FireNet

FireNet is an open ML training dataset for visual recognition of fire safety equipment. Our dataset directly links the objects to their respective Uniclass, the construction sector’s classification scheme to name objects.

FireNet has been designed as a ML training dataset for experimentation and therefore fulfills multiple machine learning scenarios (classification, object detection, semantic segmentation). It is intended as a domain specific dataset to refine pre-trained standard architectures. Uniclass & Description

Pr_40_50_28_64 Portable fire extinguishers

Pr_75_75_30_50 Manual call points

Pr_40_50_28_29 Fire protective blankets

Pr_40_10_77_32 Fire escape route signs

Pr_40_10_77_31 Fire equipment signs

Pr_75_75_30_97 Visual alarm signal devices Pr_75_75_30_30 Fire alarm sounders

Pr_75_75_30_65 Point smoke detectors

In a perspective of transparency and to ensure the coherence of government action in the field of artificial intelligence (AI), the Minister of Cybersecurity and Digital Technology issued, on February 28, 2024, decree number 2024-01 concerning information resource requirements with regard to the use of artificial intelligence by public organizations. This new obligation allowed the Ministry of Cybersecurity and Digital Affairs (MCN) to document the use cases of AI in public administration and to establish a portrait of them. The information contained in the file therefore comes from a collection carried out from public bodies. The portrait shows all the initiatives that are currently under development or in production within public organizations. However, for obvious security reasons, cybersecurity initiatives are excluded. For the same reason, the commercial name of solutions and solution providers has been changed to a generic name. The data contains the name of the IT asset, project or AI initiative, its category, the responsible public body, the ministerial portfolio to which the organization is attached, the associated benefits and its status. The categories are: * Decision Support, Planning, and/or Prediction: AI system used primarily as a tool to support decision, planning, or to make predictions. * Behavior/Feeling Analysis: AI system used primarily to do behavior analysis or for the analysis of feelings. * Assistant/Conversational Agent: AI system using natural language communication and primarily used to assist with specific tasks or to maintain a conversation. with a user* Automation: AI system used primarily for the automation of defined and targeted tasks and/or processes. * User experience - Personalization: AI system used primarily for improving the user experience on platforms or websites (example: facilitating searches, customizing content, etc.). * AI-assisted training and learning, etc.). * AI-assisted training and learning: AI system used primarily for training and learning assistance. * Geomatic and geospatial management: System of AI used for forest, geological, cartographic and geomatics management. * Laboratory and equipment: Initiative or project aimed at the establishment of an experimental laboratory and/or the acquisition of computer equipment dedicated to the development of AI. * Connected electronic system and ambient intelligence: AI system integrated into initiatives or projects aimed at the implementation of an intelligent electronic system, ambient intelligence or connected objects. * Image processing: AI system used mainly for treatment and the image analysis. The statuses are: * Solution development: This status includes projects or initiatives that are in the development/acquisition stage or in the testing and validation phase of the development cycle of an AI system. * Solution in production: This status includes projects or initiatives that are in the deployment/integration stage or in the maintenance and support of the development cycle of an AI system. * Information unavailable: This status indicates that the information could not be confirmed by the body responsible for the project or initiative.