CPC G06Q 30/0633 (2013.01) [G06F 18/2431 (2023.01); G06T 7/20 (2013.01); G06T 7/70 (2017.01); G06V 20/52 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/20132 (2013.01); G06T 2207/30232 (2013.01)] | 17 Claims |
1. A method of detecting a cart-based loss incident in a retail store, the method comprising:
decoding one or more video frames of a video stream to obtain one or more motion vectors therefrom,
wherein the video stream includes an H.264 video stream encoded using the one or more motion vectors;
filtering the one or more motion vectors according to a position and an orientation of each motion vector from the one or more motion vectors to extract one or more substantially horizontal orientated motion vectors located in a pre-defined region of a cash register lane;
comparing a sum of intensity values of the extracted substantially horizontal orientated motion vectors with a threshold value;
detecting motion of a shopping cart within the pre-defined region of the cash register lane bounded by a pre-defined tracking start point and a pre-defined tracking end point when the sum of intensity values exceeds the threshold value;
tracking a location of the shopping cart till the shopping cart reaches the pre-defined tracking end point, upon detection of motion of the shopping cart;
dynamically classifying the shopping cart in one of a plurality of classification statuses, based on recognition of one or more items present in the shopping cart, till the shopping cart reaches the pre-defined tracking end point; and
generating an alert signal, when the shopping cart is classified in a pre-defined classification status from the plurality of classification statuses, at an alert threshold point between the pre-defined tracking start and end points.
|