These options are available only to the registered users after sign-on using your EO-SSO ID. This service is open to well-known, registered users of the U-TEP portal and to become one, you have to register with one of the U-TEP processing centres. The registration form is located on the “Develop and offer content” page.
The submission of the registration form generates the ticket within the platform’s ticketing system. The helpdesk operator is automatically informed about the new ticket and the following use case specific communitation is performed via this ticketing system.
You have the following options regarding your own dataset which you want to use on the platform:
When you log into the platform you can find the appropriate request forms on the “Develop and offer content” page. To view the request forms use “Managing your own datasets” link.
Fill the “Requesting upload of a user-provided dataset” form to upload your dataset for the publication as a product or as a dataset for processing.
You will be contacted by a selected processing center operator via the helpdesk system about the method of the dataset upload. The processing center operator ingests the data e.g. by harvesting it from a remote location, or by offering you an FTP drop-down point for the processing center. In the end, the help desk operator informs you about the finalisation of your request.
Fill the “Requesting publication of a dataset” form to publish the dataset you have generated on the platform as a new product.
This form creates a new ticket within the helpdesk system of the platform. When the necesarry configuration is done, the help desk operator informs you via this ticket about the finalisation of your request.
To upload your own tool you have to use the processor development environment package available on the “Develop and offer content” page.
The whole proces of creating, uploading and publishing your own processor (tool) is described in detail in the user manual’s “Processor Development Environment” tab. This process could be summarized in the following steps:
Donwload the VM image from the following link. It is a linux VM that comes with the a graphical user interface providing you the tools to create your own processor, test it locally within the VM and upload it to a selected processing center to make it available on the platform just for you or also for other users.
Copy the content of the virtual machine image directory to a local directory on your machine. Unless done, install Oracle Virtual Box on your machine. Start the VM and log-in using the “urbanuser” password for the default username.
The desktop of the Urban-TEP VM shows some icons of pre-installed applications:
The Urban TEP processor development VM comes with a few Earth observation products from different missions and sensors. The test data is located in directory /urbantep/eodata/<type> in the virtual machine:
The development environment contains one processor example as a template how local test, packaging and upload to a processing centre can be performed. You can copy this example into your home directory of the VM with /urbantep/software/urbantep-dev/bin/setup-example.sh.
The result is a directory called example with the following content:
Processors are usually packaged with Docker to allow for a constant environment of the processor even in different processing centres. The descriptor XML is a declaration that is independent of the target processing centre. It declares names and versions, the processor script to be started to run the processor, formal parameters, and dependencies.
urbanuser@urbandev:~ $ cat /urbantep/software/urbantep-dev/example/descriptor.xml
<?xml version="1.0" encoding="utf-8"?>
<utep:descriptor xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns:utep="http://urban-tep.eo.esa.int/schema/urban-tep-schema.xsd"
xsi:schemaLocation="http://urban-tep.eo.esa.int/schema/urban-tep-schema.xsd urban-tep-schema.xsd">
<utep:processor>
<utep:name>Fmask8</utep:name>
<utep:executable>fmask-and-merge.sh</utep:executable>
<utep:title>Urban TEP Fmask for Landsat 8</utep:title>
<utep:description><p>Performs cloud detection for Landsat 8 L1 products.</p></utep:description>
<utep:inputTypes>Landsat8</utep:inputTypes>
<utep:parameters>
<utep:parameter>
<utep:name>threshold</utep:name>
<utep:type>string</utep:type>
<utep:description>cloud probability threshold</utep:description>
<utep:default>0.2</utep:default>
</utep:parameter>
</utep:parameters>
<utep:packaging>
<utep:name>fmask</utep:name>
<utep:version>3.2</utep:version>
<utep:type>Docker</utep:type>
<utep:dependencies>
<utep:dependency>
<utep:name>snap</utep:name>
</utep:dependency>
<utep:dependency>
<utep:name>mcr_root</utep:name>
<utep:version>v81</utep:version>
</utep:dependency>
</utep:dependencies>
<utep:resources>
<utep:resource>
<utep:name>memory</utep:name>
<utep:value>7000</utep:value>
</utep:resource>
<utep:resource>
<utep:name>timelimit</utep:name>
<utep:value>3600</utep:value>
</utep:resource>
</utep:resources>
</utep:packaging>
</utep:processor>
</utep:descriptor>
The example processor can be tested locally with or without the docker container. There is a README file in the urbanuser’s home directory with all steps to be performed. The result can also be inspected using the installed SNAP by opening the result NetCDF product in the working directory of the example processor.
After the successful local test the processor can be packaged and uploaded to a processing centre for deployment and later application to larger datasets. The same way an updated processor can be installed to supersede a previously installed processor version (with explicit versioning, you are responsible for version numbers yourself).
Upload tool requires user name and password of a well-known user of the respective processing centre.
Brockmann Consult processing center
IT4Innovations processing center
After the successful upload, the processor should be available directly on the platform in the “Test my processor” thematic application.
All uploaded processors are listed in the “Services” tab in “Test my processor” thematic application.
If you want to publish your processor for all users or just selected community, you can do it via the request form on the “Develop and offer content” page. This form will initiate new ticket in the helpdesk system. The help desk operator informs you about the finalisation of your request.
If you want your custom processor to systematically generate the data, you need to submit an appropriate request form on the “Develop and offer content” page. This form will initiate new ticket in the helpdesk system. The help desk operator informs you about the finalisation of your request.
