Provided are systems and methods for smoke and fire detection. An example method includes receiving two or more image frames that represent images of two or more areas, combining the two or more image frames to generate a composite image frame that represents an image of an area comprising the two or more areas, comparing the composite image frame to a background frame associated with the area to determine one or more pixels of the composite frame having a pixel value that is different from corresponding pixels of the background frame, determining a smoke frame, wherein pixels of the smoke frame that correspond to the one or more pixels of the composite frame comprise a pixel value associated with smoke, determining a smoke value corresponding to pixel values of the pixels of the smoke frame, determining that the smoke value exceeds a smoke threshold value, determining a smoke or fire condition based at least in part on determining that the smoke value exceeds a smoke threshold value, and providing an indication of the smoke or fire condition.

A solution for monitoring an area for the presence of a flame and/or a leak, such as from a pressurized fluid, is provided. An imaging device can be used that acquires image data based on electromagnetic radiation having wavelengths only corresponding to at least one region of the electromagnetic spectrum in which electromagnetic radiation from an ambient light source is less than the electromagnetic radiation emitted by at least one type of flame for which the presence within the area is being monitored. An acoustic device can be used that is configured to acquire acoustic data for the area and enhance acoustic signals in a range of frequencies corresponding to a leak of a pressurized fluid present in the area.

The invention relates to a machine learning system (10) which is configured, on the basis of a plurality of images captured in succession, to detect smoke (12) within the images. The machine learning system (10) comprises a convolutional recurrent neural network. The invention also relates to a method for detecting smoke by means of this machine learning system.

Embodiments of this disclosure provide a smoke detection method and apparatus. The apparatus includes a processor configured to detect foreground blobs in images of a plurality of frames in a video; determine motion areas of detected foreground blobs; determine a first degree of overlapping of the foreground blobs in images of at least two frames, and/or determine a second degree of overlapping of the foreground blobs and motion areas to which the foreground blobs correspond; determine interfering foreground blobs according to the first degree of overlapping and/or the second degree of overlapping; and extract features from motion areas of remaining foreground blobs with the interfering foreground blobs being removed, and detect smoke according to the features. With the embodiments of this disclosure, accuracy of smoke detection may be increased, and false detection resulted from such interfering objects as water blobs, or light spots, etc., may be avoided.

A device for characterizing a fire comprises at least one stereovision system and at least one processing unit. The at least one stereo vision system comprises a first and a second image capture unit. The at least one processing unit is configured to determine at least one geometric characteristic of the fire. The processing unit is also configured to determine a radiative flux of this fire on the basis of a calibrated linear relationship established between a radiative flux of a reference fire as a function of at least one of the geometric characteristics of the fire and of at least one fire category in order to be able to determine zones exposed to a radiative flux that exceeds a reference threshold. The present disclosure also relates to a method for determining radiative fluxes that implements such a fire characterizing device.

A method of detecting smoke or a fire involves the installation of properly produced cameras in the field. The cameras send recorded photos to a server. A computer downloads the photos from the server and uses attempts to automatically detect smoke with artificial intelligence software. Photos with a detected fire are sent to users which interface with a non-transitory computer readable medium, such as a desktop computer or a mobile phone, capable of executing a software application for verification. Verified smoke or fire threats signal alarms to the designated fire department in an effort to prevent fires from spreading and thus causing unnecessary damage to communities, lives, and ecosystems.

In one aspect, data characterizing a video feed acquired by a camera oriented toward and including a field of view of a forecourt of a fueling station can be received. The video feed can be monitored for hazards, and the monitoring of the video feed can include performing automatic hazard detection on the video feed using at least one predictive model that predicts a presence of a hazard within the forecourt of the fueling station. A command can be transmitted in response to the detecting of the presence of the hazard within the forecourt of the fueling station. Related apparatus, systems, methods, techniques, and articles are also described.

The invention relates to an intelligent evacuation system and method used in a subway station fire. A fire alarm module discovers a fire and transmits fire information, a fire source positioning module positions a fire source according to the collected picture information and smoke information of the fire, an indoor pedestrian positioning module positions pedestrians in a subway station; a relay control module is connected with the fire alarm module, the fire source positioning module and the indoor pedestrian positioning module, and the relay control module combines fire information, fire source positioning information and pedestrian positioning information; and a master control module is connected with the relay control module, and the master control module plans a dynamic escape route for each pedestrian according to the combined information, pedestrian information collected by a subway APP and pedestrian information collected by a map management system, and generates an escape route guide map.

Systems and methods are provided for automatically performing an action based on video content. One example method includes receiving, at a first computing device, a video and determining, with a content determination engine, content of the video. An action to perform at the first computing device and/or at a second computing device is generated, based on the content of the video. If the action is to be performed at the second computing device, the action is transmitted to the second computing device. The action is performed at the respective first and/or second computing device.

Virtual sensor technology, in which a camera is controlled to capture at least one configuration image of an area monitored by a monitoring system and input is received that defines one or more characteristics of the at least one configuration image that enable sensing of an event in the area. Based on the received input, configuration data used in sensing the event is generated and stored. After storage of the configuration data, the camera is controlled to capture one or more images of the area, the one or more images are analyzed based on the configuration data, and occurrence of the event is detected based on the analysis of the one or more images. Based on detecting occurrence of the event, a transmitting device is controlled to send, to a controller of the monitoring system, a signal that indicates the detection of the occurrence of the event.