Large Dataset Processing in eCognition: Overlapping Tile Method

This video is part of the AmericaView eCognition software tutorial series. It explains a work flow known as "Overlapping Tile Processing." Overlapping tile processing is an efficient method for processing very large datasets using eCognition Server. It is one way of overcoming the memory limitations that are inherent to object-based image analysis.

This presentation is recommended for intermediate to advanced users of eCognition.

We are making some changes to the REDDnet mirror of the Hazards Data Distribution System (HDDS) in an attempt to make REDDnet as much like HDDS as possible. The new procedures are included here. They have changed only slightly:

To access HDDS disaster_restricted data via REDDnet, proceed as follows:

Use a web browser to navigate to: "https://se1.accre.vanderbilt.edu/lodn/"

To access disaster_restricted data, log in with the disaster_restricted user name and password

double-click the disaster_restricted subdirectory (left side of the window)

double-click the the desired event, i.e. 201001_Earthquake_Haiti

continue navigating through the subdirectories to the desired data

To access public disaster data, select "browse files as guest.

double-click the "Disaster" subdirectory (left side of the window)

double-click "pub" and then= "data" subdirectories

continue navigating through the subdirectories to the desired data

Top download a file, click on it. Allow the request to access your computer (necessary to write files to your local disk)

A Java Web Start window will appear. Select the Output file location and number of connections.

Experiment for best performance. Connection values >20 are sometimes effective.

Optimal setting depends on network connection, CPU, etc.

Click the check boxes on several files and select "Download" under the "Operation:" pull down menu to download several files at once. PLEASE DO NOT SELECT ANY OTHER OPERATION!

Performance should be up to 10 times faster than downloading from HDDS, but results will vary.

Hazards Disaster Data System(HDDS) via REDDnet

The AmericaView/REDDnet project is now mirroring EDC Hazards Data Distribution System (HDDS) data. At present, only the disaster_restricted data is available. The HDDS disaster_restricted password is required and must be obtained from Brenda Jones at EDC before accessing these data. Later, the non-restricted data will be added as well and will be accessible without a password.

Only HDDS data for current events are mirrored on REDDnet. At the time of this posting, 201001_Earthquake_Haiti, 200909_Samoa_Tsunami and 200909_indonesia_Earthquake are available.

To access HDDS disaster_restricted data via REDDnet, proceed as follows:

Use a web browser to navigate to: "https://se1.accre.vanderbilt.edu/lodn/"

Enter "RSWG" for the user name and the HDDS disaster_restricted password.

On the left side of the screen select: "disaster_restricted"

On the left side of the screen select the event you are interested in. for example: "201001_Earthquake_Haiti"

Continue navigating through the directories until you get to the data you want. The directory structure matches HDDS.

Click on any filename to download it. Allow the request to access your computer (necessary to write files to your local disk)

A Java Web Start window will appear. Select the Output file location and number of connections

(experiment with the latter for best performance. Numbers larger than 20 are sometimes effective. Optimal setting depends on network connection, CPU, etc.)

Performance should be up to 10 times faster than downloading from HDDS, but results will vary.

Please let me know how it works for you.