Methods and systems for reducing a risk of spread of disease among people in a space

Patent metadata

Field	Value
Publication number	US-12142385-B2
Application number	US-202117328356-A
Country	US
Kind code	B2
Filing date	May 24, 2021
Priority date	Jun 22, 2020
Publication date	Nov 12, 2024
Grant date	Nov 12, 2024

Abstract

Methods and systems for location tracking or maintaining a count of people in a building or space. An illustrative method may include storing a background image of a field of view of a video camera and receiving a video stream from the video camera. Background subtraction may be performed to identify one or more blobs in the field of view of the video camera. The size of the one or more blobs may be compared to an expected size of the blob at a similar distance from the camera. When the size of the blob is greater than the expected size of a person at the determined distance of the corresponding blob by more than a predetermined threshold the blob may be counted as two or more people.

First claim

What is claimed is:

1. A method for counting a number of people in a space of a building, the method comprising:
storing a background image of a perspective field of view captured by a non-overhead video camera;
receiving a video stream from the non-overhead video camera;
calibrating the perspective field of view captured by the non-overhead video camera, including:
determining a number of pixels of the non-overhead video camera that correspond to a known-sized object at a known location at a near end of the perspective field of view of the non-overhead video camera, and a number of pixels of the non-overhead video camera that correspond to a known-sized object at a known location at a far end of the perspective field of view of the non-overhead video camera;
generating a calibration map that maps real world distances in the perspective field of view of the non-overhead video camera against x-and y-pixel positions of the non-overhead video camera based at least in part on the number of pixels of the non-overhead video camera that correspond to the known-sized object at the known location at the near end of the perspective field of view of the non-overhead video camera and the number of pixels of the non-overhead video camera that correspond to the known-sized object at the known location at the far end of the perspective field of view of the non-overhead video camera;

subtracting the background image from each frame of the video stream to identify one or more blobs in the perspective field of view of the non-overhead video camera;
determining a real-world distance between the non-overhead video camera and each of the one or more blobs based at least in part on the calibration map;
comparing a size of each of the one or more blobs to an expected size of a person at the determined distance of the corresponding blob;
when the size of the blob is greater than the expected size of a person at the determined distance of the corresponding blob by more than a factor of at least 1.5 times, counting the blob as two or more people, otherwise counting the blob as one person or no person; and
determining a count of the number people in the perspective field of view of the non-overhead video camera based at least in part on the count of people assigned to each of the one or more blobs.

2. The method of claim 1 , wherein when the size of the blob is not greater than at least 1.5 times the expected size of a person at the determined distance, using deep learning to determine whether to count the blob as one person or no person.

3. The method of claim 2 , wherein the perspective field of view of the non-overhead video camera covers only part of the space of the building, and wherein a perspective field of view of one or more other non-overhead video cameras cover one or more other parts of the space, the method further comprising:
determining a count of the number people in the perspective field of view of each of the one or more other non-overhead video cameras; and
aggregating the counts of the number of people from all of the non-overhead video cameras that have a perspective field of view that covers part of the space to identify a total count of the number of people in the space.

4. The method of claim 3 , further comprising updating the total count of the number of people in the space at predetermined time intervals to obtain a plurality of total counts over a period of time.

5. The method of claim 4 , further comprising determining a rate of change of the total count of the number of people in the space.

6. The method of claim 5 , further comprising displaying a map of the one or more spaces of the building and shading each of the one or more spaces of the map based on the determined rate of change the total count of the number of people in the corresponding space.

7. A method for counting a number of people in a space of a building, the method comprising:
storing a background image of a field of view of each of two or more video cameras;
calibrating a field of view of each of the two or more video cameras, where for each of the two or more video cameras:
determining a number of pixels of the respective video camera that correspond to a known-sized object at a known location at a near end of the field of view of the respective video camera, and a number of pixels of the respective video camera that correspond to a known-sized object at a known location at a far end of the field of view of the respective video camera;
generating a calibration map that maps real world distances in the field of view of the respective video camera based at least in part on the number of pixels of the respective video camera that correspond to the known-sized object at the known location at the near end of the field of view of the respective video camera and the number of pixels of the respective video camera that correspond to the known-sized object at the known location at the far end of the field of view of the respective video camera;

receiving a video stream from each of the two or more video cameras;
subtracting the background image from one or more frames of the respective video stream to identify one or more blobs in the respective field of view of each of the two or more video cameras;
determining a distance between the respective video camera of the two or more video cameras and each of the one or more blobs that are associated with the respective video camera based at least in part on the corresponding calibration map;
comparing a size of each of the one or more blobs to an expected size of a person at the determined distance of the corresponding blob;
when the size of the blob is greater than the expected size of a person at the determined distance of the corresponding blob by more than a predetermined factor that is greater than one, counting the blob as two or more people, otherwise counting the blob as one person or no person; and
determining a count of the number people in the respective field of view of each of the two or more video cameras; and
aggregating the counts of the number of people from all of the video cameras that have a field of view that covers part of the space to identify a total count of the number of people in the space.

8. The method of claim 7 , further comprising updating the total count of the number of people in the space at predetermined time intervals to obtain a plurality of total counts over a period of time.

9. The method of claim 8 , further comprising determining a rate of change of the total count of the number of people in the space.

10. The method of claim 9 , further comprising displaying a map of the one or more spaces of the building and shading each of the one or more spaces of the map based on the determined rate of change the total count of the number of people in the corresponding space.

Assignees

• Honeywell Int Inc

Inventors

• Dhamija Vijay
• Eswara Lalitha M
• Shaik Ismaiel
• Balasubramanian Mourian
• Jain Abhisekh
• Sonkamble Siddharth
• Chaurasia Jitendra
• Thomas Manu

Classifications

• G06V20/44

  Event detection · CPC title

• G06V40/20

  Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16) · CPC title

• G06V20/53

  Recognition of crowd images, e.g. recognition of crowd congestion · CPC title

• G06V20/41

  Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items (segmenting video sequences G06V20/49) · CPC title

• G16H40/20

  for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms · CPC title