# US 2023/0410623 A1 · Safety violation detection

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0410623 A1
- Document ID: US20230410623A1
- Status: Published application
- Kind: A1
- Application number: 17/807,063
- Filed: 2022-06-15
- Published: 2023-12-21
- Earliest claimed: 2022-06-15
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: John S. Werner; Aishwarya Gopinath; Arkadiy O. Tsfasman
- Web page: https://patentgrep.com/patent/US20230410623A1

## CPC

- G08B 21/02: Alarms for ensuring the safety of persons
- F16P 3/142: {using image capturing devices}
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/7715: {Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods}
- G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
- G06V 10/82: using neural networks
- G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69)
- G08B 21/22: responsive to presence or absence of persons
- H04W 4/029: Location-based management or tracking services
- G08B: SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- F16P: SAFETY DEVICES IN GENERAL; {SAFETY DEVICES FOR PRESSES}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

A method, computer system, and a computer program product for safety violation detection is provided. The present invention may include receiving a camera feed of a facility. The present invention may also include analyzing the received camera feed using an AI model to detect a safety violation in the facility. The present invention may also include tracking a position of a user device in the facility associated with a user moving through the facility. The present invention may further include generating a proximity alert responsive to determining that the tracked position of the user device is within a predetermined proximity of the safety violation in the facility. The present invention may also include outputting the generated proximity alert to the user device for preventative action by the user.

## Claims

### Claim 1

A computer-implemented method, comprising:
receiving a camera feed of a facility;
analyzing the received camera feed using an artificial intelligence (AI) model to detect a safety violation in the facility;
tracking a position of a user device in the facility, wherein the user device is associated with a user moving through the facility;
generating a proximity alert responsive to determining that the tracked position of the user device is within a predetermined proximity of the safety violation in the facility; and
outputting the generated proximity alert to the user device for preventative action by the user.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
training the AI model to recognize a plurality of safety violations based on supervised learning of a set of training images tagged with a corresponding set of risk assessment scores.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
training the AI model to recognize the plurality of safety violations based on a set of safety specifications.

### Claim 4 (depends on 3)

The method of claim 3, wherein the set of safety specifications is selected from the group consisting of:
Occupational Safety and Health Administration (OSHA) specifications, Centers for Disease Control and Prevention (CDC) specifications, American National Standards Institute (ANSI) specifications, National Fire Prevention Association (NFPA) specifications, and internal specifications.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
receiving the camera feed of the facility from a facility manager device; and
providing the facility manager device with a report of the safety violation detected in the facility, wherein the report indicates a timeframe for correcting the safety violation detected in the facility.

### Claim 6 (depends on 1)

The method of claim 1, wherein a risk assessment score is assigned to the safety violation detected in the facility; and
wherein the predetermined proximity is based on the risk assessment score associated with the safety violation detected in the facility.

### Claim 7 (depends on 6)

The method of claim 6, wherein the risk assessment score is assigned to the safety violation based on a risk assessment matrix associated with an accident occurrence likelihood and a severity of consequence.

### Claim 8 (depends on 1)

The method of claim 1, further comprising:
generating an initial alert for the user device responsive to determining that the user device is entering the facility including the safety violation.

### Claim 9 (depends on 1)

The method of claim 1, wherein the generated proximity alert is responsive to determining that the safety violation detected in the facility includes a risk assessment score that is above a risk assessment threshold.

### Claim 10 (depends on 1)

The method of claim 1, wherein the safety violation detected in the facility is highlighted on an augmented reality display associated with the user, responsive to determining that the safety violation is in a field of vision of the user.

### Claim 11

A computer system for safety violation detection, comprising:
one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage media, and program instructions stored on at least one of the one or more computer-readable tangible storage media for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising:
receiving a camera feed of a facility;
analyzing the received camera feed using an artificial intelligence (AI) model to detect a safety violation in the facility;
tracking a position of a user device in the facility, wherein the user device is associated with a user moving through the facility;
generating a proximity alert responsive to determining that the tracked position of the user device is within a predetermined proximity of the safety violation in the facility; and
outputting the generated proximity alert to the user device for preventative action by the user.

### Claim 12 (depends on 11)

The computer system of claim 11, further comprising:
training the AI model to recognize a plurality of safety violations based on supervised learning of a set of training images tagged with a corresponding set of risk assessment scores.

### Claim 13 (depends on 12)

The computer system of claim 12, further comprising:
training the AI model to recognize the plurality of safety violations based on a set of safety specifications.

### Claim 14 (depends on 13)

The computer system of claim 13, wherein the set of safety specifications is selected from the group consisting of:
Occupational Safety and Health Administration (OSHA) specifications, Centers for Disease Control and Prevention (CDC) specifications, American National Standards Institute (ANSI) specifications, National Fire Prevention Association (NFPA) specifications, and internal specifications.

### Claim 15 (depends on 11)

The computer system of claim 11, further comprising:
receiving the camera feed of the facility from a facility manager device; and
providing the facility manager device with a report of the safety violation detected in the facility, wherein the report indicates a timeframe for correcting the safety violation detected in the facility.

### Claim 16 (depends on 11)

The computer system of claim 11, wherein a risk assessment score is assigned to the safety violation detected in the facility; and
wherein the predetermined proximity is based on the risk assessment score associated with the safety violation detected in the facility.

### Claim 17 (depends on 16)

The computer system of claim 16, wherein the risk assessment score is assigned to the safety violation based on a risk assessment matrix associated with an accident occurrence likelihood and a severity of consequence.

### Claim 18 (depends on 11)

The computer system of claim 11, further comprising:
generating an initial alert for the user device responsive to determining that the user device is entering the facility including the safety violation.

### Claim 19 (depends on 11)

The computer system of claim 11, wherein the generated proximity alert is responsive to determining that the safety violation detected in the facility includes a risk assessment score that is above a risk assessment threshold.

### Claim 20

A computer program product for safety violation detection, comprising:
one or more computer-readable storage media and program instructions collectively stored on the one or more computer-readable storage media, the program instructions executable by a processor to cause the processor to perform a method comprising:
receiving a camera feed of a facility;
analyzing the received camera feed using an artificial intelligence (AI) model to detect a safety violation in the facility;
tracking a position of a user device in the facility, wherein the user device is associated with a user moving through the facility;
generating a proximity alert responsive to determining that the tracked position of the user device is within a predetermined proximity of the safety violation in the facility; and
outputting the generated alert to the user device for preventative action by the user.
