Thursday, June 17, 2010

CTCOG Aerial photos

Below is a map showing the 2 foot aerial photos that have been processed and delivered to CTCOG so far.   We’ll be creating County-wide mosaics (JPEG 2000 format) of the information as it becomes available.  


Once completed, the mosaics will be free to download at the following FTP site:


We have not received any of the 1 foot photos of the urban corridor between Copperas Cove and Temple yet, but here is a sample of the 2 foot resolution, showing Mill Pond Park in San Saba.


Tuesday, March 16, 2010

Bad directions? Set 'em straight!

How many times have you tried to give someone directions to your house or a meeting by sending them a link to Google Maps, MapQuest, or a similar service? How often do you find the maps produced by those websites are, ahem, less than accurate?

It turns out that most GPS and online mapping services pull their data from two services - TeleAtlas and NavTeq. They each have reporting tools for locals to help them provide the most current map data.

TeleAtlas Map Feedback

NavTeq Map Reporter

Who knows your area better than you?  Start sharing your knowledge!

Friday, February 12, 2010

Data request

I'm looking for GPS data for trails in our area.  Specifically in the parks operated by the Corps of Engineers or local state parks.  The data will be used to produce updated maps for a website we recently resurrected:


Send them to my email in .shp or .gpx format. Any tracklogs from Bell, Coryell, Lampasas, Hamilton, Mills and San Saba County are appreciated.

Thanks!

Tuesday, February 9, 2010

Processing Aerial Imagery for use in GIS applications

Steve and Jason from CTCOG drove down to San Antonio last week to visit the firms who are collecting and processing our next set of aerial imagery.

Williams-Stackhouse Incorporated has been providing photogrammetric mapping and related services since 1960, and have produced imagery in our region in the past.  Their staff walked us around their offices, explained every step of the collection process, and patiently answered our questions.  The process from planning to final delivery is more complex than I would have guessed, but the attention to detail is what makes the final product so useful.

First, Williams-Stackhouse sits down with the aviation company to plan a series of flights that will cover the entire area, such that the images overlap.  This information is plotted into the pilot's GPS navigation system.




Once the route is planned, the pilot feeds the shot data, airspeed, and altitude into a custom-built controller that times the shutter release on the camera.  One of the pilots is showing Steve the display that tracks the flight path and allows him to control the camera.

Here's the camera, minus the film cartridge.  The mount is automatically gyro-stabilized in three dimensions, so any small corrections to yaw, pitch and roll don't cause distortion in the image.  For example, if the plane is flying in a crosswind, the pilot will adjust the rudder so that the nose is pointing upwind but the flight path remains straight.  The mount will swivel the camera opposite the direction that the plane is pointing, keeping the film square to the planned flight path.  However, the mount has a limit to how much it can correct, so the pilots only fly in the best possible weather.  


Weather is a consideration not just for a stable flight path, but also for ground visibility.  Clouds between the ground and the camera will obscure the features of the terrain.  High-altitude clouds can cast undesirable shadows beneath the plane. 

Once the flight is complete, the film negative is developed and sent to Williams-Stackhouse for processing. 


Steve is looking at the negative on a light table.  Even at this size, the detail is impressive.

The next step is scanning the film into a digital format.  This scanner automatically scans and advances the roll of film, scanning each frame down to a resolution of 2 microns.  


That means that each pixel in the scanned digital image corresponds to an distance of 0.0000787 (7.87 × 10-5) inches on the film. 

The images are then adjusted for color, contrast, and brightness on the workstation to the left. 
You can see on the monitor how the image is tiled as the scanner processes each section of the image.



The next step is adjusting for variations in elevation.  The shutter is timed so that there is at least 60% overlap between images, which means we can use stereoscopic glasses to perceive the terrain in thee dimensions (very similar to the glasses you wear at a 3D movie).  Here's one of the technicians showing Steve how he can make adjustments to the images in 3D.


This is the 3D mouse used to process the images so that elevation differences can be calculated.


After several more steps that include controlling color variations, and determining where the edges of the separate images will be cut, another technician will mosaic the tiled images together. Here's a shot of the Temple High School football field being adjusted in Photoshop.


 If there are any smears or distortions that let you see where one image ends and another begins, the problem area is reprocessed and the mosaic is performed again.

Once the mosaic is complete, the image is stored on an external hard drive and shipped to the client.  Based on the number of images, and the resolution and file size, we expect to receive about 2 Terabytes of data when the project is complete.



The aerial acquisition for the entire 7 county region at 2 foot resolution, and the urban corridor between Copperas Cove and Temple at 1 foot resolution, is complete at this time.  They are currently processing the photos, and expect to have their final product delivered to us sometime in June.

Thanks again to Randy Holt, Dave Calkins, and all the staff at Williams-Stackhouse and Krawietz Aerial Photography for an enjoyable and informative visit!

Friday, January 29, 2010

CTCOG Aerial Photos

As of December 31st, approximately 70% of the aerial photos had been captured for the region. Weather has once again held up the progress, but they should be able to continue pretty quickly on the remainder of the project. I’ve attached a map showing their progress.



Click to view larger image.

Wednesday, December 30, 2009

Plotting Points from a Spreadsheet

In a previous post, we worked through importing points from a GPS receiver into Excel and converting them to a format that we can easily plot in ArcMap. The next step is getting those points into ArcMap. Most of the time we can save the spreadsheet as a text file, usually a comma-separated volume (.CSV). However, I’ve run into problems doing that ever since Office 2007 came out. Sometimes it works, sometimes it doesn’t. It boils down to field names and cell formatting. I don’t pretend to understand exactly what 2007 does, because it doesn’t do it to me all the time. (Anyone who knows, feel free to comment or email me and I'll be glad to post your explanation). You can still save as a text file and then edit the field names (which seems to help) but that adds another step to your workflow. In previous versions of Excel, you could also save directly to a database file (.dbf), but it’s not an option in 2007. However, there’s a workaround: Using ArcCatalog to import the spreadsheet data into a table, saved as filename.dbf. This will put it into a .dbf format that ArcMap will play nicely with. By the way, I've included a lot of screenshots with this post, but don't be intimidated: the process is relatively quick once you get the hang of it.

First, continuing from the previous example, let’s save our changes. Now, let’s open ArcCatalog and navigate to the folder where our spreadsheet is saved.

Click for larger image

NOTE: For some reason, if you only have the one file in the folder, ArcCatalog doesn’t show it. If you associate the file type [xlsx] in ArcCatalog, it opens with Excel when you double-click it. We don’t want that to happen, so save the file as a 97-2003 Workbook and remove the file association. Now we’re in business!

Click for larger image

Double-click the file, and choose the sheet containing your data – it should look something like Sheet1$. Right-click this file, and choose Export > To dBase (single)… This opens the “Table to Table” tool dialog box.

Click for larger image

Browse to the folder where you want to save this table in the “Output Location” text box. Just highlight the folder where you want to save, and click “Add”. Next, give it a name under “Output Table”. Make sure you give it a name – the Table to Table tool won’t do this for you automatically.

Click for larger image

Click for larger image

Note that you can also create an SQL expression and pick out certain data from the spreadsheet. This is something I haven’t played with yet, but it could prove useful. For now, we’ll leave it blank. Click OK and your table is created.

Click for larger image

Bring it into ArcMap by using the “Add Data” button or simply drag it from ArcCatalog into the Table of Contents in ArcMap.

Click for larger image

Click the Source tab at the bottom of the Table of Contents – yep, there it is. But we don’t have points on the map yet! Why not?

We have to tell ArcMap to go find our coordinate information and plot a point for each record in our table. That’s easily done as well. Go to Tools, and choose “Add XY data…”. We do this and a new dialog box opens. If you’ve named your fields LAT and LON, ArcMap will automatically populate them into the correct text box. If not, you can pick them yourself from a drop-down list of fields.

Click for larger image

Choose a coordinate system – since these were imported from a GPS receiver, WGS 84 is your best bet.

Click for larger image

Click OK and you’ll see a new layer – waypoints events.

Click for larger image

This only a temporary layer, but you can save it as a shapefile by right-clicking the layer and choosing Data > Export…

Click for larger image

Again, this is the hard way to do it. There are a number of utilities that can import GPS data and save it directly to ArcMap. I use the DNRGarmin utility developed by the Minnesota Department of Natural Resources, but you can search the web and find others. However, it’s nice to know what that utility is doing for you so you can do it by hand in a pinch.

To quote my friend Craig Collins over at Temple College, “They pay 747 pilots $100,000 a year to watch the plane fly itself – because they know what to do when the plane STOPS flying itself.”

World Population Data

Nordpil has published a dataset examining and projecting urban populations.

World database of large cities

Be sure to watch the video and play with the Google Earth visualization. Pretty interesting!

Reposted from the GIS Lounge Facebook page