Charts II: NGA - 2700 Charts

[cagney]

Originally Posted by nohal

...[*]We will loose some QA options as for the panels we can't be sure that the calibration points are in the corners and sometimes there will even be just 2 or 3 of them. That's why I want to keep insets separated from the base charts a bit more than absolutely necessary, not to loose options to check the quality for as many charts as possible automatically.

...
Pavel

Pavel
Thanks for all the hard work!

There might be a way to only have the corners as ref points, irrespective of shape, when we are all done.
This include an intermediary step where other ref pts are used temporarily. As long as we document the pixels for each corner, we can later, after the PLYs are defined, get the lat&long for the corners and use the corners as REFs.
I'm not sure that we will benefit from this extra complication though.

Thomas

[cagney]

Originally Posted by bcn

54362 (and a lot of other Greek insets)
no geo references at al - sketches with depth, but not lat/Lon....

Eyeball navigation

Hubert

They are still possible to do manually. Do a first, best guess, for lat & long, based on the main or best available chart, GE or other sources. Adjust for "best fit".
Much more work, for sure.

Thomas

[bcn]

Another example for things we have to be able to correct:

54423 Antalya:
the inset is "Approaches to Antalya" and has a scale of 1:75000 (not 15.000 as from the database). The main map is a 1:15000 one

Hubert

[cagney]

Originally Posted by bcn

Another example for things we have to be able to correct:

54423 Antalya:
the inset is "Approaches to Antalya" and has a scale of 1:75000 (not 15.000 as from the database). The main map is a 1:15000 one

Hubert

You can do that right now. Use the "pencil" in the lower left corner to edit selected row.
Thomas

[bcn]

Ok, this solves as well the "Guadalquivir" case with the 3 panels named "X"

Thanks for the tip!

Hubert

[bcn]

Pavel,

will you provide a kind of statistics for the insets like for the main charts? That makes it a lot easier to search for "work to do".

Hubert

[nohal]

Hubert...
Yes, of course, inset statistics are on the to-do list.

Pavel

[wdkester]

I'm assuming that we draw the box to the outside border to include lat/lon marks etc

[wdkester]

Anybody want to describe the difference between Inset, Plan, and Panel. This is a pull down menu for adding a line to the Identification block

[z.stadler]

Originally Posted by wdkester

I'm assuming that we draw the box to the outside border to include lat/lon marks etc

Yes. Moreover, including additional area will not harh and can even be safer for skewed charts.
For charts composed of non-overlapping panels I started from relevant external corners. See chart 25607 for example.

Zeev

[nohal]

Originally Posted by wdkester

Anybody want to describe the difference between Inset, Plan, and Panel. This is a pull down menu for adding a line to the Identification block

+1
As far as I was trying to decipher it, it seems to me that at NGA and Maptech their software has a similar dropdown and their employees must have similar questions which nobody answers...
In the NGA catalog, almost all the insets are marked as PLAN, followed by INSET and just a few are PANEL. I'm really not sure whether I would call a plan something at 1:500000, but maybe for them the main key is that the plan is any more detailed chart of an object also visible on the base chart?
I don't think it's very important though - the only place where this value is used is the .BSB file listing the charts originating from the same paper source (and not used by any chartplotter I've ever seen)

Pavel

[nohal]

Some preliminary QA stats for the insets: NGA Charts project stats

Pavel

[z.stadler]

Originally Posted by nohal

Some preliminary QA stats for the insets: NGA Charts project stats

Pavel

Excellent!

Could you do us a small favor and add hyperlinks to the appropriate entries - http://opencpn.info/en/nga-charts-status-<Region>?page=<Page>#<Chart>

Separating the regions into pages reduces the wait for the region to complete loading, but it complicated the ability to get to the right page for a given chart.

Zeev

[bcn]

Originally Posted by z.stadler

Yes. Moreover, including additional area will not harh and can even be safer for skewed charts.
For charts composed of non-overlapping panels I started from relevant external corners. See chart 25607 for example.

Zeev

And have a lookout for extrusions as well. Give additional margin for those.

Hubert

[bcn]

Originally Posted by nohal

Some preliminary QA stats for the insets: NGA Charts project stats

Pavel

Pavel,

that helps!
Perhaps some numbers about the insets defined/delimited?

Hubert