I'm a tech savvy, outdoorsy, surfer/scuba diver/swimmer/water enthusiast, and I am creative enough to want to start making videos and shooting photos of my favorite outdoor activities. I decided to invest in a GoPro HD Hero a few years ago. Here's what I have learned since then.
!Feel free to browse the main points since this is such a long article:
You need a high end computer to edit high end video
You need a flat lens to shoot with a GoPro underwater
Choosing video editing and conversion software
Choosing photo editing software
Image processing techniques
Camera Selection Part 1 - Do you need a GoPro?
Camera Selection Part 2 - Consider a point & shoot with a housing
1) You need a high end computer to edit high end video:
Having seen friends transfer their camcorder video to their laptops and stitch some movies together in the past (several years ago), I figured that I might only have a little lag and jumpy playback to endure while editing. I was completely wrong. Even after shooting 720p video (great, but reduced quality) and further converting it to a reduced format and resolution, my video editing software (a.k.a. Non-Liner Editing system, or simply an "NLE") would simply crash after shortly after I added any of the video files to a project. Thinking harder about those videos my friends produced, the size of the videos were always pretty small.
If you plan to invest in a camera, make sure you have a computer that can handle editing high definition video. Otherwise, be prepared to upload full clips directly to YouTube, etc. Be advised that I will immediately down-vote and close your video if the clip begins and ends with you fiddling with the camera. Some form of basic editing is a must - it doesn't take much.
I'm in the process of building a computer that will hopefully be able to handle editing 1080p video. Here are some recommendations:
CPU - The Intel Core i7 series is currently the industry standard, but it comes with a high price. The i5 line is known to get you by.
I'm interested in trying to go with an AMD processor for around $100 less. A great AMD processor seems to be the Phenom II X6 1090t. It has six cores running @ 3.2 GHz. Very fast.
RAM - Memory is relatively inexpensive and the costs will continue to decline. Start with at least 8 GB (if not 12 or 16 GB) if you're looking into a new machine. Make sure your motherboard can upgrade to 32 gigs for future upgrades, but 8-16 gigs will get you started. Shoot above the mid-point of the frequency rating for the motherboard's supported ram speed. 1600-1800 MHz should be plenty of speed for a while; moving into the 2000 MHz range is just costing more without really providing that much more power.
Motherboard - Upgrade potential is key! Unless you're getting top of the line components, try to find a motherboard that will offer flexibility to upgrade your RAM or your processor in the future. For instance, if your AMD processor is socket AM3, think about getting an AM3+ motherboard. Research if you're going to need to upgrade to dual video cards (I will not for this build). Think about other PCI expansion cards (sound, serial ports, RAID, TV Tuner cards, etc.) you may need in the future and make sure there's enough room for them. Remember, a powerful video card will usually take up two bay spaces in a computer case and it may cover a critical PCI slot.
Hard Drives - If you can afford a solid state drive (SSD), great. They're incredibly fast, and I've even heard of people using two of those at RAID 0, which seems almost unimaginably fast in 2012. Use one of those around 100 GB or so for $100 to run your operating system and your NLE (video editing software) but you'll need at least slave drive (hopefully two, running Raid 1 for redundancy) for real storage.
Otherwise, I suggest using 2 Western Digital (Segate drives are inexpensive however they're lower quality and have had a high failure rate for my friends and I) hard drives that run at least 7200 RPM.
!Upon receiving my two WD hard drives, I found that one was dead on arrival. This goes to show that all hard drives have a failure rate (although it's quite low), but it illustrates how important RAID 1 and other data redundancy strategies are. See my upcoming post on building a building a DIY video editing computer for more information on this hard drive failure.
Run them together using Raid 0 for increased efficiency and speed. This means that the computer will treat them as one hard drive, but it can write to both at the same time; thus it can write twice the information at once.
Video Card - A video card is essentially a mini computer with its own processor and memory whose sole function is to process and output video to a monitor. Whenever possible, use a video card that is not built into the motherboard. These are easier to upgrade, and using video card that is separate from the motherboard will divert the processing load of rendering video away from the CPU so your main processor can concentrate on running your operating system and NLE.
I've been thinking about the EVGA GeForce GTX 550 Ti because it's relatively inexpensive (≈$120) and it has some power to it. If you're thinking about using Adobe Premier with this card, you will need to alter a registry file to unlock these non-supported video cards to be able to use non-supported graphics card properly (check out a list of supported graphics cards for CS6).
I've read recommendations to use the GTX 460 (identical to the GTX560, but cheaper) with your intro video editing workstation.
If you have the money (starting around $250-300), use Adobe's list of supported graphics cards as a guide to begin selecting your video card.
Check out the Videoguys' blog (heavy on Intel suggestions) and Tom's Hardware and for some more discussion on building a video editing workstation. I disagree with having to rely on Intel processors, but I'm a spatial analyst, not a video professional, and I haven't tried it for myself yet. Intel is the industry standard, but they can be significantly more expensive. I finished my custom build for just over $1,000 (closer to $1,100 after building the case and adding a few parts like lighting and a hard drive rack)
2) You need a flat lens to shoot with a GoPro underwater
! As of October 17, 2012, the new GoPro Hero3 series waterproof cases come with flat lens standard, negating the need for additional equipment.
For the same reason that humans need a mask to see properly under water, the GoPro Hero & Hero2 cameras standard curved lens produces a blurred, unfocused image under water. High end camera housings have precisely aligned hardware that overcome this problem. The GoPro simply needs a mask (a flat lens and an air space between the water and the camera) to make full use of the HD capabilities.
Since GoPro finally released their version (I waited almost two years for the official GoPro Dive Housing), I would recommend their flat lens housing, however there are a number of options available to overcome this problem (don't forget, you can make your own):
• Official / guaranteed (to a point)
• Fair price
• No vignetting at any resolution (although some of my photos have slight shading around the corners which I don't mind)
• Looks awesome/professional
• Rated to ≈200'
• Currently no official red/cyan filter
• I've heard any of these housings can begin to crack/leak around 140' so I'd recommend considering an insurance plan to help cover it for any deep(>100')/deco diving
• Removable filters (red/UV/polarizing/etc.) that fit standard camera lens sizes
• Looks good/professional
• Comes with replaceable desiccant balls in the lens to absorb moisture (rad idea!), keeping the lens free of "fog" on the inside
• A bit expensive
• Tested to 130' - it's important to have a buffer. 200-300' is preferred for 100' diving
• Vignetting @ 960p
• Pressure tested to 180' (same ballpark as official version)
• Looks good. Solid.
• Adapter for filters
• No reflection issues
• Super expensive
• Filters not included (dumb)
3) Choosing video editing and conversion software
! Windows Movie Maker will NOT edit video straight from a GoPro or most other HD video cameras without a format conversion.
GoPro and other cameras that shoot HD video in a highly compressed video format called H.264. The file format that the user will be most familar with is .mp4 or .mov which is more of a container than the actual format, but don't worry about that. Read more about the H.264 codec or just assume that this HD video format allows high quality video to stream very efficiently over the Internet. Examples are Flash video (YouTube, Vimeo, Google's video chat), iTunes content, broadcast television, cable & satellite TV, etc.
In the same way that a computer needs specific software to display an image, play an MP3, or compose a text document, you need special software to be able to read (decode & uncompress) the .mp4 (MPEG-4 / H.264) format. Thankfully most operating systems come with a player built in to read this format, but sometimes you have to go download Quicktime / VLC to use as video players. Otherwise you may need to just download a codec or update your current software (like Windows Media Player, etc.) to view video that you shot.
Editing video is another story. Most video editing software cannot edit the MP4/H.264 that the GoPro/Canon/Nikon shoot for video, so it is often necessary to convert the video to something your computer can handle natively. GoPro offers their own software package for free called CineForm Studio that will convert your video or stitch together your time lapse into .AVI or .MOV file formats, then it'll allow you to do some basic editing, trimming, and color correction.
Otherwise, I suggest using another free tool called Format Factory. It let's you easily convert anything to anything. Seriously. It's fantastic. It does a very good job of converting video between compressed (.mp4 /.mov) and uncompressed (.wmv / .avi / .mov) formats without loosing too much quality. You can also choose to reduce the quality and save some space and processing power if necessary.
Easily convert between video formats using Format Factory
From here there are a few paths you can take on your way to begin trimming the ends of your video, stitching clips together, and overlaying music and adding effects to your movies.
Free - Go low budget. Convert your video and reduce the quality a bit and use Windows Movie Maker. As long as you your computer can handle the video you throw at it (or you are close to finding nirvana in your meditation sessions), WMM is actually pretty effective. Forget have much control over placing text or other visual layers/objects over your video, though.
! Update: Adobe just released a free copy of CS2 for download - no strings attached! That's a free copy of Premiere (video editing), Photoshop (raster graphics), Illustrator (vector graphics), and more. It's somewhat dated, but much of the most powerful functionality you can expect to use today is available. This can be a great tool to learn how to use this software.
Another higher end free option is Lightworks: an open source, professional NLE. I've heard that some of the more powerful tools are only available for purchase. I have yet to try the free/light version.
Under $2000 - Get a Mac and use iMovie. You're paying way too much, but they're great quality and you will definitely get what you pay for with Macs; you'll be happy. Also, no additional assembly required!
Over $2000 - Invest thousands in a serious work station, GPU video cards, and use Adobe Creative Suite or Final Cut Pro
4) Choosing photo editing software
I highly recommend using Google Picasa for intro photo editing. The software is free, powerful, and highly organized. It's great if you're a beginner with no knowledge of photo editing, all the way up to playing with impressive image touch-up tools, color adjustments, fun filters, and photo grids/collages.
Free, powerful photo editing software
Further, if you're interested in working your way up to more powerful software packages - like GIMP(also free), Photoshop, etc. - Picasa is great introductory package that provides an easy to use playground to begin learning about what goes into digital image correction, enhancement, and processing - essentially, how to make your photos pop. 5) Image processing techniques
Watch the video above for a demo of what the software can do, but the following is the general process I use to process an image taken out on one of my adventures. The screenshots below correspond to Picasa version 3.9.
Basic fixes: Straighten the image! A grid is overlaid for reference. Make sure your horizon or your subject are aligned properly using the slider bar.
Remove any red eyes.
If there is minimal backscatter or any other unsightly spots that you would like to "photoshop" out of the image, use the Retouch tool to replace unwanted marks/areas across an image - similar to a clone tool. Anything from a blemish on someone's face to a shadow of the camera creeping into a shot, to an annoying kid in the background can be quickly removed with the Retouch tool.
Crop out excess scenery if it takes emphasis off of the subject. I am a fan of using the background as the main subject, but a busy shot can take the focus away from fine detail on a reef or an interesting formation, etc.
Begin with basic fixes and touch-ups
Tune the image and illuminate the scene:
Adjust the Fill Light, Hilights, Shadows, Color Temperature, and (especially for underwater photography) apply a Neutral Color adjustment. Start by using the magic wand button (aka "I'm feeling lucky button") to see what the software thinks will make the image look best (basically, an algorithm attempts to normalize the distribution on the histogram by adjusting these settings). Feel free to undo that statistical guess and/or make further adjustments until you're happy. You can always undo all edits and start over.
Use "Tuning" controls to bring out some missed detail
Fill Light - Controls the overall brightness of an image. Be careful if there are many areas that are already bright or there is a high contrasts between the bright areas and darker sections that are being illuminated with this control, as the bright areas can quickly become overexposed. This shifts color values to the right side of the histogram. Highlights - Increases the range between bright and dark colors, making a richer, vivid image. This shifts the span of color space beyond the upper bounds of the histogram, potentially overexposing certain images. Shadows - Amplify the dark areas -- often representing shadows -- across an image. The color space is shifted beyond the lower bounds of this histogram; inverse to the method used on the Highlights control. Color Temperature - Tune a warmer (more red) or cooler (more blue) tone to the image. Moving the slider to the right (warmer/red) compresses the color distribution on the histogram. Moving to the left (cooler/blue) expands the colors Neutral Color Picker - Select which spot is the "whitest" and the image is adjusted accordingly. This is especially important when filming underwater where the color spectrum is reduced. A blue or green tint is guaranteed to be present at any depth without the use of significant lighting. This is this post-production version of adding a red filter to a camera rig and is a very easy way to restore a natural appearance to a significantly green/blue image. You might need to Undo this and select a few different spots on the image until you are satisfied. Use this along the Color Temperature slider for further adjustment.
Adjust the texture of an image:
There are a few specific tools out of the 36 total processing tools on the last three tabs that can adjust the focus and bring out shapes like fish that are hiding behind poor visibility in the water. Sharpen or Soften the texture using these tools on tabs 3 & 5, respectively.
Applying a touch of the "HDR-ish" tool on tab 4 can really help bring out the texture of a reef or a fish through poor underwater visibility. Further, it gives a unique, highly processed look to a photo with vivid colors. Read more about the High Dynamic Range photography if you're interested.
The last three tabs are a variety of 36 different filters and fun effects
Adjust the colors... again:
Now that your image has an improved color balance and texture, you can further process the photo to really make it pop. Here are a few of my favorite filters:
Graduated Tint (at the bottom right of tab 3) - Use this especially for beach shots or photographs where the sky is present to give the sky a more rich color that may be washed out from excess light on a bright day, or from a slightly overcast sky. This tool will apply a selectable color tint to the top portion of the image and it will gradually fade into the natural color of the rest of the image. Turn the
Boost (on tab 5) - This tool is great for emphasizing rich colors across the image. Move the slider all the way to the left and experiment with adding small amounts of this tool at a time. Just a touch can be plenty.
If necessary, finish with a wild filter:
Play around with any of the other filters. Especially for underwater photography, consider a standard black & white photo for deep dives where shades of blue are essentially the only color that will be observed. I find it difficult to view a rich blue photo changed to black and white, so consider applying the black & white filter early in the processing, and move on to another photo. Come back when your eyes will be fresh and your retinas can readjust to a colorless image.
Additionally, play with Focal B&W to zero-in on a vivid color to highlight.
The 1960's filter gives an image a softer photo with some interesting yellows and greens (which looks great on beach shots with some bright, overexposed areas against a lot of water).
For underwater shots, adding a Vignette, or using a filter with a shaded/darkened border can give a tunnel vision effect, giving the feeling of wearing a mask underwater if the image is in landscape orientation (wide, rather than tall). Be sure to think about how the image is Cropped to focus the attention on a specific object like a fish or a shell in the center of the image.
Practice!
There's no single correct answer to creativity. Apply these fixes and filters to a series of pictures. Try to make a series of photos flow well with each other by attempting to match color schemes over a few photos. Try transitioning from a very natural looking scene to a highly processed image with wild, vivid colors across a few images as you descend on a reef.
Search galleries from other photographers online and try to recreate their work. Make a copy of your processing, and start again from scratch. Compare the same photo with different styles.
6) Camera Selection Part 1 - Do you need a GoPro? I'll ask you these questions:
Do you do cool things (don't worry if you're actually good at doing them or not)?
Would you like to see those things through a fisheye lens?
If you were to bring sensitive electronics with you while you do your things, will the electronics likely be destroyed?
Is anyone remotely interested in seeing what you do?
Do you have a Facebook/YouTube/Vimeo/Flikr/Twitter/Instagram/etc. account?
If you can answer yes to any of the above questions, you should probably get a GoPro camera. Here are some pros and cons of investing in a GoPro camera
Pros:
Cons:
• Super cool and easy to use
• Very small/light so it can fit anywhere you try to mount it
• There are lots of mounts (suction cup, head/chest/wrist/base plate/helmet/surfboard/snowboard/handlebar/rollbar mounts, etc.) that allow you to get very createive
• It can take a beating
• See the world in new ways
• High speed video mode allows you to critique your techniques in sports, etc. There used to be up to 60 fps, but the Hero3 Black has modes that shoot up to 120 fps in HD, and 240 fps in a non-HD mode!
• It's very simple to create cool time lapse videos. There are a range of rates from half a second to 60 seconds.
• It shoots amazing video, and really impressive photos. Seriously, that fish-eye lens is really cool if you've never used one before
• Wi-Fi capabilities are cool. The Wi-Fi Remote is easier to use than looking at the camera to check settings, and the higher Hero3 models can use an iPhone as a viewfinder
• The new Mic Adapter allows the use of better microphone setups. Very cool.
• Your current computer may not be powerful enough to edit video
• A flat lens is imperative to shoot underwater (scuba, etc.), but the Hero3 series come standard with a flat lens housing. Previous HD Hero & Hero2 versions require the add-on dive housing
• Get used to the fisheye effect if this is your main camera
• No flash / strobe sync
• The old LCD screen and the new touch screen are $80 add-ons. The fisheye lens provides a much wider field of view than regular cameras, so it's not absolutely imperative to use a viewfinder but it definitely helps set up shots. Also, it's a little hard to see the older LCD screen in bright sunlight (not sure if the new touch screen is improved)
• Mounts and add-ons are addicting. This can become an expensive hobby in itself, so make sure you use it
• The Original HD Hero does not perform very well in low light: the image turns to distorted pink and white hot spots. The Hero 2 is supposed to be a bit better with this
• SD card not included. I recommend at least an 8 GB class 10 card. A 4 GB fills up rapidly, but will work just fine.
• The small, light camera will shake like crazy in the water, but it can easily be mounted to a camera tray with some weight
Weigh the pros and cons of each setup. The Hero2/Hero3 Gold are additional hundred bucks more expensive than the original HD Hero/Hero3 White models, but depending on how you use them it may or may not be worth the extra cash. If you just want a sweet camera to play around with, grab a Hero3 White Edition and add some accessories later. If money is not an issue, absolutely go with the Hero2/Hero 3 Silver. You'll be happy with any of them. Be sure to do some research before you do buy and of them, though. Keep your ear to the ground for new cameras and accessories to be released. A lot of people (including small businesses) invested in the original HD Hero, then one day out of nowhere the Hero2 hit the streets. The same thing happened with the Hero3 series. I originally wrote this article in September, 2012, not knowing anything about the Hero3. A week prior to the release, they completely dropped the HD Hero 960 (see below), and dropped the price on the HD Hero. Then one night the Hero 3 series was announced, the HD Hero is gone, and the GoPro Web site doesn't mention the old products anymore. At least Apple lets you know that they're ready with a big development - and their old products don't drop out of existence.
! Here were the three options prior to October, 2012 for choosing a GoPro camera: $150 - HD Hero 960: The 960 is no longer available as of October-ish, 2012!
Shoots up to 960p (which is still really good, and is a higher resolution than I've been using for 2 years. I'm perfectly happy shooting 720p. 1080p not included
5MP photos
Does not shoot 60 frames per second. I really like the high speed video for slowing down action sports.
No bus port for add-ons like a bigger battery pack/LCD back plate/WiFi accessory/3D/etc.
$200 / $130 HD Hero: The HD Hero is no longer available as of October 17, 2012
+$30 for the surf accessory kit / +$40 for the helmet kit / + $40 for the motorsports kit (the suction cup mount is amazingly strong!
Shoots 1080p / 960p / 720p HD HD video
5MP photos
Shoots high speed video (60 fps) @ 720p
Bus port included for expansion packs
$300 - HD Hero2 (includes selected accessory kits):
Improved lens
Improved image processor
Is said to do better in low light
11MP photos
7) Camera Selection Part 2 - Consider a point & shoot with a housing For the amount of money you'll drop on a GoPro rig (camera + LCD backplate + dive housing + mounting accessories = at least $300 to $400), you should consider a simple Canon point & shoot camera with matching housing. Most Canon point & shoots have a good, dedicated "underwater mode" that automatically adjusts the white balance for blue/green saturation, while still giving plenty of room for manual control. I suggest browsing the features and comparing a few different cameras and find a matching Canon or Ikelite underwater housing.
Canon PowerShot ELPH 100HS
After much research and consideration, I chose the Canon PowerShot ELPH 100 HS with the Canon WP-DC39housing. This camera shoots unbelievably rich video up to 1080p, and takes beautiful stills @ 12.1 megapixels. There is a lot of control over white balance, ISO selection, shutter speed (long exposures can be set to anywhere between 1 & 15 seconds). Additionally I was pleased with a few other technical specifications, and the amount of built in effects and filters.
Canon WP-DC39 waterproof housing for the ELPH 100HS
After comparing the cameras in the ELPH series, I decided that the higher costs of the newer models did not justify the slight improvements. I am extremely happy with the quality of photos, video, and with the control the camera gives me to shoot. Although the Ikelite digital camera housings are a bit more expensive, they may give a user more control over all of the buttons for a supported camera that may not be available on the Canon version. Compare the back of the Canon WP-DC39 (this image is another model, but the DC39 looks almost identical) with the Ikelite 6243.01 housing. I was able to find the Canon housing for around $80 cheaper. While the Canon housing does not have full control over every button, thus every setting on the camera, I still have significant control over custom exposure settings (most important). Using the Canon housing, a user is limited to three modes:
Automatic - though manual controls are still available
Beach - designed for bright, sunny scenes over highly reflective surfaces (sand/glare)
Underwater - rapid white balance, which does a great job at removing blue/green overtones
Although the built in effects filters are really impressive, they're 90% useless while scuba diving. Many of those effects can be recreated in post-production, anyway. The dedicated underwater setting with manual controls are far more effective in capturing the best image possible. I am absolutely thrilled with what I've shot so far and have not fully met this camera's potential. I could use a more powerful video light with a wider beam. I'm currently using a hand mounted Intova Ultra Nova 220 lumen LED in the video below, which is bright, but it has a tight beam and a limited cone which isn't great for shooting video. Additionally, I would like to invest (will probably just build) a stronger underwater tripod. I'm currently using the Joby Original Gorillapod, however the legs are not quite strong enough to make me completely happy. It's a great tool, but the camera housing is a bit too large to make the small tripod 100% effective. I would suggest a larger version, or build your own mount. One additional item that would help would be weights and a tray to hold the camera while shooting underwater. The camera is almost completely neutrally buoyant (just slightly negative so it sinks), so additional weight would help stabilize shooting long exposure photos in a cave, and a heavy tray would minimize shaking during moving shots (video especially) into a flowing spring or current. I'll end with a couple basic some samples of what I've shot so far with my Canon ELPH 100 HS rig: Freediving Wakulla Springs(click to enlarge the images):Cavern dive at Jackson Blue Spring:
I've done some geometric correction using other software packages, but have not yet tried to do it in ArcMap. It turns out that there's a pretty small learning curve in the world of georeferencing in ArcMap. I recommend simply turning on the Georeferencing toolbar and playing around with the hand full of tools that it holds. All of the most useful tools are hidden away in either the Georeferencing menu on the left of the toolbar or in the small menu of three manual alteration tools to the right of drop down menu of selectable layers. In the hour or so that I have been attempting to georeference a couple images, I already want to pull the tools out of the drop down menus and place them on the main toolbar for faster access.
A screenshot of the Georeferencing toolbar in ArcMap
Turn on the Georeferencing toolbar:
To verify, turn on the Georeferenceing toolbar by either right clicking in empty gray space beside tool bars/file menus at the top of the ArcMap window, or by going to Tools :: Customize and on the Toolbars tab, place a check mark in the box next to the Georeferencing toolbar option (then click Close). The toolbar (see above) will appear.
Add imagery:
Use the Add Data button to add a raw, unreferenced images (.jpg, .png, .bmp, etc.) to an ArcMap project. Expect to receive an "Unknown Spatial Reference" message that explains that the image cannot be projected. If there are spatial layers already added to the project, the images will probably be displayed somewhere off in a corner. Right click on the image layer in the table of contents and select Zoom to Layer to view it.
Here are a few good resources to consider to obtain imagery:
LABINS (Land Boundary Information System): DOQQ imagery, high resolution imagery, Digital Elevation Models, DRGs, and other spatial datasets for the state of Florida.
Georeferencing:
I have access to rectified imagery and other reference datasets (transportation networks, hydrography, etc.) so I'll use those to illustrate some techniques (very quickly and roughly). I'll just be using aerial imagery from 2005-2009 and a coverage of State Routes.
Zoom to the area where the image should cover. This example is looking at an area in Florida around Taylor Creek, just north of Lake Okeechobee:
Choose an image to manipulate from the Layer menu on the Georeferencing toolbar
Since my .jpg image (in this case) is hiding down at the grid origin, somewhere in the Gulf of Mexico, I will have a hard time matching similar locations in the relation process. Open the Georeferencing menu on the toolbar and select Fit to Display. The image will shift into view and be resized to fill your map view.
Begin by using the three tools in the menu to the right of the Layer menu to roughly align the image:
Primarily, it seems that the images are rotated about 90° clockwise (north is pointing east). Use the Rotate tool to fix this. The original top of the raw JPEG is now facing west:
(This can also be performed from the Georeferencing menu under Flip or Rotate :: Rotate Left)
Rotated the image 90° anti-clockwise so north
(Click to enlarge images)
Next, use the Scale tool to shrink the image a bit:
Shrink the image with the scale too
Finally, use the shift tool to move the image closer to its correct position:
Moved the image using the shift tool
Open the Link Table by clicking the button on the far right of the Georeferencing toolbar. This will keep track of the control points that will be added in the next step.
Zoom in to a location that is easily distinguishable in the rectification image as well as any reference data. Roadways are excellent references, especially when they intersect with each other or with a stream, etc. A less effective reference feature can be a lake or stream, for example, because the shorelines can change over time. Roadways often have less variation. Local knowledge is very beneficial in these situations.
Below is an example of a roadway intersection that is a decent enough reference point to use. The red circle indicates where the control points will be placed:
(Broken GIF: the image should be animated
to illustrate the two control point locations)
Use the Add Control Points tool to add pairs of control points. The first click will represent the "from" location that will correspond to a point on the image that is being rectified (the black & white image in this case). The second click will represent the "to" position, where the image will be moved or stretched. This pair of control points creates a vector that is used to rotate, stretch, and skew the image.
Add a few of these control points across the extent of the rectification image. Each time a control point is completed, the information will appear in the Link Table, and the image will move (if the Auto Adjust option is checked in the Georeferencing menu).
Note the Total RMS Error that is reported in the lower right corner of the Link Table window. Root Mean Square Error is "a measure of the difference between locations that are known and locations that have been interpolated or digitized" (Heather Kennedy. 2000. ESRI Dictionary of GIS Terminology, ESRI Press, Redlands, CA.). In general, lower values are desirable. Read more about RMS Error.
It helps to turn layers on and off in the table of contents pane. Another useful technique is to use the Swipe Layer tool (found on the Effects toolbar). Make sure that the image/layer/group layer is selected in the Effects toolbar selected layer menu, then select the Swipe Layer tool and click the map view. Then simply click and drag the tool across a map to reveal lower layers on the map.
Image rectified with various control points
* Caution! Changing the selected layer on the
Georeferencing toolbar will remove any editing
or work that has been done to an image. Use the
following methods to preserve your referencing
Save image referencing:
From the Georeferencing menu, simply choose Update Georeferencing
This adds two files ( [image name].aux & [image name].JGw ) to the directory where the image is located. These two files are referenced when the image is added to a project,future
(Optional) Export image:
From the Georeferencing menu, choose Rectify to bring up the Save As dialog
Change the Output Location (choose a folder or workspace), and change the name of the image if necessary. The remaining parameters can be left with the default settings. Click Save to continue.
Use the Add Data button again to add the new image to the project, or distribute as necessary
Remove null values:
After adding the image, you may notice a black box surrounding the image. This is simply the blank space left when the new image was created from an image that was rotated slightly. These null values have the value of 0 in the dataset and can be easily removed:
Right click on the image layer in the table of contents pane and choose Properties (or double click on the layer)
Navigate to the Symbology tab
Place a check next to Display Background Value setting. Ensure that 0 is the denoted value, and that the color selector is set to No Color. Choose Apply or OK to continue
Removing null values from the layer
Several georeferenced images
One problem remains. There are still some black outlines on the edges of these images. Those areas are inherited from the original raw images. These can be handled in two ways;
The images can be cropped from the original JPEG files before they are georeferenced to remove the outlines. Be sure to make copies of the original images before any editing is applied.
Manipulate the geographic extent of each layer:
Open the layer properties for an image
Navigate to the Extent tab
Change the "Set the extent to" option to "a custom extent entered below"
Increase (Left & Bottom) or decrease (Top & Right) the Visible Extent values by small intervals and click Apply to view the results.
* Reset the values if necessary by changing the "Set the extent to" option to "the default extent of this layer"
Several georeferenced images with altered geographic extents
If this is still not suitable, you may need to pre-process the image with some other graphics software.
Water resources are challenging to model in a geographic information system because these features are often so temporally variable. John P. Wilson, et al. discuss in the article, “Water Resource Applications of Geographic Information Systems” many methods of water resource management and assessment. Often, these topics reiterate the relation of data sharing with the increasing level of technology. These advancements allow a greater number of less professionally trained individuals to contribute what can possibly be high quality data with a varying degree of uncertainty. For this reason, future technology will need to incorporate these individuals’ level of training and geographic and scientific understanding into the framework of tool building. Thus interfaces for using such tools must be created to suit a wide audience efficiently. Otherwise, difficulty in using technology directly translates to uncertainty. Scale is addressed often as well when reviewing other issues ranging from the integration of spatial data acquisition technology to data sharing practices.
Primarily, it is stressed that raster grid resolution selection should attempt to match the size of the smallest features that are being modeled. Accuracy of these resources is quickly degraded due to excess aggregation of topographic features. Generalization in this regard minimizes the number possibilities of variation for a given area where a smaller stream feature, for instance, could exist in reality. Using a raster resolution of 30 meters rather than the 2-10 meter resolutions suggested by several studies mentioned easily smoothes over smaller changes in topography. Further, a 30 meter cell size yields only 21% - 30% of correct slope gradients. However, if only larger (major) features are required to be analyzed or communicated, 30 meter resolution can sometimes be quite sufficient.
Many water resource applications are GIS-based. The National Hydrography Dataset is a prime example of how to digitally store and communicate a hydrologic network by logically displaying geographic information with its organizational and descriptive parameters embedded within in a GIS environment. An example of temporal water quality monitoring is the Florida DEP's Watershed Monitoring program and EPA & FDEP’s STORET (“STOrage and RETrieval” database) program, where water quality data from across the state of Florida is collected, managed in databases, and can be served internally or publicly via internet mapping service, while being provided to in-house personnel for GIS analysis via ArcSDE & ArcGIS Server. Advancements in technology such as these increase the dissemination of information to a growing number of skilled as well as improperly trained stewards and contributors of information. Increased availability must be mirrored by an increased accounting of data collection information in the form of metadata to ensure reliability. A method of controlling uncertainty for every step of data collection is currently available by reporting appropriate metadata parameters prior to distribution of any data. Although this information is mainly included on official final-drafts of published datasets, more can be done deeper within an organization to include metadata to aide in the quality assurance of final versions of disseminated data.
Overall, it will be important to develop efficient models of water resources that can be used by an audience with contrasting degrees of familiarity with geographic information. By conducting further research on the most effective way to properly incorporate data with varying degrees of uncertainty, as well as investigating the most efficient scales at which modeling hydrologic networks should take place, the GIS community can continue to accurately model water resources in the future.
Multispectral sensors such as the Landsat series, SPOT, IKONOS, and QuickBird acquire anywhere from three to ten simultaneous bands of information across a scene. Each of these bands cover a relatively broad spectral range of electromagnetic radiation observation. Hyperspectral remote sensing, as its name implies, is generally composed of a greater number of spectral bands which observe a more precise (narrow) spectral threshold. The MODIS sensor uses 36 bands, while NASA's AVIRIS sensor captures as many as 224 bands.
John R. Jensen continues to describe in his book, Introductory Digital Image processing: A Remote Sensing Perspective, two methods of multi- and hyperspectral image acquisition methods: the whiskbroom system, and the linear and area array technique. In both methods, instantaneous field-of-view (IFOV) radiant flux – observed reflectance from the Earth's surface – is passed through a spectrometer in order to disperse, or separate the light into separate bands; ranging from blue to near infrared (NIR) and IR wavelengths, which are passed onto a spectrometer where it is dispersed and focused onto a array of detectors which digitally record the field of view. The whiskbroom method uses a rotating mirror to reflect and direct radiant flux through the spectrometer to a linear array detector which individually measures the value of radiation of each band that has been separated. This technique is best suited for capturing broad spectral ranges and is utilized by multispectral sensors such as Landsat MMS and SPOT, etc. These sensors undersample the observed radiant flux by making only a few measurements from wide spectral bands; as wide as several hundred nanometers which may cover more than one color of the spectrum simultaneously. Although the use of a dispersing element is similarly used in each method to separate incoming light into individual bands, the alternative method of does not use a scanning mirror; thus allowing a longer amount of time for a detector to record the incoming radiant flux of a given area. This extra duration of detection yields improved geometric and radiometric accuracy.
These methods have dramatic implications on the type of information that is produced. The varying techniques influence spectral, temporal, radiometric, and even spatial resolution of a produced image. Various types of investigation require unique parameters in the type of data that will be used in analysis. Having an understanding of the different types of information that are produced by multi- and hyperspectral imagery with a respect to ground conditions that are being observed will ensure the most accurate results are obtained during image analysis and exploration.
The process of extracting information from multispectral and hyperspectral imagery are largely similar, however there are a few preliminary steps one must complete before analyzing hyperspectral datasets. Various forms of calibration – radiometric, geometric, etc. – are common between these two types of imagery, however the larger volume of highly specific spectral bands associated with hyperspectral imagery permit the construction of "spectra" that closely resemble the quality of spectral signatures captured by spectroradiometers in laboratories. Further, initial image quality assessment of hyperspectral can be a much more tedious undertaking, although it is occasionally possible to use hyperspectral images with poor quality for atmospheric correction. Due to this more time consuming and redundant task of visual examination, many image processing packages have animation functions that provide a more efficient means of inspecting up to hundreds of images in a single session.
Overall, the dimensionality is the most distinguishing characteristic between these two types of remote sensing techniques. The low data dimensionality of multispectral imagery is significantly more accessible and less difficult to work with in basic research due to the low number of spectral bands that make up an image. The higher number of bands/images involved with hyper- and ultraspectral imagery generate numerous obstacles, ranging from data storage to processing abilities. Numerous images of highly specific spectral bands – often representing bandwidths of just 10 nm each – produce certain amounts redundant information. Statistical analysis aids in identifying, and removing or transforming data in order to reduce the dimensionality of the overall hyperspectral dataset, improving the efficiency of exploration and analysis. Although there are indeed variations that exist with regard to image processing techniques between these two types of remote sensing data, the causes again are largely due to dimensionality. An analyst must decide whether few, spectrally broad images or numerous and specific images will best suit his or her analysis.
Presented 3 February 2009 Spring 2009 Advanced Remote Sensing Florida State University Reviewing the work of Everitt, J.H., Yang, C., Sriharan, S., and Judd, F.W. 2008 Using High Resolution Satellite Imagery to Map Black Mangrove on the Texas Gulf Coast. Journal of Coastal Research, 24:6, 1582-1586
This is the (very abbreviated) presentation I gave summarizing the article above. My presentations usually go into illustrative detail about each point, so the information in the Power Point document itself is not as informative. The presentation discusses a brief introduction on the species, their location and salt-tolerant characteristics, importance of mangroves habitat, etc. followed by more technical information regarding remote sensing techniques, methodology and sampling design, accuracy assesment and field verification, and finally a brief discussion (mostly omitted from the presentation) regarding analysis of the results of the study.
When downloading imagery (Landsat, IKONOS, ASTER, SRTM, etc) I continually encounter the .tar extension/format of compressed datasets. 7-zip is a great (light and free) package to use.
A Hydrothermal Composite Enhancement produces a multi-colored, multi-band image that is composed of three ratioed bands of Landsat TM imagery (MIR[1.55μ-1.75μ]/MIR[2.08μ-2.35μ], Red/Blue, and NIR/Red). This image highlights both hydrographic temperature spikes across the area of interest as well as sources of thermal contribution, such as beaches and urban land cover.
to add a raw, unreferenced images (.jpg, .png, .bmp, etc.) to an ArcMap project. Expect to receive an "Unknown Spatial Reference" message that explains that the image cannot be projected. If there are spatial layers already added to the project, the images will probably be displayed somewhere off in a corner. Right click on the image layer in the table of contents and select Zoom to Layer to view it.
to fix this. The original top of the raw JPEG is now facing west:
to shrink the image a bit: 
to move the image closer to its correct position:
to add pairs of control points. The first click will represent the "from" location that will correspond to a point on the image that is being rectified (the black & white image in this case). The second click will represent the "to" position, where the image will be moved or stretched. This pair of control points creates a vector that is used to rotate, stretch, and skew the image.
(found on the Effects toolbar). Make sure that the image/layer/group layer is selected in the Effects toolbar selected layer menu, then select the Swipe Layer tool and click the map view. Then simply click and drag the tool across a map to reveal lower layers on the map.
