Differentiating abandoned and removed object using temporal edge information

摘要

Disclosed is a computer-implemented method for classifying a detected region of change of a video frame as indicating an abandoned object or a removed object, the detected region of change is classified as an removed object if the temporal change of the edge consistency is from a more consistent state to a less consistent state, and the detected region of change is classified as an abandoned object if the temporal change of the edge consistency if from a less consistent to a more consistent state.

主权项

1. A computer-implemented method for classifying a detected region of change of a video frame as one of an abandoned object and a removed object, wherein a plurality of boundary blocks define a boundary of said detected region of change, said method comprising the steps of:
(1) for each one of the plurality of said boundary blocks:
(a) determining a first difference between a predicted edge orientation and an observed edge orientation of said boundary block;(b) determining a second difference between a historical edge orientation of said boundary block and at least one of the predicted edge orientation and the observed edge orientation of said boundary block; and(c) determining a temporal change of edge consistency based on the first and second difference; and (2) classifying said detected region of change as one of the abandoned object and the removed object based on one or more of the plurality of determined temporal changes of edge consistency, wherein the classifying step comprises, in regard to each one of the set of boundary blocks, the steps of:
assigning a first block score to the boundary block for the removed object if the determined temporal change of edge consistency of the boundary block indicates a transition from a more aligned state to a less aligned state;assigning a second block score to the boundary block for the abandoned object if the determined temporal change of edge consistency of the boundary block indicates a transition from a less aligned state to a more aligned state; anddetermining said overall score for said region of change by determining a sum of said individual block scores and weighting said individual block scores using said observed edge orientation of each one of said boundary blocks.