Improving Large Image Viewing in the Crisis Management System SécuRéVi Mehdi Tahan, Jean Vareille, Laurent Nana Laboratoire des Sciences et Techniques de l’Information, de la Communication et de la Connaissance Brest, France e-mail: {Mehdi.Tahan, Jean.Vareille, Laurent.Nana}@univ-brest.fr Olivier Danjean Inovadys Brest, France e-mail: Olivier.Danjean@inovadys.com Hervé Mahoudo, Gilles Cloarec SDIS29 & EDF Brest, France e-mail: Herve.Mahoudo@sdis29.fr, Cloarec_Gilles@hotmail.com Abstract— This paper deals with the processing and integration of images in SécuRéVi platform, a crisis management system, with an emphasis on large images whose handling leads to specific difficulties. Indeed, in the domain of crisis management, images are key elements for understanding and taking proper decisions. Images of different nature (map, aerial view, 360 ° view, etc.) are helpful and adequate solutions that need to be provided in case of sinister. After describing the SécuRéVi platform, we shortly present its images processing approach and establish some design constraints and criteria for the development of a large image viewer. Then, we present our proposal for improving large image handling and its implementation in the SécuRéVi platform. The paper ends by conclusion and future works. Keywords-crisis management; image processing; large image; image segmentation I. INTRODUCTION Crisis management requires a good understanding of the sinister environment for adopting necessary attitudes for the resolution of incidents. Different prevention measures are provided for this purpose, such as emergency planning. The SécuRéVi [1] platform is inscribed in this process. It is based on the Global Safety Plan (GSP), which is a knowledge base aimed at gathering all the information needed by preventive approaches dedicated to crisis management. The understanding of a situation is a major challenge for crisis resolution and images are key elements for such understanding (like in geospatial imagery, scientific visualization or immersive application [2]). It is therefore important for a crisis management tool, to provide adequate solutions for images handling. In the context of crisis management, images can come from various sources (such as satellite [3], camera [4] or industrial plan) and are incorporated before or during crisis management. They are extracted from camera shots or shots assembling [5]. The size of the used images goes from tens to hundreds of millions of pixels. The image generation systems offer increasingly higher resolutions, but often, visualization tools do not allow their full restitution because the screen resolution is most often lower than that of camera sensors. Given the large size of images generated, the use of conventional tools quickly becomes problematic and they can't be used by any type of computer [6]. Therefore, we decided to develop a tool allowing a better use of these images. We start by describing the SécuRéVi Platform (section 2). Then, we show the importance of images for crisis management and explain their handling in the SécuRéVi platform (section 3). Thereafter, we present our proposal for image processing and precise the key elements for a successful implementation (section 4). Afterwards, we describe the implementation of this process and its evolution to obtain better performance (section 5). Finally, we conclude and propose some lines of thought for future works (section 6). II. SÉCURÉVI PLATFORM The SécuRéVi Platform [7] is a platform dedicated to communication and understanding, as well as decision making and the workmanship (see Figure. 1). It allows different actors from various professions to communicate and collaborate for a crisis resolution. It is based on the GSP integrated into a Geographic Information System (GIS), as well as various sources of external data (measurement tools, camera, business oriented computer tools, etc.). The whole interacts with a monitoring and events forecasting system (accident, incident, etc.) with commitment of resources (staff and equipment) in which each object, event or staff, can index data (e.g., description sheet, procedure, user manual, etc.) . SécuRéVi allows an approach in both space and time. It provides real time monitoring which makes it possible to create recording for replays that will feed the "lessons learned" database associated (feedback). Finally, in the platform, a 2D world can be associated with a 3D world, and this makes it possible to go from the virtual world to the real world and inversely. SécuRéVi works with a standard computer which can be enriched with other tools (video projector, interactive whiteboard, etc.). 59 Copyright (c) IARIA, 2014. ISBN: 978-1-61208-342-1 CONTENT 2014 : The Sixth International Conference on Creative Content Technologies