Loading...
 
ESA > Join & Share > Technology Projects > MIMS Project
Print

MIMS Project

Project Title   MERIS Information Mining Services System
Project Acronym   MIMS
Contractor(s)   ACS

 

Image Context
          Objectives
             Architecture
          How it Works
Output

 


Context   Top

In recent years our ability to store large quantities of data has greatly surpassed our ability to access and meaningfully extract information from it. Emerging needs from big applications (e.g.: change detection, global monitoring, disaster management support, etc.) and the continuous increase in archives' size and Earth Observation (EO) sensors' variety require new methodologies and tools for information mining and management, supported by shared knowledge. The manual process performed by experts to mine information from images is currently too complex and expensive to be applied systematically on even a small subset of the acquired scenes. This limits the full exploitation of the petabytes of archived or new data. The issue might become even more challenging in future since more missions - including constellations - are being planned, with broader sensor variety, higher data rates and increasing complexity. This raises the need for knowledge supported automatic or semi-automatic feature recognition and classification techniques to replace the manual interpretation activity. Preliminary results from recent R&D activities are showing the possibility to use Image Information Mining (IIM) techniques to ease the access to the imagery also through their information content.

IIM techniques include:

  Image Probabilistic Information Mining (PIM) methodologies
  Image Feature Extraction Algorithms (FEAs)

The system implemented through the MIMS project shall permit the application of PIM  to MERIS Level 1 products, possibly extending to other sensors. The environment emerging from the project represents the basis for the implementation of a prototype system for a full exploitation of IIM techniques (PIM and FEAs) on a variety of EO products, and for a simple provision of related services in the ESA Service Support Environment (SSE).


Objectives   Top

The MIMS project aims at simplifying access to and therefore at possibly expanding the use of multi-mission EO data. Contribution to this achievement comes from the use of emerging technologies for Probabilistic Information Mining (PIM), which permit selection, retrieval and understanding  of images through their information content. The system implemented through the MIMS project shall permit to apply Probabilistic Information Mining to MERIS Level 1 Reduced Resolution (RR) and Full Resolution (FR) products, possibly extending to other sensors.

The objective of the project is to develop a prototype permitting:

  Image Content Based Image Selection (e.g. for the identification of MERIS RR images free from clouds over the user Area of Interest)
  Image Content Based Image Retrieval and Information Discovery for other features different from clouds (e.g. algal blooms)
  Image Information Discovery and Scene Understanding (e.g. on MERIS FR images)
  Image Application of Probabilistic Information Mining on different products (e.g. Landsat, SPOT, MODIS, etc.)
  Image Exploitation of system functionalities also on user-provided images
  Image Provision of related SSE services (e.g. for extracting features and thematic maps from images)

The system can be installed, configured and used in different contexts, ranging from large archive and processing sites to specialised service providers' environments.


Architecture   Top

Image

Figure 1: Architectural overview

Figure 1 provides an overview of the MIMS architecture. On one hand, the system includes all basic functionalities of the KIM prototype for processing EO products and extracting information from them. MIMS will provide also an improved user interactive interface for ingesting new EO products and performing PIM on them, and will permit also the link with external data sources for a systematic ingestion of EO products. On the other hand, MIMS will provide catalogue services on the SSE, which will permit to:

  Image Activate database searches using features already defined (e.g. clouds)
  Image Select images relevant for the user, through the identification of their content
  Image Notify the user on product availability (e.g. via e-mail)
  Image Download products (via FTP) from archives
  Image Extract features and thematic maps from images

Furthermore, Figure 1 also shows the different roles which interact with the system:

  Image Expert
  Image User

The Expert has access to all MIMS functionalities and is capable to:

  Image Use the interactive interface for training the system and recognizing specific features
  Image Make the identified features reusable for the provision of catalogue services
  Image Publish detected features on SSE

The User can access to MIMS functionalities via dedicated SSE services and is capable to:

  Image Query and browse products into specific collections
    - using spatio-temporal criteria
    - using available features provided by the system (i.e. features previously defined by experts)
  Image Select relevant images and retrieve product identifiers and/or feature maps
  Image Download selected products


How it Works   Top

 

Interactive Interface

The MIMS system permits to interactively train the system to identify relevant features on ingested images and to perform catalogue searches through their content. The following figure show an example of the user interface, in the case of the cloud identification on MERIS RR images.

Image

Figure 2: MIMS user interface

 

SSE services

The MIMS system permits also to activate, on the basis of extracted features, SSE services for catalogue searches and feature extraction. One example is the MERIS Cloud-free Subscription Service for MERIS Level 1 RR products transiting in the ESA rolling archives. The service actively informs users - for the specified subscription period and with a selected time interval – when new MERIS Level 1 RR products touching the specified area of interest and having, within the area of interest, a cloud coverage below a specified threshold, are available in the rolling archives. The user is also provided with the product identifiers and the link for product download. 

 

     Image     

Figure 3: Snap shot of two MERIS Cloud-free Subscription Service web pages (left side: selection of area of interest and cloud coverage threshold; right side: display of results of the catalogue search).


Output   Top

The MIMS project aimed at developing a prototype system permitting (either via an interactive interface or through SSE services) to fully exploit Probabilistic Information Mining (PIM) capabilities:

  Image Creation and management of images collections for Content Based Image Selection and Retrieval, Information Discovery and Scene Understanding
  Image Extraction of relevant features and thematic maps from ingested images
  Image Search and Download of selected products from archives
Ø

The MIMS prototype, as well as related SSE services, has been delivered at ESRIN on July 30, 2007.

    Top

Contributors to this page: .

Page last modified on Wednesday 15 of December 2010 16:06:41 CET by .