GADM Team

GADM Team

Automatic Management of Multimedia Documents and Imaging

Team Leader: Pr.SERIDI HAMID

Overall objectives:

Automatic Document Management (GADM) is a computerized process for organizing and managing information and multimedia documents (text, image and video) within an organization. It mainly implements systems for recognition, indexing, segmentation, search and classification of multimedia documents.

GADM participates in the processes of collaborative work, capitalization and exchange of information. The automatic management of multimedia documents is the fruit of the contributions of several disciplines. This requires not only conventional skills in image processing and machine vision, but also machine learning and pattern recognition skills. Video Mining research is also a new focus that aims to advance this established field and attempts to meet the needs of users in different areas.

Scientific foundations

The GADM & Imagerie research team will deal with several topics related to the automatic management of multimedia documents, namely: video mining, text mining and handwriting recognition.

1) Video Mining
Research activities in this field will be based essentially on the following axes:

detection, recognition and tracking of different forms in a video;
the development of a generic system for indexing and searching video in a data warehouse.
the interpretation of gestures in a video.
2) Text Mining
Research activities related to this field are organized around:

development of text mining methodology within the information search process.
the integration of these techniques in an information search context.

3) Recognition of Manuscript writing
Research activities in this discipline consist of:

propose architectures of the systems of recognition of the handwriting
improve the different recognition modules (preprocessing, segmentation ...)
develop systems recognition of writing in different fields (reading literal checks, administrative forms, preservation of Algerian heritage writings ...)
Keywords: Information Retrieval, Recognition of Shape, Imaging, Recognition
Handwriting, Text Mining, Video Mining.