Charts II: NGA - 2700 Charts

[LeaseOnLife]

Originally Posted by Netsurfer

I am doing this completely by PHP scripts. One for the download on my server and the other for stitching the tiles on my local system. Stitching 3.000 tiles together won't take a minute.
So imho there is no need to use a certain script for this task if you ensure to always use the highest quality level.

Which ones are you downloading & stitching?

Originally Posted by Netsurfer

Are you talking about any future changes to the chart source images or what's the point for "a revision control system"?

Both: for the time after downloading and stitching, but we may see updates to the original files too, from NGA!

[cagney]

It was a bit silly to say that just the python script would do. The important thing is to download and stitch together the "raw" jpg's, without further treatment, to start with.

I'll try to get the python script going on the server, and will then start going through all charts that has not been claimed.

Thomas

[cagney]

I'm working on the whole of region 2, Central & South America Caribbean etc.

[Netsurfer]

Originally Posted by LeaseOnLife

Which ones are you downloading & stitching?

I am working on Region 3 starting from 35000.

Both: for the time after downloading and stitching, but we may see updates to the original files too, from NGA!

Ah OK. But wouldn't it be sufficient for such cases if we would add a 'rev-date-in-ISO-format' to the original filename (being just the chart number)? We all know that our original charts are from July 2011.


Gunther

[bcn]

I noticed that imgkap will put the current date as issuing date of the chart.

That might be ok with a privately created chart, but in our case we should perhaps maintain the NGA issuing information. That can manually done by patching the kap header information in the chart, but this is for sure not the safest way.
What do you opinion?

Naming conventions:
- for the header using Cagney's format my header files are named 52nnnhed.txt
- Stiched jpgs have just the nga number: 52nnn.jpg
- The charts carry the name from NGA (no blanks, no accents, etc.)

19 out 34 charts from 52000 to 52143 are currently downloaded and stitched.
Waiting for an upload place. No hurry.

Hubert

[Netsurfer]

Originally Posted by bcn

I noticed that imgkap will put the current date as issuing date of the chart.

Are you talking about the 'CED/ED' parameter?
Which format did you use for the date?

That might be ok with a privately created chart, but in our case we should perhaps maintain the NGA issuing information. That can manually done by patching the kap header information in the chart, but this is for sure not the safest way.
What do you opinion?

As I already set up a DB with all available chart data (including edition and edition date) I could offer two options:

1. I can make a website where you enter the chart number (or a range) and you'll get the basic header(s) as txt file
2. I can create all headers and you download your desired ones

Naming conventions:
- for the header using Cagney's format my header files are named 52nnnhed.txt
- Stiched jpgs have just the nga number: 52nnn.jpg
- The charts carry the name from NGA (no blanks, no accents, etc.)

We should use nothing but only the chart number. All the rest is logical given by the file type.
Anything else in the file names will only complicate things.

All the rest like searching a chart by name could be easily achieved by appropriate scripts running on a website.

Gunther

[svrevelations]

Originally Posted by Netsurfer

...........
My proposal is to use the existing 'opencpn.xxx' websites for this (and any other site you like). ............. Thanks Gunther

Where will the completed .kap files be available for downloading?

[r.fairman]

As promised here is a kap file of bodrum harbour with 4 rough calibration points and outline defined by mapcal
Header file was created from chartcal.dir using mc2bsbh renamed as txt and the img2kap used to create kap file using -n 127 switch

https://docs.google.com/leaf?id=0B3S...0ZTEw&hl=en_US

Here is the header from the kap file

! 2011 imgkap 1.11 - at 10/07/2011 from BODRUM.png
! Created by mc2bsbh beta07 - Use at your own risk!
VER/2.0
BSB/NA=Bodrum Harbour,NU=1000
RA=7286,5763,DU=179
KNP/SC=10000,GD=WGS84,PR=MERCATOR,PP=UNKNOWN
PI=UNKNOWN,SP=UNKNOWN,SK=0.0,TA=90.0
UN=METERS,SD=UNKNOWN
DX=1.42,DY=1.42
REF/1,6977,359,37.0356667,27.454
REF/2,252,5465,36.97,27.35
REF/3,167,278,37.0367083,27.3486867
REF/4,6979,5468,36.9699617,27.45403
PLY/1,37.0402176,27.3461815
PLY/2,36.9662301,27.3461802
PLY/3,36.9662304,27.4587002
PLY/4,37.0402172,27.4587013
DTM/0,0
OST/1
IFM/7

Works well in OpenCPN Seaclear Kapview and Photoshop Fugawi BSB Viewer

please test in any other software you may have

regards

Richard

[Meark56]

Hello everyone !

I'm new and working on region 5. A big one !
In first excuse me for my poor english
I joins me to you because I think it would be ridiculous that another person do the same work .

Finish east atlantic :51002 to 52143, and 57000 to 57487
just downloaded, no convert, no rotate

Nothing done for mediteran.sea and black sea :52160 to 56222.

Now, i go away sailing for 2 weeks. When i'll come back, i restart the job !

question : I have some problem with 51261, 52047 and 57183.
dezoomify is ok, but impossible to download

[Netsurfer]

Originally Posted by MooseMoney

Where will the completed .kap files be available for downloading?

I think we have to differ between

1. the physical location of the files
2. download links

The answer to 1. is "on your server".
And for the latter "everywhere (on very website) you'd like". What I meant is that imho we do not need to also setup any website on your server as it would be sufficient if the links are propagated on the already existing 'opencpn.xxx' websites.
I hope this makes things clearer?

Thanks
Gunther

[HappySeagull]

Originally Posted by r.fairman

As promised here is a kap file of bodrum harbour with 4 rough calibration points and outline defined by mapcal
....etc

..........
please test in any other software you may have

regards

Richard

Hi Richard...


for whatever reason,the lat/long does not line up...see my mark in screen grab?on a chart that purports to be WGS 84...?It says it is in the chart title .(I'm well aware that older charts' lines of lat/long are NOT WGS 84 and need shifting to agree)

Otherwise looks fine....

so an issue...frankly,I make it very clear in created kaps that these are home-made....I use EX in the name,or similar un-missable....

also,(per BCN's post above) why mess with mdj's imagekap creation date?(see http://www.cruisersforum.com/forums/f134/imgkap-kap-file-converter-tool-56600-9.html#post720088)
because these are created from pictures stitched together after all...

[LeaseOnLife]

Originally Posted by r.fairman

As promised here is a kap file of bodrum harbour with 4 rough calibration points and outline defined by mapcal
Header file was created from chartcal.dir using mc2bsbh renamed as txt and the img2kap used to create kap file using -n 127 switch

https://docs.google.com/leaf?id=0B3S...0ZTEw&hl=en_US

Here is the header from the kap file

Richard:

• works in Fugawi Marine ENC 4.5 via the non-MapTech-BSB import, trying to import via the Maptech BSB route fails, like any other chart I have produced. I didi not check the calibration like HappySeagull. BTW the pilot charts don't load in Fugawi any way I tried.
• does not work in Raytech 6.2. But so don't the pilot charts....

[LeaseOnLife]

Can we have a knowledge base / Wiki for the needed header info, per application?

Maybe something for Netsurfer:

before crafting all the headers, we should understand the needs of applications, other than just OpenCPN.

Examples:

• Fugawi needs the NU filled, and more than two Cal points, maybe it needs 4.
• I have not been able to make the Raymarine Raytech program display any of the homemade bsb charts and I probably need help evaluating the special needs of this application. Raytech can overlay the radar returns over a BSB chart, I really would like this feature, using the NGA charts.
• Polarview => brak should be able to tell us?

[NavPak ECS]

<< I see IFM 7 quite often >>

Yes, me too. That is because when the BSB spec was lifted from a patent application and leaked into the public arena, it did not come with all the technical papers necessary to implement it correctly.

[r.fairman]

Thanks to all about the test kap charts i uploaded. I agree calibration is off and I have now corrected. I too cannot get the chart to load in explorer or raymarine
Have just retested with Polar Navy (Polar View + Polar COM) and the recalibrated kap chart loads and looks great, BUT the chart appears skewed by about 2-3 degrees any ideas Brak

here is the bodrum chart 54412.kap

https://docs.google.com/leaf?id=0B3S...0Yzg3&hl=en_US

regards

Richard