Charts II: NGA - 2700 Charts

[paulr]

Originally Posted by MooseMoney

I'm sure many users would like to help but feel frustrated with this thread. There are no step-by-step instructions for non script experts. Barring the people who knows what is going on - for the rest of us, this thread is a confusing mess. Unless somebody is going to take the time to post a detailed step-by-step procedure, you "experts" will have to do all the work by yourselves. Help me to help you is what I say.

MooseMoney

I sympathise. If you are talking about the download part of the operation, getting a download-and-stitch script working is not as easy as it should be, and given that there are still 600 charts required, I expect your help for this task would still be very welcome.

If you are running Windows XP, I'll gladly supply simple instructions to get you up and running with fewest complications. Just post back here to confirm. I expect others could chip in for other operating systems.

But as Pavel says, the real work, which is going to need some very careful and intensive human input and quality control (in fact some of the qualities that professional cartographers have), is still some way off.

Paul

[Surveyor]

Originally Posted by nohal

Moose...
We are already preparing the recipe in our kitchen, but you guys have to be a bit patient. We have to calibrate 2700 charts, no doubt there will be more than enough work for everybody interested
Right now the bodycount is 2109, so we have to download and stitch about 600 charts to have the raw data for the whole set. Then we have to burn quite a few CPU cycles to transform them to a form anybody will be able to calibrate, finish the needed tools and scripts and write the instructions. All this needs it's time.

Pavel

2109! I have been wondering where we stood and I am pleasantly surprised that we are getting so close. I should make us all T-shirts with "I survived the great download of 2700!"

[svrevelations]

Originally Posted by paulr

MooseMoney

I sympathise. If you are talking about the download part of the operation, getting a download-and-stitch script working is not as easy as it should be, and given that there are still 600 charts required, I expect your help for this task would still be very welcome.

If you are running Windows XP, I'll gladly supply simple instructions to get you up and running with fewest complications. Just post back here to confirm. I expect others could chip in for other operating systems.

But as Pavel says, the real work, which is going to need some very careful and intensive human input and quality control (in fact some of the qualities that professional cartographers have), is still some way off.

Paul

Hi Paul - I have Windows 7

[OErjan]

Hi, Örjan from Sweden here, I am new here and just joined to say thanks to all of you.

Oh and just a bit of a heads up, the chart 44260 is not 100% accurate, I have checked against my paper charts covering the region (live there), several shoals and some points of land are not exactly where they should be, error is not large - within 0.20' - but could lead to trouble if no paper chart is used or area is unfamiliar...

[Surveyor]

Originally Posted by OErjan

Hi, Örjan from Sweden here, I am new here and just joined to say thanks to all of you.

Oh and just a bit of a heads up, the chart 44260 is not 100% accurate, I have checked against my paper charts covering the region (live there), several shoals and some points of land are not exactly where they should be, error is not large - within 0.20' - but could lead to trouble if no paper chart is used or area is unfamiliar...

Orjan hello!

This is a topic I have been thinking much about lately. I have a feeling that these charts, without the NTMs annotated, could serve to perpetuate already known errors.

I am not sure if this project will mature to the point that NTMs or other known errors will one day be shown, (or come with an OCPN layer file showing corrections ) but I am sure it will be discussed.

That being said, all charts are erroneous the moment they are published. In 99.9% of all cases, charts are created from data that was collected in the past. As with any navigational situation, total reliance upon charts, paper or otherwise, is foolish. Local knowledge is what keeps you off the rocks and a wide berth together with day light keeps those not so familiar with the area safe.

I propose that all of these charts be given a watermark statement such as "not for navigation" or "for planning purposes only". Personally, I would only use them for planning or open water navigation where I know there is not a hazard ahead. Not all sailors are prudent though and it might be wise to take that fact in to consideration.

One day, after the war and greed machines are defeated, everyone will sail with a simple device that will capture satellite data that can be rendered on a handheld and shows everything including accurate mapping and current weather dynamically! HAA! (the idea is not that far fetched)

[LeaseOnLife]

Originally Posted by OErjan

Hi, Örjan from Sweden here, I am new here and just joined to say thanks to all of you.

Oh and just a bit of a heads up, the chart 44260 is not 100% accurate, I have checked against my paper charts covering the region (live there), several shoals and some points of land are not exactly where they should be, error is not large - within 0.20' - but could lead to trouble if no paper chart is used or area is unfamiliar...

Hi Örjan

• what scale are the charts you comparing? NGA 44260 is 1:225000
• could you scan a section of an area of concern and post it here? Would be interesting if these anomalies are in deep water or outside shipping areas. I doubt the NTM would have all the changes but we should check.

Dirk

[OErjan]

sorry have no scanner
the paper chart I compared with is scale 1:250 000 (what I hat that was roughly equivalent), Swedish chart 42, also did quick checks using the 1:50 000 charts of same area, Swedish charts where somewhat better, not perfect when compared to the 1:50k but closer than 44260

[cagney]

NGA chart 44260 is based on Finnish and Swedish Charts to 1977 and corrected to 1996. Every chart has a link to the applicable NTM correction. Here is the link to 44260: 44260 corrections.

The NGA coverage for Scandinavia is of limited use for small craft navigators doing inshore navigation and piloting, as it seems aimed at ships doing transits and offshore navigation with the odd larger harbor plan available.

It is interesting to note that the Swedish "Hydrographica" is one of the few commercial outfits hat still produces BSB3 charts, and has excellent detailed charts for many Swedish areas. They protect their charts differently than encrypted BSB4 charts. The bottom line is that they can still be used with OpenCPN.

Thomas

[OErjan]

Cagney; if these charts are based on 1977 charts that will certainly explain the errors, we had a extensive revision of charts covering the area in question few years back using new data acquired using modern techniques (GPS, advanced 3D sonar...)

[r.fairman]

All ntm's since 1996 are listed up to july 2011. you could apply these to the original image file (using photoshop)to bring your chart upto date then recalibrate and create kap file.

regards Richard

[r.fairman]

This is the sound of a can of worms opening

All the 2700 charts have links to the most uptp date NTM's Applying these to the charts is a straightforward process (but time consuming). As its impossible to apply the NTM's to the charts post processing (convert to kap) is there a case to apply them before processing to ensure they are up to date. This would put our chart resource way ahead of any others including the NGA. Our charts would be really worldbeating and unquestionably free to use. We would be adding real value to the original charts. In some cases there are up to 20 NTMS to apply and in other cases only a few NTMs to apply The only penalty is time. Perhaps there is a case for continuing the convert to kap with running in paralell the application of NTMs. an alternative might be to include a html link in OpenCPN when a chart is opened to its NTM for display.(as a layer) I regard the NTM's stored by NGA as a resource we should mine. I am not sure what Marco did with his NZ charts

regards

Richard Fairman

[Surveyor]

So what is going on? We are getting near the end of the download/upload process and it seems that communication has come to a halt here and on github.

[dougwm10]

Congrats to everyone who is involved and working on the download and processing of these charts.

In my opinion it would be very shortsighted if the NTM.s were not applied to the charts before the processing is finalised and the *.kap charts released for general use.

I fully appreciate and personally would willingly accept whatever the delay would be in order to complete this task so that we end up with a fully updated set of charts that can be used with OpenCPN.

It would be helpful to all OpenCPN users if whoever is organising and controlling the chart acquisition process could post regular (monthly or weekly) updates on the current position with regards to this project and possibly a time line to completion and availability for downloading the finished charts.

Best wishes to all involved

Originally Posted by r.fairman

This is the sound of a can of worms opening

All the 2700 charts have links to the most uptp date NTM's Applying these to the charts is a straightforward process (but time consuming). As its impossible to apply the NTM's to the charts post processing (convert to kap) is there a case to apply them before processing to ensure they are up to date. This would put our chart resource way ahead of any others including the NGA. Our charts would be really worldbeating and unquestionably free to use. We would be adding real value to the original charts. In some cases there are up to 20 NTMS to apply and in other cases only a few NTMs to apply The only penalty is time. Perhaps there is a case for continuing the convert to kap with running in paralell the application of NTMs. an alternative might be to include a html link in OpenCPN when a chart is opened to its NTM for display.(as a layer) I regard the NTM's stored by NGA as a resource we should mine. I am not sure what Marco did with his NZ charts

regards

Richard Fairman

[nohal]

Originally Posted by Surveyor

So what is going on? We are getting near the end of the download/upload process and it seems that communication has come to a halt here and on github.

Jim,
the count is currently 2584, so we are really approaching the completion of the download, but there's still some 5% left to do.
The web calibration process is worked on, but not yet ready to use, as Gunther is a bit busy with OpenCPN 2.5 release which takes place "just now" as you've sure seen.
Once we are ready with the download, I will do the crunching of the charts on the server, and we can start to calibrate the rectangular charts and collect the needed info. This will come in a couple of days I would say.
Then we have work for months (or years if we decide to apply the NTMs)

Pavel

[Surveyor]

Originally Posted by nohal

Jim,
the count is currently 2584, so we are really approaching the completion of the download, but there's still some 5% left to do.
The web calibration process is worked on, but not yet ready to use, as Gunther is a bit busy with OpenCPN 2.5 release which takes place "just now" as you've sure seen.
Once we are ready with the download, I will do the crunching of the charts on the server, and we can start to calibrate the rectangular charts and collect the needed info. This will come in a couple of days I would say.
Then we have work for months (or years if we decide to apply the NTMs)

Pavel

Excellent info, thank you. If anyone needs a hand grabbing and stitching please let me know. That was kinda what I was getting at, sorry if I came across any other way.