Abstract: A computer system determines a likelihood of the presence of flame in a scene. The system obtains a series of digital infrared and optical images of the scene; identifies a candidate region in a location of the scene based on analysis of the infrared images; identifies an optical image slice based on analysis of the optical images and the location of the candidate region; and determines a likelihood of the presence of flame in the scene based on analysis of the optical image slice. Analysis of the infrared images includes detecting a high-temperature image region that exceeds a threshold temperature; detecting a turbulent motion image region; and determining whether the turbulent motion region is within a specified proximity of or overlaps with the high-temperature region. The optical image slice may be provided to a trained neural network, which returns a degree-of-confidence value that indicates whether flame is present.

Abstract: A computer system determines a likelihood of the presence of flame in a scene. The system obtains a series of digital infrared and optical images of the scene; identifies a candidate region in a location of the scene based on analysis of the infrared images; identifies an optical image slice based on analysis of the optical images and the location of the candidate region; and determines a likelihood of the presence of flame in the scene based on analysis of the optical image slice. Analysis of the infrared images includes detecting a high-temperature image region that exceeds a threshold temperature; detecting a turbulent motion image region; and determining whether the turbulent motion region is within a specified proximity of or overlaps with the high-temperature region. The optical image slice may be provided to a trained neural network, which returns a degree-of-confidence value that indicates whether flame is present.

Abstract: The present invention is directed to a facility for distributing network security information. The facility receives network security information and recipient selection information specifying a characteristic of perspective recipients to be used in selecting recipients for the security information. The facility then compares the received recipient selection information to each of a plurality of perspective recipient profiles. Each perspective recipient profile corresponds to one or more perspective recipients and indicates one or more characteristics of the perspective recipients relating to the receipt of network security information. Based upon this comparison, the facility selects at least a portion of the plurality of perspective recipients as recipients of the network security information, and addresses the network security information to each of the selected recipients.

Abstract: The present invention is directed to a facility for adapting a network security policy model for use in a particular network. The facility retrieves the network security policy model, which comprises network security rules each specified with respect to one or more aliases. Each alias represents a role in a network for one or more network elements. The facility receives, for each alias included in the network security policy model, a list of one or more network elements in the network serving the role represented by the alias. The facility replaces each alias in the network security policy model with the received list of network security devices specified for the alias to produce a network security policy adapted for use in a network.