| CPC G06V 20/44 (2022.01) [G06V 10/82 (2022.01); G06V 20/17 (2022.01); G06V 20/53 (2022.01); G06V 40/172 (2022.01); G06V 40/20 (2022.01); H04L 9/008 (2013.01)] | 17 Claims |
|
1. A system to proactively detect in real time at least one threats in crowded areas, the system comprising:
a processor; and
a memory communicatively coupled to the processor, wherein the memory stores processor-executable instructions, which, on execution, causes the processor to:
capture in real time a plurality of images selected from a set of image types, the set of image types consisting of still images and a plurality of video feed, using a plurality of imaging sensors within a pre-defined area, wherein within the pre-defined area a plurality of individuals is present;
encrypt the plurality of images and the plurality of video feed using advanced encryption techniques to protect individual identities;
identify at least one threats based on results of at least three identification tests performed, wherein identification of the at least one threats requires at least one identification test to have a positive result and said tests are selected from a set of tests consisting of:
identifying at least one suspected individuals from the plurality of images using a machine learning model;
identifying at least one suspicious objects within the at least one of the plurality of images; and
identifying at least one behavioral patterns associated with at least one individuals using the machine learning model after further classifying the at least one behavioral patterns into one of a suspicious behavior and a non-suspicious behavior;
determine a precise location of the identified at least one threats;
perform facial recognition on the encrypted plurality of images and the encrypted plurality of video feed;
generate anonymized or de-identified facial recognition results while preserving privacy;
output recognized individuals' identities for authorized purposes, ensuring privacy compliance in crowded areas; and
transmit a notification of the identified at least one threats to the at least one authorities, wherein the notification contains at least one type of information selected from a set of information types consisting of: the precise location of the threat, the identified at least one suspected individuals, the identified at least one suspicious objects, the identified at least one behavioral patterns, and imagery from the plurality of images
wherein the machine learning model is configured to dynamically adjust at least one facial recognition parameters based on complexity of the crowded areas, presence of at least one occlusions, and a level of background noise in the plurality of images and the plurality of video feed.
|