Computer-implemented method of handling an emergency incident, communication network, and emergency processing unit
A computer-implemented method of handling an emergency incident can include receiving information on an emergency incident that includes at least one image of the emergency incident, applying a Convolutional Neural Network (CNN) object recognition and classification process for identifying and marking objects on the at least one image that are related to the emergency incident and that may cause at least one secondary hazardous situation, processing the data relating to the identified and marked objects by applying a deep learning algorithm to the data in a Relational Network (RN) architecture, wherein the image on the basis of the identified and marked objects is correlated to a set of recognized objects in a database for classifying the emergency. A communication network, communication apparatus, and an emergency processing unit are also provided. Embodiments of such machines and systems can be configured to implement embodiments of the method.
1 . A computer-implemented method of handling an emergency incident reported to a Public Safety Answering Point PSAP the method comprising:
receiving information on the reported emergency incident, the information comprising at least one image of the reported emergency incident showing the emergency incident as an unknown primary hazardous situation,
applying a Convolutional Neural Network, CNN, object recognition and classification process for identifying and marking objects on the at least one image, the identified and marked objects being part of a defined set of identifiable emergency objects;
processing data relating to the identified and marked objects by applying a deep learning algorithm to the data in a Relational Network (RN) architecture designed to attain an understanding of hazardous relations between emergency objects, the hazardous relations causing at least one secondary hazardous situation, to infer hazardous relations between the identified and marked objects; and
outputting an alert in response to verification that the at least one secondary situation may occur.
2 . The method of claim 1 , wherein the RN has a Long-Short-Term-Memory (SLTM) architecture.
3 . The method of claim 1 , comprising applying image processing to the at least one image using image correction software run by an emergency processing unit that is communicatively connectable to the PSAP.
4 . The method of claim 1 , wherein the CNN is applied based on convolution and max pooling.
5 . The method of claim 1 , wherein the primary hazardous situation is a fire the at least one secondary hazardous situation includes inflammable materials.
6 . The method of claim 1 , further comprising providing a suggestion for handling the primary hazardous situation and a suggestion for handling the at least one secondary hazardous situation.
7 . A communication apparatus for handling emergency incidents, the communication apparatus comprising:
an emergency processing unit communicatively connectable to an Emergency Service Network (ESINET) via which information on emergency incidents are exchangeable with a Public Safety Answering Point (PSAP), the emergency processing unit comprising a processor connected to a non-transitory computer readable medium having code stored thereon, the code defining a method that is performed by the emergency processing unit when the processor runs the code, the method comprising:
receiving via the ESINET information on a reported emergency incident, the information comprising at least one image of the emergency incident showing the emergency incident as an unknown primary hazardous situation,
applying a Convolutional Neural Network, CNN, object recognition and classification process for identifying and marking objects on the at least one image, the identified and marked objects being part of a defined set of identifiable emergency objects;
processing data relating to the identified and marked objects by applying a deep learning algorithm to the data in a Relational Network (RN) architecture designed to attain an understanding of hazardous relations between emergency objects, the hazardous relations causing at least one secondary hazardous situation, to infer hazardous relations between the identified and marked objects.
8 . The communication apparatus of claim 7 , wherein the emergency processing unit comprises a deep learning unit and an image correction unit.
9 . The communication apparatus of claim 7 , wherein the emergency processing unit comprises a deep learning unit.
10 . The communication apparatus of claim 7 , wherein the emergency processing unit comprises an image correction unit.
11 . The communication apparatus of claim 7 , further comprising the ESINET and the PSAP.