Granted Patent B2

US 12,646,348 · App. 18/045,930 · Granted Jun 2, 2026

Inventors: Lukasz Luczak (Skawina, PL); Jakub Baranski (Cracow, PL); Grzegorz Szurek (Cracow, PL); Dominika Dubowik (Cracow, PL)

Assignee: MOTOROLA SOLUTIONS, INC.

G06V40/103G06F3/14G06T5/77G06V10/26G06T2207/30196G06T2207/30242G06V2201/07

Abstract

Claims (70)

1 . A method for detecting people in video data using non-visual sensor data, the method comprising:

receiving, at an electronic computing device, from a camera, video data captured by the camera during a time period;

analyzing, at the electronic computing device, the video data, using an electronic processor executing a video analytics engine comprising an object classifier that is trained to identify an instance of a person depicted in the video data, to detect a number of persons visible in a field of view (FOV) depicted by the video data;

determining, at the electronic computing device, a degree of obscuration of the FOV;

upon determining that the degree of obscuration is above a predefined threshold, transmitting an electronic request from the electronic computing device to a non-visual sensor to provide non-visual sensor data captured by the non-visual sensor data during the time period corresponding to the FOV depicted by the video data;

receiving, at the electronic computing device, from the non-visual sensor, non-visual sensor data that was captured by the non-visual sensor associated with the FOV during the time period;

detecting, at the electronic computing device, a feature in the non-visual sensor data that indicates that an additional person not detected with the video analytics engine was present in the FOV depicted by the video data during the time period; and

providing, at the electronic computing device, an electronic notification indicating that an additional person not detected with the video analytics engine was present in the FOV depicted by the video data during the time period.

2 . The method of claim 1 , further comprising:

applying a computer-vision technique to the video data to determine the degree of obscuration of the FOV.

3 . The method of claim 1 , further comprising:

obtaining additional sensor data captured by a smoke detector, a fog detector, or a light detector during the time period;

determining the degree of obscuration of the FOV based on additional sensor data.

4 . The method of claim 1 , further comprising:

identifying a location of the additional person based on the non-visual sensor data;

mapping the location of the additional person to a pixel range of the video data; and

inserting an indication of the pixel range into the electronic notification.

5 . The method of claim 4 , further comprising:

rendering, on an electronic display, an image that illustrates the FOV as captured by the camera during the time period; and

rendering, on the electronic display, a visual element that identifies the pixel range in the image.

6 . The method of claim 1 , wherein the non-visual sensor comprises at least one of: a Bluetooth Low-Energy (BLE) beacon, a weight sensor, a Global Positioning System (GPS), a Wireless Fidelity (WiFi) Positioning System (WPS), an Ultra-Wideband (UWB) positioning system, a laser perimeter sensor, a radar, a Light Detection and Ranging (LiDar) sensor, a fiber optic detection system, an electrostatic field disturbance sensor, a spot vibration sensor, a passive infrared (PIR) sensor, an active ultrasonic sensor, a temperature sensor, a microwave motion sensor, a window sensor, a door sensor, a capacitive sensor, or a seismic sensor.

7 . The method of claim 1 , wherein the non-visual sensor is included in an electronic device associated with an identifier uniquely identifying the additional person, the method further comprising:

retrieving the identifier uniquely identifying the additional person from the non-visual sensor data obtained from the non-visual sensor; and

inserting the identifier uniquely identifying the additional person into the electronic notification.

8 . The method of claim 7 , wherein the electronic device is a cellular phone, a Bluetooth tracker, or a Global Positioning System (GPS) tracker.

9 . An electronic computing device, comprising:

a communications unit; and

an electronic processor communicatively coupled to the communications unit, the electronic processor configured to:

receive, via the communications unit, from a camera, video data captured by the camera during a time period;

analyze, the video data, using an electronic processor executing a video analytics engine comprising an object classifier that is trained to identify an instance of a person depicted in the video data, to detect a number of persons visible in a field of view (FOV) depicted by the video data;

determine a degree of obscuration of the FOV;

transmit, in response to determining that the degree of obscuration is above a predefined condition, an electronic request, via the communications unit, to a non-visual sensor to provide non-visual sensor data captured by the non-visual sensor data during the time period corresponding to the FOV depicted by the video data;

receive, via the communications unit, from the non-visual sensor, non-visual sensor data that was captured by the non-visual sensor associated with the FOV during the time period;

detect a feature in the non-visual sensor data that indicates that an additional person not detected with the video analytics engine was present in the FOV depicted by the video data during the time period; and

provide an electronic notification indicating that an additional person not detected with the video analytics engine was present in the FOV depicted by the video data during the time period.

10 . The electronic computing device of claim 9 , wherein the electronic processor is configured to:

apply a computer-vision technique to the video data to determine the degree of obscuration of the FOV.

11 . The electronic computing device of claim 9 , wherein the electronic processor is configured to:

obtain additional sensor data captured by a smoke detector, a fog detector, or a light detector during the time period;

determine the degree of obscuration of the FOV based on the additional sensor data.

12 . The electronic computing device of claim 9 , wherein the electronic processor is configured to:

identify a location of the additional person based on the non-visual sensor data;

map the location of the additional person to a pixel range of the video data; and

insert an indication of the pixel range into the electronic notification.

13 . The electronic computing device of claim 12 , wherein the electronic processor is configured to:

render, on an electronic display, an image that illustrates the FOV as captured by the camera during the time period; and

render, on the electronic display, a visual element that identifies the pixel range in the image.

14 . The electronic computing device of claim 9 , wherein the non-visual sensor comprises at least one of: a Bluetooth Low-Energy (BLE) beacon, a weight sensor, a Global Positioning System (GPS), a Wireless Fidelity (WiFi) Positioning System (WPS), an Ultra-Wideband (UWB) positioning system, a laser perimeter sensor, a radar, a Light Detection and Ranging (LiDar) sensor, a fiber optic detection system, an electrostatic field disturbance sensor, a spot vibration sensor, a passive infrared (PIR) sensor, an active ultrasonic sensor, a temperature sensor, a microwave motion sensor, a window sensor, a door sensor, a capacitive sensor, or a seismic sensor.

15 . A method for detecting people in video data using non-visual sensor data, the method comprising:

receiving, at an electronic computing device, from another computing device, a redacted version of video data captured by a camera during a time period;

analyzing, at the electronic computing device, the redacted version of video data, using an electronic processor executing a video analytics engine comprising an object classifier that is trained to identify an instance of a person depicted in the redacted version of video data, to detect a number of persons who have been redacted from a field of view (FOV) depicted by the redacted version of the video data;

determining, at the electronic computing device, a degree of obscuration of the FOV;

upon determining that the degree of obscuration is above a predefined threshold, transmitting an electronic request from the electronic computing device to a non-visual sensor to provide non-visual sensor data captured by the non-visual sensor data during the time period corresponding to the FOV depicted by the video data;

receiving, at the electronic computing device, from the non-visual sensor, non-visual sensor data that was captured by the non-visual sensor associated with the FOV of the redacted version of the video data during the time period;

detecting, at the electronic computing device, a feature in the non-visual sensor data that indicates that an additional person not redacted in the redacted version of the video data was present in the FOV during the time period; and

providing, at the electronic computing device, an electronic notification indicating that an additional person not redacted in the redacted version of the video data was present in the FOV during the time period.

16 . The method of claim 15 , further comprising:

in response to detecting the feature in the non-visual sensor data that indicates the additional person was not redacted in the redacted version of the video data, automatically redacting the additional person from the FOV of the redacted version of the video data.

17 . The method of claim 15 , further comprising:

applying a computer-vision technique to the video data to determine the degree of obscuration of the FOV.

18 . The method of claim 15 , further comprising:

obtaining additional sensor data captured by a smoke detector, a fog detector, or a light detector during the time period; and

determining the degree of obscuration of the FOV based on additional sensor data.

19 . The method of claim 15 , further comprising:

identifying a location of the additional person based on the non-visual sensor data;

mapping the location of the additional person to a pixel range of the video data; and

inserting an indication of the pixel range into the electronic notification.

20 . The method of claim 19 , further comprising:

rendering, on an electronic display, an image that illustrates the FOV as captured by the camera during the time period; and

rendering, on the electronic display, a visual element that identifies the pixel range in the image.

Assignments (1)

Continuity (1)

References Cited (17)