Monday, December 31, 2007

Comparison - WVS, SWBD, NGA PGS Global Shorelines and EVS Precision Shorelines

Shoreline Comparison

Three Vector Shoreline's Compared - WVS, SWBD and EVS Precision

One of my readers suggested I evaluate SWBD shorelines in comparison to other freely available shorelines. Previously, I compared these shorelines with my own EVS precision shoreline. I'll do it once more.

The WVS is widely available and manageable in size. There is another popular version of it called the GSHHS. It repairs varous WVS anamolies. In addition, it conveniently breaks the dataset into land, lakes, ponds and islands. At only 56MBs zipped, it is a very attractive option. As one can see in the two images, the WVS is the least precise of all large scale global shorelines. It is designed to be used at scales of 1:250,000 and above. For most of us designing projects for our classroom, employers or customers, this dataset does a fine job, it is free and the size is reasonable. It is when we need to do large scale precise shorelines that we require the next best global shorelines

Although the SWBD is a quality product and provides the end-user with precise SRTM water-defined shorelines, it tracks the water. The SWBD attempts to identify the place where the water meets the shore. That should be the shoreline, but it isn't. The top image shows the SWBD shoreline tracking the reef edges and ignoring the numerous motus within the reef. For us island mappers, the SWBD is not precise enough to identify the many shoreline features we require.

The same can be said for the NGA PGS shorelines. They are extremely precise, using Landsat ETM+ imagery as their base imagery. This is the same imagery I use to construct my EVS precision maps. I have the NGA PGS shoreline loaded into Marplot, a mapping package I use. The dataset is divided into 24 separate groups. My global coverage, using this dataset, is outstanding - except where clouds or other atmospherics obscure the shoreline or the shoreline consists of water-inundated areas (marsh, swamp, mudflats), which the NGA PGS product didn't identify. The cloud obscured areas on Landsat ETM+ imagery are validly poorly mappable to unmappable. The water-inundated areas are mappable, but the NGA contract called for the exclusion of these shoreline features, thus they were not mapped. I beg to differ. This is shoreline that must be mapped. There are massive river deltas NGA PGS mapped that depict shorelines that look nothing like the original Landsat ETM+ image, which is full of features that the NGA PGS product did not map. They just happen to be water-inundated. They are features you and I would describe as shoreline. Make no mistake, when the shoreline is cloud-free and not water-inundated, which most Landsat imaged shorelines are, this global shoreline dataset is outstanding. It's just all of those river delta islands that they didn't map that I have a problem with.


Comparison 4-Vector Shorelines (1)

Four Vector Shoreline's Compared

However, my EVS precision shorelines are pretty outstanding as well. When one compares my hand-drawn shorelines to programmed color tracking shorelines (or many other automated processes), my hand-drawn shorelines track quite precisely. Something about using one's brain processing powers, manual dexterity and years of experience that consistently allows one to create more precise shorelines than the WVS, SWBD and NGA PGS products.

As I work with Landsat ETM+ imagery and learn more about the digitizing craft, I am certain software is available that could do what I do more accurately, faster and in much less time. But it wouldn't be as much fun! Would it?

Enjoy!

Wednesday, December 26, 2007

Thoughts - Landsat Island Image Mosaic

I have over 600+ Landsat ETM+ Orthorectified Mosaics that two summers ago I patiently downloaded and copied onto a series of DVDs. These mosaics cover all of the shorelines and oceanic islands throughout the world. It is an impressive collection. This collection, coupled with Global Mapper v9.01 (GM), allows me to load and view the individual mosaics as I desire. Each mosaic, even compressed in MrSID format, is typically 100 to 250MBs. Them is some big files!

Over a year ago I had a discussion with a reader who wished to work with these files to make his own maps. I described where he could download the files. But, alas, he said the files, even in MrSID format, were too large for his system. Was there a smaller file of just individual islands, he asked. Not to my knowledge, I guessed and went back to making maps of islands.

That was until this past month and my work in image mosaics. First I worked with DigitalGlobe imagery. Sadly, they have never responded to my product idea. Oh, well, so goes life. Lately, I've been working with Landsat ETM+ again. And the following is my latest idea - Landsat Island Image Mosaic or LIIM.

Greek and Turkish Islands -  Landsat ETM+ Mosaics

Greek and Turkish Islands - Landsat ETM+ Mosaics

Some of my earliest mapping efforts were of Greek and Turkish islands located in the Aegean Sea. This area is virtually cloud-free and the clarity is outstanding. However, it takes seven Landsat mosaics to cover the area. Using a raster export feature within GM, I experimented with various raster formats and came up with georeferenced JPG. Pixel size of each resulting LIIM is 14.5-meters, 300 DPIs and projection is Mercator. The resulting image is outstanding.


Nisos Limnos - Landsat ETM+ N-35-35_2000

Nisos Limnos - Landsat ETM+ N-35-35_2000

I zoomed into my target area encompassing the primary island and created the georectified JPG. The above Nisos Limnos LIIM taken from GM gives the user the complete resolution power of Landsat ETM+ without the size problems. The above file, along with it's accompanying DEM, is only 7.2MBs.


Greek and Turkish Islands -  SRTM Tiles

Greek and Turkish Islands - SRTM Tiles

To really bring the Nisos Limnos LIIM to life, the SRTM data is the ticket. To cover the Aegean Sea one needs 83 individual SRTM tiles. Loads of data for most of us. Many of these tiles require some fixing, which is done using SRTM Fill (freeware) to fill holes. Once this data is loaded, GM allows for DEM creation. One is able to take the Nisos Limnos DEM and drape the Nisos Limnos LIIM on top and get a variety of 3D views of the island.


Nisos Limnos - SRTM N039E025

Nisos Limnos - SRTM N039E025

So I exported the underlying area into DEM format. This allows for some nifty looks at the island



Nisos Limnos - 3D View

Nisos Limnos - 3D View

Pretty cool! This is the Nisos Limnos LIIM draped over the Nisos Limnos DEM. All of this wrapped up in a 7.2MB package. I like it. How about you?

I should add that I have created 16 Greek and Turkish Island LIIMs. The following islands have LIIMs - Bozcaada, Gokceada, Nisida Fournoi, Nisida Gioura, Nisos Agios Efstratios, Nisos Alonnisos, Nisos Chios, Nisos Ikaria, Nisos Lesbos, Nisos Limnos, Nisos Pelagos, Nisos Samothraki, Nisos Skiathos, Nisos Skopetos, Nisos Skyros and Nisos Thasos. I intend to complete coverage of the islands of the Aegean before I head back to school on Jan 03, 2008.

Enjoy!

Thursday, December 20, 2007

How To Colorize a Black and White Landsat WRS Image

Tsu-Shima - Landsat from USGS WRS - Original Image (1-500,000)

Tsu-Shima - Landsat from USGS WRS - Original Image (1:500,000)

I did my "load and explore" exercise, where I take a random Landsat ETM+ mosaic, load it and see what looks interesting. Fortunately, I traveled to a part of the world I studied while in college, Japan. Although I speak some Mandarin Chinese and very little Japanese, my emphasis within my Asian Studies major was Japan. The islands making up Tsu-Shima rest between Japan and Korea. Both Korea and Japan claim ownership, but it appears Japan has "squatters rights" to the islands. The islands have a long history of human occupation, but the current human footprint is relatively small - over 85% of all of the islands land area is in a natural state.

So, why not transform the black and white Landsat WRS image into a useful colorized image. I decided to give it a try.


Tsu-Shima - Landsat from USGS WRS - Creating Land Mask (1-500,000)

Tsu-Shima - Landsat from USGS WRS - Creating a Single Color Sea Mask - First Pass (1:500,000)


Tsu-Shima - Landsat from USGS WRS - Sea Mask a Single Color (1-500,000)

Tsu-Shima - Landsat from USGS WRS - Sea Mask as a Single Color (1:500,000)

My first task was to isolate the land area making up Tsu-Shima islands. I did this using a feature within Global Mapper v9.0 that allows you to isolate certain colors and to make them transparent. One merely identifies a RGB (i.,e., 010,215,215) color and turns it off. The first image in the two above shows a first pass. It took seven passes to convert the sea mask into a single color. Each pass required that I save the modified raster image as a TIFF file. After a number of passes, I ended up with the second image. The cloud-like features in the lower portion of the image are clouds. They were dealt with later.


Tsu-Shima - Landsat N-52-30_2000 (1:500,000)

Tsu-Shima - Landsat N-52-30_2000 (1-500,000)

The Landsat ETM+ image taken from N-52-30, has way too much black and the shoreline is tough to clearly makeout. I am certain others could easily determine shorelines with a smart routine that identifies shoreline colors, but I don't have this level of expertise or the software. So, I forged ahead in my pedestrian manner. Taking the above Landsat ETM+ image, which is georectified, I overlaid my modified Tsu-Shima Landsat WRS image on top of it. Again, using a Global Mapper v9.0 feature, I made the top image (Landsat WRS) 60% transparent, allowing the green vegetation layer to appear to drape onto the top image.


Tsu-Shima - Landsat from USGS WRS - Land Mask Colored (1-500,000)

Tsu-Shima - Landsat from USGS WRS - Land Mask Colored (1-500,000)

I saved the resulting image, moved it into ImageForge, my "poor man's Photoshop", and touched it up. I made the sea mask a single blue color and the urban areas were changed to a gold color. I saved the modified image. I loaded this final Tsu-Shima image into Global Mapper, where it was georectified. Tsu-Shima image is good to use between 1:500,000 to 1:62,500. Anything larger and the shorelines are a series of clunky pixel shapes.


Tsu-Shima - From Landsat WRS to Landsat Modified

Tsu-Shima - From Landsat WRS to Landsat Modified - Central Detail

From a black and white Landsat WRS to my Landsat modified image took 4-hours to complete. I have a series of georectified TIFF image of the islands. As with all of my previous projects, they are available at no charge. Just drop me a line and let me know what you need. I like the look of my Landsat modified image of Tsu-Shima. Don't you?

Enjoy!

Monday, December 17, 2007

Tana Island and the Ark of the Covenant

Lake Tana - Landsat N-37-10_2000 (1-500,000)

Lake Tana - Landsat N-37-10_2000 (1:500,000)


Lake Tana - EVS Precision Marplot Map (1-500,000)

Lake Tana - EVS Precision Marplot Map (1:500,000)

At times I like to open a Landsat ETM+ of a region of the world and explore. This time I traveled to Africa, specifically to Ethiopia. I know the country of Ethiopia is perched on highlands and Lake Tana is the primary source of the Blue Nile. I also have read the story of the Ark of the Covenant being spirited away from Israel first to upper Egypt at Elephantine Island, then to Tana Island and finally to St Mary of Zion Church in Axum. My curiosity about Tana Island got the best of me. My first task was to research to see if any maps of the island were available via the Internet. Nothing! Not even of the lake!

So my first task was to rectify the mapping of Lake Tana. The only mapping I have depicting this lake are at a scale of 1:1,000,000 (DCW). The shorelines are very coarse approximations of the Landsat imaged shorelines as depicted on N-37-10_2000. I have spent the last 6-hours mapping the shoreline of this lake. I worked at a scale of 1:31,250 for all mapped shorelines. Smaller islands in the lake were mapped at an even larger scale. The lake's shorelines were easy to discern and digitizing was uneventful, but tedious.


Nargadaga Deset - Landsat N-37-10_2000 (1-62,500)

Nargadaga Deset - Landsat N-37-10_2000 (1:62,500)


Nargadaga Deset - EVS Precision Marplot Map (1-62,500)

Nargadaga Deset - EVS Precision Marplot Map (1-62,500)

This is the largest island within the lake and the largest island in Ethiopia. I carefully digitized it's shorelines, figuring the largest island had to be Tana Island. That is until I finished and began working in Marplot. It is there that I have my EVS Island name file (86,000+ named islands). It turns out bigger is just bigger. It is not Tana Island but Nargadaga Deset. Although it is a holy island, it wasn't the holy island I wanted.

After a little searching I found Tana Island located along the eastern shore of the lake. Not a large island, but at one time, the reputed resting place of the Ark of the Covenant.


Tana Island - Landsat N-37-10_2000 (1-31,250)

Tana Island - Landsat N-37-10_2000 (1:31,250)


Tana Island - EVS Precision Marplot Map (1-31,250)

Tana Island - EVS Precision Marplot Map (1:31,250)

Here is Tana Island, all of 500-meters by 50-meters. It is a holy island, with a monastery located on it. No woment are allowed to visit the island. The priest living on the island will show any men visitors where the Ark was kept while on the island.

This was a fun mapping project. I got to map a number of islands at EVS precision. I went back to my mapping roots using Landsat ETM+ as my base imagery and didn't have to worry about offending any image owners and I mapped a major lake to boot! Now, that's living the high life.

Enjoy!

Tuesday, December 11, 2007

Thoughts - Ask the Owners of the Images! Not the Users!

Last week, in my naive excitement to share a product idea with readers of this blog, I did something wrong. Sharing the product idea - that was fine. However, it was brought to my attention that my proposed product sounded like a real product. IT IS NOT A REAL PRODUCT! It is merely an idea I shared to solicit feedback on its potential viability as an image tool.

I admit, as I reread my post,it sounded as if I were offering the proposed product for sale, something I would NEVER do without permission from both DigitalGlobe and Google Earth. It seems one of my readers took offense at the language of the post and an offer to let readers view a product prototype, at no cost. After thinking about my offer to allow readers to download the proposed product, I decided that it was a poor idea for a number of reasons, so I withdrew that post and previous posts that discussed this proposed product. I will handle this proposed product idea with the parties involved and not with my readers

Now I want to share what I felt as my motives and character were being discussed by strangers via a comment thread on the blogsite Technical Ramblings. First, I felt pissed that my innocent motives were being challenged by someone who doesn't know me, and only knows my posts. I felt I was being called a thief for allegedly stealing images from both DigitalGlobe and Google Earth. Worst, this individual intimated that I was doing this knowingly. I attempted to defend my motives and apologize for any misleading statements I might have made, to no avail. The scolding continued and a suggestion that I read a September thread discussing the proper uses of Google Earth images. I read the thread and, thank God, after reading it I snapped out of it.

The next morning, I contacted a well-placed representative of DigitalGlobe about the "fair use" of their imagery taken from Google Earth. This individual is well aware of my work and has viewed it a number of times. He assured me that my use of their product, to derive my island maps from, was okay! His only caveat was that my maps and the use of their images was not to be used in a commercial operation and I DO NOT sell my maps or their images, DigitalGlobe is fine with what I do. As to my proposed product idea, that is now between me and DigitalGlobe.

When I first began making maps of islands, using DigitalGlobe images fround in Google Earth, I spoke with a Google Earth representative who repeated the same approval of my island mapping and the use of their imagery to better share my mapping efforts. I was told to include all appropriate trademarks on the images and to not produce maps or to use their images for profit. This person suggested I contact DigitalGlobe to secure their permission as well, which I did.

This is my lesson learned - avoid threads, no matter how well intended the parties might be. Most often the comments are strictly opinions of people like you and me, they just sound more learned then we do. The bottom line is if you want to determine once and for all if your actual or intended use of images or derivatives taken from those images is "fair use" or allowed or how ever you want to describe it, ask the owners of the images, in my case DigitalGlobe and Google Earth. Both parties approved my use of their images. So, that's that.


Thursday, December 6, 2007

Ancient Small Martian Island - If Only There Was Flowing Water!

Small Martian Island - Mars Mission Image

Ancient Small Martian Island - Mars Odyssey THEMIS Image (19-meter)


Small Martian Island - EEVS Map

Ancient Small Martian Island - EVS Map (19-meter)

What if the surface of Mars could be flooded to a depth of 20-meter? There would be islands where hills stand today. One of these islands would be this unnamed small island located in Ares Valles (16-lat - 30.4W lon). It would look like a barrier island, a 2-meter above water sand spit, shaped by tides and wind.

This is my first extraterrestrial island project. It took about 1-hour to complete. The circles scattered about are meteor plunks, most probably the large one's ejecta. The link to the Mars Odyssey website gives the curious reader loads of information about the area and the dry island features.

Perhaps if we pray for rain, Martian rain, and an increase in temperature, the seas will flow once again and this small island will live once more.

Enjoy!