Trianglar Mouth
Long Hind legs and Normal Front legs
Long Tails.

The machine can pick up the different because it have gave a lot of unique identifier on the image, and it can access the UI and draw a matrix table, even some of the feature on either cat would also feature in other cat breed as well (such as Russian Blue) but if only a subject process all the trait (cat breed all have the same trait), then that sample could only be of particular breed. In the end, the only things the Computer will need to do is to detect the UI (Such as Ears and Nose in the cat) and you can put the matrix up and compare each subject.

On the other hand. Detecting a Carrier in the middle of the ocean is another problem, carrier is not a living thing, nor was it multi-dimensional polygon, meaning it will be very hard to find UI to build a matrix, in term of the computer, a Carrier is roughly a noisy rectangle.

Consider this

The Black Line is the Carrier Abraham Lincoln profile as appear on a computer, and this

The black line denoted the Profile Liaoning.

Now tell me how many UI you can pick up from the image?

If you solely consider the Black Line (The profile) before scaling, I would say the Profile is 80-90% similar.

@jhungary
The Indian Army has specifically asked for new stealth technologies to counter the advances made in radar and optical sensors.
http://www.telegiz.com/articles/213...ouflage-makes-fighting-vehicles-invisible.htm
The army is asking industry to develop an "electronic camouflage solution" that enables AFVs to blend with their surroundings. This solution will also change the image of the hull in concert with the changing environment.

Another but more fantastic solution the army wants developed is "Quantum Stealth," a system being developed by a Canadian company where a special material makes a vehicle invisible by bending the light waves around the vehicle. The special material also suppresses infrared and thermal signatures, and also the shadow of the vehicle.

NO idea why you quote me on this.

Quantum stealth is another issue not really related to Image Recognition. It uses UV Spectrum to make the item reflex the same UV/EM Spectrum as the surrounding environment, in essence, it is the basic form of protection from FLIR. Which the Taliban have mastered it 10 years ago.

 

[Penguin]

To sink ships in WW-2 most effective way was to hit torpedoes on one side of ships so water can enter from one side and capsize ship quickly and it is still effective against any ship if its defenses can be breached.

Edit.

This is how huge ships capsize and sink with torpedo hit.

USs Wasp (CV7) after being torpedoed. 1942 (sunk)

Story: https://en.wikipedia.org/wiki/USS_Wasp_(CV-7)#Loss

USS Franklin (CV13) after being bombed 1945. (Survived)

Story: https://en.wikipedia.org/wiki/USS_Franklin_(CV-13)#19_March_1945

USS Hornet (CV8), last carrier ever to be sunk by enemy fire.
In a 15-minute period, hit by three bombs, two torpedoes, and two kamikaze plane strikes. Later on strike once more by 1 torpedo. Then ordered sunk by own admiral. During American atempts to scuttle her, she absorbed nine torpedoes, many of which failed to explode, and more than 400 5-inch (130 mm) rounds from USN destroyers. Two Japanese destroyers finally finished her off with four 24-inch (610 mm) Long Lance torpedoes.

Story https://en.wikipedia.org/wiki/USS_Hornet_(CV-8)#Battle_of_the_Santa_Cruz_Islands

https://en.wikipedia.org/wiki/List_of_sunken_aircraft_carriers
https://en.wikipedia.org/wiki/List_of_aircraft_carriers_by_country
Using these two pages you could calculate loss rates.


http://ww2live.com/en/content/world...ft-carriers-sunken-world-war-ii-12-sad-images
https://www.quora.com/How-many-US-a...y-been-lost-since-as-a-separate-number-please
http://ffhiker.tripod.com/index-7.html

While clearly not unsinkable, (modern) aircraft carriers are VERY resilient. Design features as well as damage control and firefighting measures and practises are key to survival. See also post WW2 incidents on USS Oriskany, Forrestal and Enterprise.

 

[The Accountant]

Go

USs Wasp (CV7) after being torpedoed. 1942 (sunk)

Story: https://en.wikipedia.org/wiki/USS_Wasp_(CV-7)#Loss

USS Franklin (CV13) after being bombed 1945. (Survived)

Story: https://en.wikipedia.org/wiki/USS_Franklin_(CV-13)#19_March_1945

USS Hornet (CV8), last carrier ever to be sunk by enemy fire.
In a 15-minute period, hit by three bombs, two torpedoes, and two kamikaze plane strikes. Later on strike once more by 1 torpedo. Then ordered sunk by own admiral. During American atempts to scuttle her, she absorbed nine torpedoes, many of which failed to explode, and more than 400 5-inch (130 mm) rounds from USN destroyers. Two Japanese destroyers finally finished her off with four 24-inch (610 mm) Long Lance torpedoes.

Story https://en.wikipedia.org/wiki/USS_Hornet_(CV-8)#Battle_of_the_Santa_Cruz_Islands

https://en.wikipedia.org/wiki/List_of_sunken_aircraft_carriers
https://en.wikipedia.org/wiki/List_of_aircraft_carriers_by_country
Using these two pages you could calculate loss rates.


http://ww2live.com/en/content/world...ft-carriers-sunken-world-war-ii-12-sad-images
https://www.quora.com/How-many-US-a...y-been-lost-since-as-a-separate-number-please
http://ffhiker.tripod.com/index-7.html

While clearly not unsinkable, (modern) aircraft carriers are VERY resilient. Design features as well as damage control and firefighting measures and practises are key to survival. See also post WW2 incidents on USS Oriskany, Forrestal and Enterprise.

Good to see you on this thread... I have always seen your post as the most balanced and sound ... Can you please share a little bit more about your opinion specially on the options discussed such as Tarpedos, antiship ballistic missiles, saturation attacks etc or a combination of smart glide bombs, cruise missiles and ballistic missiles ...

 

[randomradio]

This is a paper identifying the problem associated to the image extraction and rescale, it does not actually offer a solution. Hence this is a theory, and it may not applies in the following circumstance

Read the last paragraph, extract as follow



and



Hence it basically saying this theory may be able to put in the multi-target function, it mostly talking about how to transpose an object in term for non-geometric identification, this can already be done by today technology.



Same thing, this is a theory running SIFT Feature and Harris Feature, notice the image they compare are "UNIQUE" meaning a car can be of similar make but have unique marker which can draw with inline extraction (Draw two Unique Identifier (UI) into a line of matrix.)

No where it does works with non-unique object (without UI, such as same car but different branding), or dynamic object (where UI keep changing).

notice;

At 5 : Result



Denoted that interest point matching only provide 30% accuracy (30 out of more than 100 interest point). And;



Denoted that matching probability is small and limited.



A. The detection of an aircraft used by such alrogithm is not 100% (not even 90% match) depending on the algorithm used and the sample supplied, it reaches from 53% to 93%.

B.) An aircraft is not as dynamic as an aircraft carrier.

As I said before.

View attachment 393090

THis, and This,

View attachment 393091

Is the same Aircraft Carrier, but the aircraft on deck will make too much "noise" to the image and would give you a negative reading. While in an aircraft, the "noise" itself is not as great as the body of the aircraft will not distorted within flight.



Again, this is a geometry recognition. We are talking about Dynamic and Non-Geometric

They system can do that because, for example, a Siamese Cat have different Unique Identifier than say, A British Short Hair.

View attachment 393093

British short hair cat (Above) have;

Wide and fat ears,
Fat Cheek
Wide Eyes
Triangular Nose,
Blunt Ear tips
Short Fore and Hind Legs
Short Tail

View attachment 393094

Siamese, on the other hand have

Double Fold ears
Close Eyes gap
Sharp face
Round nose
Trianglar Mouth
Long Hind legs and Normal Front legs
Long Tails.

The machine can pick up the different because it have gave a lot of unique identifier on the image, and it can access the UI and draw a matrix table, even some of the feature on either cat would also feature in other cat breed as well (such as Russian Blue) but if only a subject process all the trait (cat breed all have the same trait), then that sample could only be of particular breed. In the end, the only things the Computer will need to do is to detect the UI (Such as Ears and Nose in the cat) and you can put the matrix up and compare each subject.

On the other hand. Detecting a Carrier in the middle of the ocean is another problem, carrier is not a living thing, nor was it multi-dimensional polygon, meaning it will be very hard to find UI to build a matrix, in term of the computer, a Carrier is roughly a noisy rectangle.

Consider this

View attachment 393095

The Black Line is the Carrier Abraham Lincoln profile as appear on a computer, and this

View attachment 393096

The black line denoted the Profile Liaoning.

Now tell me how many UI you can pick up from the image?

If you solely consider the Black Line (The profile) before scaling, I would say the Profile is 80-90% similar.

That first link was for detection of objects in clutter. Minimization is not important when you are looking at a carrier.

The Liaoning is much smaller than the Nimitz, nearly 30m, that's plenty. Then the bridge is much bigger, even that's enough. Switching the location of aircraft does nothing. The image of a single aircraft can give away the carrier. No different from the cup in the grass.

As for that link which could differentiate between the make and model of different cars. In some cars, only the door handle is different, or the windshield design is different, that can give it enough clues. The link made it simpler to understand. A carrier is not a uniform homogeneous body at all.

http://www.eyedea.cz/make-and-model-recognition/

Anyway, this is from 2001.
http://calhoun.nps.edu/bitstream/handle/10945/2762/01Mar_Alves_Redacted.pdf?sequence=3

Small ship detection using low resolution pictures.
http://elib.dlr.de/82654/1/isprsarchives-XL-1-W1-233-2013.pdf

Autonomous ship detection using satellite images is extremely old. Only you believe it is not possible.

 

[jhungary]

That first link was for detection of objects in clutter. Minimization is not important when you are looking at a carrier.

The Liaoning is much smaller than the Nimitz, nearly 30m, that's plenty. Then the bridge is much bigger, even that's enough. Switching the location of aircraft does nothing. The image of a single aircraft can give away the carrier. No different from the cup in the grass.

As for that link which could differentiate between the make and model of different cars. In some cars, only the door handle is different, or the windshield design is different, that can give it enough clues. The link made it simpler to understand. A carrier is not a uniform homogeneous body at all.

First of all, Miniaturisation is always important, you cannot compare the matrix if you cannot miniaturize or magnify the image, unless you limited the image input.

Second of all, the Liaoning is smaller than the Nimitz, yes, but how do the computer knows when they only can see shapes without miniaturisation?

Thirdly, about cars, as long as there are 1 UI, you can identify the car between smallest detail, given if you can capture that UI, the problem is can you capture them. I have worked with Traffic cam ID software before

http://www.eyedea.cz/make-and-model-recognition/

lol, you do know how this program works, right? It does not recognize car, instead it recognize partial information (most likeky from the License Plate) from the image. You can see how the program works if you put in this photo in try me section

If you put it in the system, select automatic read number plate, then select rear view, then select A,D,C.........TK in the license plate option, it will give you.

Car (1 point)
VW (1 point)
Red (1 Point)

However, is the car red? In fact, the color is probably the most easiest part to get right by pixel matrix (each pixel have a different color code encoded in an image, something like FF00FF.

The problem now is how this system make this fundamental mistake? That can possibly mean only one thing, which the program is taught to read the license plate or something unique, then search the history (such as DSLV or some sort of motoring site) and find the detail that way.

It is further proven if you put an Australian car (With Australian Number plate) like this one

And it returned nothing.

You see how this work?

Anyway, this is from 2001.
http://calhoun.nps.edu/bitstream/handle/10945/2762/01Mar_Alves_Redacted.pdf?sequence=3

This is about Automatic Target Recognition with invariant position, rotation and scale, as per the thesis pointed out

In any pattern recognition application, it is important to select features that adequately and uniquely describe the objects to be recognized. Moreover, the features associated with an object should be invariant with respect to the position, rotation, and scale of that object in the field of view. Thus the ideal recognition system is robust to orientation variations, scale variations and boundary perturbations

In order word, the scale, position and rotation must be constant, which mean it is a geometrical match. Not a non-geometric match or dynamic match

Small ship detection using low resolution pictures.
http://elib.dlr.de/82654/1/isprsarchives-XL-1-W1-233-2013.pdf

Did you actually read the title and the article? It said NEAR REAL TIME. IN the article, it explained as 1 hours from target to end user

, enabling near real-time applications, e.g. one hour from image acquisition to end user

The Chinese have already achieve 30 minutes, well, they claim anyway. Do remember, I did not claim satellite tracking is not possible, but it's not possible to do in real time.

Plus, this is the same as the car thing you post yesterday, the system use "state differential" as a method to "recognize", this system tell you what is a ship (By it state and shape) but will not tell you what ship it is and belong to whom.

Autonomous ship detection using satellite images is extremely old. Only you believe it is not possible.

I am sorry but none of your prove show anything to be able to do automatic ship detection. And again, if you can find one that really do say they can detect target in real time, not after an hour or so calculation, let me know, otherwise, just because you said that does not make it true, I have beem very well inform with Top Secret Access, and I am telling you I am not aware any system in the current market can do that, may be you are working on one, I don't know, but I do not aware something like that is in the pipeline, and you are welcome to show that I am wrong, because I can not possibly know the world wide development of this technology in any minute.

 

[Taj_91]

Actually, it's a lot easier to detect human face than a set object.

Human face is not smooth, and it's quite unique, your jawline and my jawline is most definitely different, and hence, a lot of unique identifier can be pull from each of the human face, where there are some unique feature of my face, and some unique feature of your face. If we uses vector matrix, we can build quite good vector matrix mapping of one's face, because the unique identifier is different in each face with regard to the different part of the face (like a mole to the side of the cheek or freckled to the side of the eyes) On the other hand, a smooth surface such as ship hull, offer no "State Change", meaning it's no use to break apart an image to analyse it simultaneously (The bridge offer the same scheme, reflection than the part of the hull, unlike Human nose offer more reflection than human jaws) Hence making identifier of smooth object harder than rough surface.

No, I am talking about using one software to differentiate between objects as varying as human and carriers! Then, even in that special features for each object, for humans, jawlines, eye position, nose, etc. For carriers, their silhouette, aircrafts on deck, wake, etc.
You said PRISM has been used for both the purposes. Quite difficult!
As for differentiating between Liaoning and Abe, what is the need to do that autonomously? In the entire world, there is less than 20 carriers, so if an algo calls human for max 20 times for verification, I would say it is an awesome algo!

 

[gambit]

@gambit I am taking the shape of ships into account, but not the cluster at the top. BTW, I was just trying to show that detection is not an impossible task, can be done. Uncle Sam himself might be doing it, lol!

It is not that difficult to visually confuse shapes and outlines of ships.

Low light. Fog. Rain. Anything that can obstruct your view.

If the sensor is radar, that is how ANY body is seen by the radar computer -- as a cluster of voltage spikes.

If you want to distinguish out finder details, as in trying to tell a civilian tanker apart from an aircraft carrier, you will need increasingly complex pulse characteristics. It also helps if the environment is clear of other signals that may contaminate your radar, and that the weather is cooperative and give you clear sky.

So if all you want is shapes, you might just end up shooting at a civilian ship.

Carrier aircraft on deck actually give away the carrier. Carrier aircraft, even their individual models, the bridge, the unique silhouette of the ship, they all give away the carrier. And there's actually nothing you can do to prevent the carrier's detection except probably blind the camera using a laser or blow up the satellite.

See above. I wonder if you actually been to sea and/or to the air to know what you are talking about.

 

[Penguin]

Go

Good to see you on this thread... I have always seen your post as the most balanced and sound ... Can you please share a little bit more about your opinion specially on the options discussed such as Tarpedos, antiship ballistic missiles, saturation attacks etc or a combination of smart glide bombs, cruise missiles and ballistic missiles ...

See https://defence.pk/pdf/threads/what...-modern-aircraft-carrier.491386/#post-9414435

I doubt there is any one weapon that particularly suitable. Ship sinkings are a function of many factors, of which that damage done by one weapon or another is just one. The USN managed to save hit carriers due to better damage control and fire fighting capabilities, when similarly hurt Imperial Japanese navy carriers were lost. Then again, all it may take is one unlucky hit (see e.g. HMS Hood, HMS Barham, CVL 23 USS Princeton, DD Porter).

https://en.wikipedia.org/wiki/USS_Princeton_(CVL-23)#Loss
https://en.wikipedia.org/wiki/USS_William_D._Porter_(DD-579)#Battle_of_Okinawa

 

[Gomig-21]

If we put aside any form of aerial attack for a minute, the worst fear for a carrier group is really submarines. I think a lot of people are missing the fact that there are 2 scenarios, really, for the only way to get close enough to hit a carrier with that most effective weapon. One is much easier and is during peacetime maneuvers. This has happened several times. The Russians have crept up withing torpedo distance of several US aircraft carriers in the 70's and even near the US coast.

This is the pic from 1974 that the Russians released many years later claiming that was the USS Nimitz off the United States east coast.

The Chinese have done it as recently as 2006 off of Japan when one of their subs surfaced about 10 miles from the USS Kitty Hawk. But this was all done in peacetime or cold war periods. In peacetime, naval vessels of all types have unrestricted freedom of navigation at seas. Their only restrictions are to do so safely and avoid collisions, or COLREGS. Therefore, any submarine is free to sneak up on an aircraft carrier as long as it's done safely. In 1984, the Russians got so close they violated CORLEGS when one of their Victor subs actually collided with the USS Kitty Hawk in the waters off the Korean peninsula.

The problem is that during peacetime, alertness and defensive measures are not operating at their fullest. And with free, maritime rules and only CORLEGS observed, it's much easier. That's really the best and only time to hit a carrier but that will be like repeating Pearl Harbor. Then you have to factor hitting the carrier enough times to overwhelm its ability to seal damaged bulkheads before the submarine is spotted and retaliated against. People seem to ignore that essential fact for sinking today's aircraft carriers. These are not similarly built to the WWII carriers. Some of those were supply ships, destroyers and tankers that were modified with wooden decks to turn them into aircraft carriers. Often time they easily exploded after being hit because they weren't designed to safest standards. Jet fuel and munitions were all over the place and well in reach of damaged areas above and below the waterline. Bad analogy.

Shift to a war time scenario and it's a completely different story. Once there is imminent and impending war, the carriers deploy their disappearing strategy. They don't just lurk off the coast and wait for an attack. They basically try to vanish and go out to sea and lurk somewhere within an area of 250,000 - 300,000 square nautical miles and are constantly moving, not still. Then with all defensive measures at full alert, it becomes a practical impossibility to attack a US aircraft carrier. It becomes much more difficult for any submarine to approach within any firing distance against all deployed countermeasures, between one or several US SSN's on the lookout for underwater threats to many aircraft patrolling the area for cover and support and the ships own defensive systems. During wartime, it's almost impossible.

There is no current surface to ship or air to ship missile that can reach and impact a US carrier, let alone 'sink it'. Sorry, it just doesn't exist and there is no way to penetrate the defenses. Submarines are still the only real threat and even their task to sink a modern, US carrier during wartime is close to impossible.