Vehicular access control based on virtual inductive loop

Patent metadata

Field	Value
Publication number	US-12254697-B2
Application number	US-202217863075-A
Country	US
Kind code	B2
Filing date	Jul 12, 2022
Priority date	Jul 28, 2021
Publication date	Mar 18, 2025
Grant date	Mar 18, 2025

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for monitoring events using a Virtual Inductive Loop system. In some implementations, image data is obtained from cameras. A region depicted in the obtained image data is identified, the region comprising lines spaced by a distance that satisfies a distance threshold. For each line included in the region: an object depicted crossing the line is determined whether to satisfy a height criteria indicating that the line is activated. In response to determining that an object depicted crossing the line satisfies the height criteria, an event is determined to have likely occurred using data indicating (i) which lines of the lines were activated and (ii) an order in which each of the lines were activated. In response to determining that an event likely occurred, actions are performed using at least some of the data.

First claim

The invention claimed is:

1. A computer-implemented method comprising:
maintaining, for a field of view of a camera, a region that is in the field of view and comprises a plurality of lines;
for each of one or more lines in the plurality of lines included in the region:
determining whether an object depicted in a sequence of images captured by the camera both intersects the line and has a height that satisfies height criteria, the intersection of the line by the object having a height that satisfies the height criteria indicating that the line is activated;

in response to determining, for at least one of the lines in the plurality of lines, that an object depicted in the sequence of images both intersects the line and has a height that satisfies the height criteria:
determining data indicating which lines of the plurality of lines were activated;
detecting an order in which at least one of the lines were activated using times at which the object intersected at least one of the lines from the plurality of lines; and
determining, from a plurality of event types, a first event type that likely occurred using (i) the data indicating which lines of the plurality of lines were activated and (ii) the order in which the at least one of the lines were activated;

in response to determining the first event type that likely occurred, selecting, from a plurality of actions each of which correspond to an event type from the plurality of event types, one or more actions for the first event type; and
performing the one or more actions using at least some of the data.

2. The computer-implemented method of claim 1 , further comprising:
prior to maintaining the region that is in the field of view, performing calibration of the camera, comprising:
identifying a first mat and a second mat depicted in the region in the field of view, the first and second mat displaying a visible pattern that aid the camera to detect a location of the first mat and the second mat;
detecting the visible pattern on the first mat and the second mat in the field of view;
estimating physical locations of the visible pattern on the first mat and the second mat using the detected visible pattern;
in response to estimating the physical locations of the visible pattern, determining camera parameters of the camera using the estimated physical locations of the visible pattern on the first mat and the second mat, wherein the camera parameters comprise one or more of a focal length, a tilt angle, and a roll angle;
obtaining second image data from the camera; and
validating, using the obtained second image data, the camera parameters and the physical locations of the visible pattern.

3. The computer-implemented method of claim 1 , wherein maintaining the region that is in the field of view further comprises:
obtaining initial image data from the camera;
projecting one or more identifiers in the initial image data;
determining, for each of the one or more projected identifiers in the initial image data, a location of the region in which the identifier is depicted;
generating, for each of the locations of the one or more projected identifiers in the region, an initial line of a plurality of initial lines in the initial image data with respect to the locations of the one or more projected identifiers at the region;
determining a respective projected ground level plane in the initial image data; and
determining a first line from the plurality of lines by projecting the initial line of the plurality of initial lines onto the respective projected ground level plane in the region in the image data.

4. The computer-implemented method of claim 1 , wherein:
determining the lines that were activated based on whether the object that satisfies the height criteria is detected in the lines further comprises:
identifying (i) one or more first distinctive features along a first image plane of one or more lines of the plurality of lines from a first camera of two or more cameras and (ii) one or more second distinctive features along a second image plane of one or more lines of the plurality of lines from a second camera of the two or more cameras;
matching the one or more first distinctive features along the first image plane from the first camera to at least some of the one or more second distinctive features along the second image plane from the second camera;
estimating a plurality of cross sectional outlines of an object that intersected with the plurality of lines using the matched first distinctive features and the matched second distinctive features, wherein the plurality of cross sectional outlines comprise a height and a width of the object that intersected with the plurality of lines;
determining, for each cross sectional outline of the plurality of cross sectional outlines of the object, whether the cross sectional outline satisfies the height criteria; and
in response to determining at least some of the cross sectional outlines of the object satisfy the height criteria, the method comprises identifying the lines of the plurality of lines intersected by the object as activated.

5. The computer-implemented method of claim 4 , further comprising:
in response to determining at least some of the cross sectional outlines of the object satisfy the height criteria, identifying a type of the object that intersected with the lines that were activated;
in response to identifying the type of the object that intersected with the lines that were activated:
determining, using the type of the object, whether to open a security gate; and
in response to determining to open the security gate, transmitting a notification to a security gate causing the security gate to open that allows passage of the object; or
in response to determining to not open the security gate, transmitting a notification to the security gate causing the security gate to remain closed to prevent passage of the object.

6. The computer-implemented method of claim 4 , wherein identifying the one or more first and second distinctive features comprises one or more of identifying a change in light intensity along the first image plane and the second image plane, a change in color along the first image plane and the second image plane, a predetermined length of a particular intensity along the first image plane and the second image plane, and a predetermined length of a particular color along the first image plane and the second image plane.

7. The computer-implemented method of claim 4 , further comprising determining a three-dimensional outline of the object by analyzing a speed of the object that intersected the lines that were activated, wherein determining whether the event likely occurred uses the three-dimensional outline of the object.

8. The computer-implemented method of claim 7 , wherein analyzing the speed of the object further comprises:
determining an amount of time taken by the object to move between a first line and a subsequent second line of the lines that were activated;
determining a distance between each cross sectional outline of the plurality of cross sectional outlines;
determining the speed of the object using (i) the amount of time taken by the object to move between the first line and the subsequent second line of the lines that were activated and (ii) the distance between each cross sectional outline of the plurality of cross sectional outlines; and
determining the three-dimensional outline of the object using the distance between each cross sectional outline of the plurality of cross sectional outlines and the speed of the object.

9. The computer-implemented method of claim 1 , wherein determining the first event type that likely occurred using (i) the data indicating wh

Assignees

• Objectvideo Labs Llc

Inventors

• Ramanathan Narayanan
• Beach Allison
• Hastings Robert S
• Yin Weihong
• Taheri Sima
• Eubanks Dana
• Park Kyoung-Jin
• Madden Donald Gerard
• Qian Gang

Classifications

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title

• G06V10/776

  Validation; Performance evaluation · CPC title

• G06V2201/08

  Detecting or categorising vehicles · CPC title

• G06T2207/30232

  Surveillance · CPC title

• E01F13/04

  movable to allow or prevent passage {(E01F13/12, E01F15/12 take precedence; removing part of barrier for occasional passage E01F13/024; gates for allowing passage through fences E06B11/02)} · CPC title