MH370 KUL-PEK Missing 8 Mar 2014: Search & Recovery [PLEASE SEE WIKI]

Subscribe
Thanks!

I put that through Google translate and didn't get much info out of it. Looks like there are resolutions of 2m, 8m, and perhaps 16m. No band information.

However, it looks pretty much like the Chinese version of Landsat or Spot.
Reply
Quote: Or try
http://de.wikipedia.org/wiki/Gaofen_1 (in German)
Much better. That entry has a further link to http://www.nasaspaceflight.com/2013/...h-2d-gaofen-1/, which could be useful in the Wiki/FAQ.

Unfortunately, the high-res sensor is a CCD panchromatic device, which is probably going to be hard to use to find anything, as I suspect it's only RGB. The multi-spectral scanner, which could be useful, has a resolution of 8m, which means that anything it would detect would only show up as a few pixels.
Reply
Quote: Thanks!

I put that through Google translate and didn't get much info out of it. Looks like there are resolutions of 2m, 8m, and perhaps 16m. No band information.

However, it looks pretty much like the Chinese version of Landsat or Spot.
Its resolution is 2m. The image color spectrum includes red, yellow, green and blue bands. 4 days are needed for taking the shots of entire globe.
Reply
Could this latest Chinese sighting possibly be the same piece that the Australians picked up and has drifted to another location?
Reply
Quote: Could this latest Chinese sighting possibly be the same piece that the Australians picked up and has drifted to another location?
That sounds like the plausible theory right now from the comparisons that are being made on the news.
Reply
Quote: Unfortunately, the high-res sensor is a CCD panchromatic device, which is probably going to be hard to use to find anything, as I suspect it's only RGB.
Not quite correct. A panchromatic sensor is a single channel detector sensitive to EM radiation (i.e., light) over a broad range of wavelengths (usually corresponding to the visible part of the spectrum). It produces a grayscale image. For an 8-bit data product (typical for satellite imaging systems), each pixel has 256 possible shades of gray (0=black, 255=white).

Quote: The multi-spectral scanner, which could be useful, has a resolution of 8m, which means that anything it would detect would only show up as a few pixels.
True. But imaging systems with pan and multispectral sensors offer the potential for pan-sharpening, i.e., fusing the two data products and essentially using the panchromatic image to "sharpen" the resolution of the spectral bands.
Reply
Quote: The multi-spectral scanner, which could be useful, has a resolution of 8m, which means that anything it would detect would only show up as a few pixels.
If they are using that then hopefully a cluster of ~10 pixels will be enough to be 'unusual' - in the sense of being a relatively large number where you don't expect many at all.

Maybe they already used that to speed up the search of the visible spectrum, which I assume the released images to be? Are both arrays used at same time (as in within a small number of minutes)?
Reply
Quote: Could this latest Chinese sighting possibly be the same piece that the Australians picked up and has drifted to another location?
CNN - some reporter type from Perth just stated that this object (75 miles away) could not be the same object; that is, could not have drifted that far in 48 hours. I'd like to hear more opinions on that possibility and I am sure we will.
Reply
Quote: If they are using that then hopefully a cluster of ~10 pixels will be enough to be 'unusual' - in the sense of being a relatively large number where you don't expect many at all.

Maybe they already used that to speed up the search of the visible spectrum, which I assume the released images to be? Are both arrays used at same time (as in within a small number of minutes)?

Due to the atmospheric scattering in the blue and green bands of visible, they'll be nearly useless for detection. It's better to split the bands out and work with red. Better yet to detect the infrared and do false color. With this the water will absorb most of the red and IR and anything metal will reflect back dramatically bright.

It looks like the Chinese plan on launching a satellite with radar capability, which would be pretty useful.

During my university days I did a bit of satellite image processing on a project for the US Forest Service.
Reply
Latest media release from the AMSA:

Quote:
23rd March, 2014: 0001 (AEDT)

Search to resume on Sunday for flight MH370: Update 10

The Australian Maritime Safety Authority search operation for Malaysia Airlines flight MH370 has
concluded for today.

During Saturday’s search activities a civil aircraft tasked by AMSA reported sighting a number of small
objects with the naked eye, including a wooden pallet, within a radius of five kilometres.

A Royal New Zealand Air Force (RNZAF) P3 Orion aircraft with specialist electro-optic observation
equipment was diverted to the location, arriving after the first aircraft left but only reported sighting
clumps of seaweed.

The RNZAF Orion dropped a datum marker buoy to track the movement of the material. A merchant
ship in the area has been tasked to relocate and seek to identify the material.

The search area experienced good weather conditions on Saturday with visibility of around 10 kilometres
and moderate seas.

The Royal Australian Air Force, the Royal New Zealand Air Force, two chartered civil aircraft and two
merchant ships supported Saturday’s search effort in a 36,000 square kilometre search area in the
Australian Search and Rescue Region.

Since AMSA assumed coordination of the search on Monday 17 March, 15 sorties have been flown and
more than 150 hours of air time has been committed by the air crews to the task.

Four military aircraft assisted in today’s search, as well as two ultra-long range jets. Ten State
Emergency Service (SES) volunteers from Western Australia were tasked as air observers today, along
with two AMSA mission coordinators on the civilian aircraft. AMSA runs a training program across the
country to train SES volunteers in air observation for land and sea searches.

The Royal Australian Navy’s HMAS Success has arrived in the search area. Two merchant ships are
also in the search area.

The search will resume tomorrow and further attempts will be made to establish whether the objects
sighted are related to MH370.

This evening China provided a satellite image to Australia possibly showing a 22.5 metre floating object
in the southern Indian Ocean. AMSA has plotted the position and it falls within Saturday’s search area.
The object was not sighted on Saturday.

AMSA will take this information into account in tomorrow’s search plans.

Enquiries: [email protected]

Media kit and latest charts: http://www.amsa.gov.au/media/incidents/mh370-search.asp
Reply
Quote: Not quite correct. A panchromatic sensor is a single channel detector sensitive to EM radiation (i.e., light) over a broad range of wavelengths (usually corresponding to the visible part of the spectrum). It produces a grayscale image. For an 8-bit data product (typical for satellite imaging systems), each pixel has 256 possible shades of gray (0=black, 255=white).
Unfortunately, that makes it even worse. While the visible spectrum makes it nice for us humans to look at things, trying to detect something in the water will be made more difficult because of the blue and green in the mix.


Quote: True. But imaging systems with pan and multispectral sensors offer the potential for pan-sharpening, i.e., fusing the two data products and essentially using the panchromatic image to "sharpen" the resolution of the spectral bands.
The way we did things in the project I mentioned in a previous post was to use clustering to bring out the likely reflectivity of various land cover we were trying to identify.

I'd imagine the same sort of process could be used in what you mention. Not so much to 'sharpen' the resolution of the bands, but to assign a specific land cover type to a specific cluster.

Now, with a view over the ocean, there is only two land covers to try to find: Water, and not water. Water will absorb almost all IR light, so almost anything that shows up in the IR with high reflectivity will be a floating object. Then again, a cresting wave may show up with a good deal of reflectivity as well.

In any case, I think it's a difficult task.
Reply
Quote: Could this latest Chinese sighting possibly be the same piece that the Australians picked up and has drifted to another location?
I think it's very likely. It's just over 1,600 miles from Perth, really close to the 1,500 mi figure given by Australia.
Reply
Looks like in addition to the Chinese satellite sighting, a civilian aircraft also saw some small debris in the AMSA search area on Saturday

"During Saturday’s search activities a civil aircraft tasked by AMSA reported sighting a number of small objects with the naked eye, including a wooden pallet, within a radius of five kilometres. "

AMSA map released 22 March showing location of debris sited by Chinese satellites in relation to these objects sited Saturday by an AMSA flight

Map:
https://s3-ap-southeast-2.amazonaws....ed+chinese.pdf

AMSA press release

http://www.amsa.gov.au/media/documen...ate10MH370.pdf
Reply
MH 370 KUL-PEK Missing: Search operations ongoing [PLEASE SEE WIKI]
FWIW, Sky News in the UK is running a decent (factual not speculative) 'special report' on the missing plane. They pulled in some apparently credible people including a senior ex 777 BA pilot and an Oceanographer. The Oceanographer reckoned there's no way the Chinese satellite debris could be the same as what Australians spotted. He seems absolutely certain it could not have travelled that far in the time frame.
Reply
Quote: FWIW, Sky News in the UK is running a decent (factual not speculative) 'special report' on the missing plane. They pulled in some apparently credible people including a senior ex 777 BA pilot and an Oceanographer. The Oceanographer reckoned there's no way the Chinese satellite debris could be the same as what Australians spotted. He seems absolutely certain it could not have travelled that far in the time frame.
2 days, 200 km... needs ~ 1m/s... backwards against the prevailiing IO circulation. I'd tend to agree.
Reply