Patent information
IP India · live recordAbstract
Embodiments of the present disclosure relate to a method and an apparatus (103) for alerting threats to users (101). The apparatus (103) may capture a plurality of signals including at least one of Electro-Magnetic (E-M) signals and sound signals. The E-M signal and sound signals are used to detect objects around the user (101). A threat to the user (101) is predicted based on the objects around the user (101) and one or more alerts are generated such that the user (101) avoids the threat. The prediction of the threat enables the user (101) to take an action even before the threat has occurred. Also, the alerts are generated based on the prediction such that the user (101) can avoid the threat well in advance of the occurrence of the threat.
Applicants
1Inventors
3Specification
Show full textCollapse
1. A method of alerting threats to users (101), comprising: receiving, by an apparatus (103) associated with a user (101), a plurality of signals comprising at least one of Electro-Magnetic (E-M) signals and sound signals, from one or more sensors (204) associated with the apparatus (103), wherein the plurality signals are related to an environment (100) around the user (101); detecting, by the apparatus (103), one or more objects around the user (101) using the plurality of signals; predicting, by the apparatus (103), a threat to the user (101) caused by the one or more objects around the user (101) based on at least one of, one or more rules and an Artificial Model (AI) model; and generating, by the apparatus (103), one or more alerts based on the predicted threat posed by the one or more objects, wherein the one or more alerts are provided to the user (101) to enable the user (101) to avoid the threat. 3. The method as claimed in claim 1, wherein the plurality of signals are time series data. 4. The method as claimed in claim 1, further comprises classifying the one or more objects into one or more categories using a Convolution Neural Network (CNN) classifier. 5. The method as claimed in claim 1, further comprises: determining a temporal difference in the time series data; determining a variation in a relative position of the user (101) with respect to the one or more objects in the environment (100) using the temporal difference in the time series data; and estimating an impact caused by the one or more objects to the user (101) based on the variation in the relative position of the user (101) with respect to the one or more objects and the one or more categories of the one or more objects, wherein the estimated impact is used to predict the threat to the user (101). 6. The method as claimed in claim 1, wherein the AI model is trained to determine a threat, wherein training the AI model comprises:
providing the one or more rules and a plurality of training signals as input to the AI model, wherein each of the plurality of signals are associated with a threat based on the one or more rules; providing a plurality of testing signals after training the AI model; and evaluating a threat output by the AI model for the provided plurality of testing signals, based on the one or more rules, wherein evaluating comprises at least one of adjusting one of, one or more weights associated with the one or more rules and a bias of the AI model. 7. The method as claimed in claim 1, wherein generating the one or more alerts comprises: determining simultaneously when the one or more objects are detected, historical alerts that were generated when historical objects similar to the one or more objects were detected; generating the one or more alerts based on the determined historical alerts and the detected one or more objects, wherein the one or more alerts are provided to the user (101) to caution the user (101) of a possible threat; and updating the one or more alerts based on the predicted threat to the user (101), wherein the updated one or more alerts are provided to the user (101) to enable the user (101) to avoid the threat. 8. The method as claimed in claim 1, wherein the one or more alerts are generated further based on a severity of the threat, wherein the severity of the threat is determined based on the one or more categories of the one or more objects and an estimated impact caused by the one or more objects to the user (101) based on the variation in the relative position of the user (101) with respect to the one or more objects. 9. An apparatus (103) for alerting threats to users, comprising: a processor (203); and a memory (202), communicatively connected to the processor (203), storing processor executable instructions, which, in execution causes the processor (203) to: receive a plurality of signals comprising at least one of Electro-Magnetic (E-M) signals and sound signals, from one or more sensors (204) associated with the apparatus (103), wherein the plurality signals are related to an environment (100) around the user (101); detect one or more objects around the user (101) using the plurality of signals; predict a threat to the user (101) caused by the one or more objects around the user (101) based on at least one of, one or more rules and an Artificial Model (AI) model; and
generate one or more alerts based on the predicted threat posed by the one or more objects, wherein the one or more alerts are provided to the user (101) to enable the user (101) to avoid the threat. 10. The apparatus (103) as claimed in claim 9, wherein the processor (203) is configured to train AI model is to determine a threat, wherein processor is configured to: provide the one or more rules and a plurality of training signals as input to the AI model, wherein each of the plurality of signals are associated with a threat based on the one or more rules; provide a plurality of testing signals after training the AI model; and evaluate a threat output by the AI model for the provided plurality of testing signals, based on the one or more rules, wherein evaluating comprises at least one of adjusting one of, one or more weights associated with the one or more rules and a bias of the AI model.