Event detection using pattern recognition criteria
Computer-implemented systems utilizing sensor networks for sensing temperature and motion environmental parameters, and performing at least operations of electronically establishing, based on pattern recognition criteria, correspondence of a plurality of representative features a plurality of characteristics of an occurrence, where a first instance of the occurrence occurred within a first time period of a plurality of time periods; electronically discovering, based on the correspondence, a second instance of the occurrence in an environment during a second time period of the plurality of time periods; and electronically causing, based on the discovery of the second instance of the occurrence, a change in the environment via an electronically-controlled device.
1. A computer-implemented method, comprising:
receiving sensed data of at least one parameter from a sensor node over a network, wherein the network comprises a plurality of remotely located sensor nodes,
wherein each of the plurality of remotely located sensor nodes captures environmental data;
receiving an input selection of a target-event having at least one representative feature;
selecting a pattern recognition criteria corresponding to the at least one representative feature of the target-event, the selected pattern recognition criteria comprising a chronological sequence of sensor data collected over time;
searching, in response to the input selection corresponding to the target-event, for sensor data correlating to the at least one representative feature using the selected pattern recognition criteria;
determining that the sensor data correlating to the at least one representative feature is found; and
based on the determination that the sensor data correlating to the at least one representative feature is found, providing, to a recipient, an output indicative of a result of the searching,
wherein determining that the sensor data correlating to the at least one representative feature is found comprises:
determining that detection of the target-event corresponds to a degraded correlating event-data identifier; and
based on the determination that the detection of the target-event corresponds to the degraded correlating event-data identifier:
labelling the detection of the target-event with the degraded correlating event-identifier; and
determining that the sensor data correlating to the at least one representative feature is found based on multiple instances of degraded correlating event data,
wherein the target-event is a gunshot; and
wherein determining that the sensor data correlating to the at least one representative feature is found based on multiple instances of degraded correlating event data comprises detecting the gunshot based on a time-frequency analysis of at least a portion of the degraded correlating event data.
2. The method of claim 1 , wherein the chronological sequence of sensor data collected over time comprises a chronological sequence of acoustic data over time.
3. A computer-implemented method, comprising:
receiving sensed data of at least one parameter from a sensor node over a network, wherein the network comprises a plurality of remotely located sensor nodes,
wherein each of the plurality of remotely located sensor nodes captures environmental data;
receiving an input selection of a target-event having at least one representative feature;
selecting a pattern recognition criteria corresponding to the at least one representative feature of the target-event, the selected pattern recognition criteria comprising a chronological sequence of sensor data collected over time;
searching, in response to the input selection corresponding to the target-event, for sensor data correlating to the at least one representative feature using the selected pattern recognition criteria;
determining that the sensor data correlating to the at least one representative feature is found; and
based on the determination that the sensor data correlating to the at least one representative feature is found, providing, to a recipient, an output indicative of a result of the searching,
wherein the chronological sequence of sensor data collected over time comprises a chronological sequence of acoustic data over time, and
wherein determining that the sensor data correlating to the at least one representative feature is found comprises determining that the sensor data indicates passage of an emergency vehicle through an intersection, including a time of passage by:
detecting that the chronological sequence of acoustic data over time includes only two frequencies; and
detecting a Doppler shift in the two frequencies on the passage of the emergency vehicle through the intersection,
wherein the two frequencies comprise a first frequency and a second frequency; and
wherein determining that the sensor data indicates passage of the emergency vehicle through the intersection comprises detecting that, in the chronological sequence of acoustic data over time, the first frequency lasts for a certain number of cycles before the chronological sequence of acoustic data over time changes to the second frequency and the second frequency lasts for the certain number of cycles before the chronological sequence of acoustic data over time changes back to the first frequency.
4. The method of claim 1 , wherein searching for sensor data correlating to the at least one representative feature using the selected pattern recognition criteria comprises image processing.
5. The method of claim 1 , wherein searching for sensor data correlating to the at least one representative feature using the selected pattern recognition criteria comprises searching using fuzzy logic.
6. The method of claim 1 , wherein searching for sensor data correlating to the at least one representative feature using the selected pattern recognition criteria comprises searching using an artificial neural network.
7. The method of claim 1 :
wherein the at least one representative feature of the target-event comprises a selected frequency pattern; and
wherein searching for sensor data correlating to the at least one representative feature using the selected pattern recognition criteria comprises searching for sensor data with the selected frequency pattern.
8. The method of claim 1 , wherein determining that the sensor data correlating to the at least one representative feature is found comprises:
determining whether detection of the target-event is tentative based on the determination that the sensor data correlating to the at least one representative feature is found; and
based on a determination that detection of the target-event is tentative, labelling the detection of the target-event with a correlating tentative event-identifier.
9. The method of claim 1 , wherein determining that the sensor data correlating to the at least one representative feature is found based on multiple instances of degraded correlating event data comprises determining that the sensor data correlating to the at least one representative feature is found based on a majority of the multiple instances of degraded correlating event data relating to the at least one representative feature.
10. The method of claim 9 , wherein at least one of the multiple instances of degraded correlating event data does not relate to the at least one representative feature.
11. The method of claim 1 , wherein determining that the sensor data correlating to the at least one representative feature is found based on multiple instances of degraded correlating event data comprises determining that the sensor data correlating to the at least one representative feature is found based on at least one of the multiple instances of degraded correlating event data relating to the at least one representative feature and at least one of the multiple instances of degraded correlating event data not relating to the at least one representative feature.
12. The method of claim 1 , wherein determining that detection of the target-event corresponds to the degraded correlating event-data identifier comprises determining that detection of the target-event includes sufficient data for a recipient to make a preliminary determination of whether the degraded correlating event data appears to include the at least one representative feature.
13. The method of claim 3 , wherein determining that the sensor data indicates passage of the emergency vehicle through the intersection comprises detecting that, in the chronological sequence of acoustic data over time, amplitude of the acoustic data is generally the same over the time.
14. The method of claim 3 :
wherein determining that the sensor data indicates passage of the emergency vehicle through the intersection comprises detecting that, in the chronological sequence of acoustic data over time, amplitude of the acoustic data is generally the same over the time.