Hazard detection method, hazard detection device, hazard detection system, and program
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
- Filing Date
- 2024-12-12
- Publication Date
- 2026-06-24
Smart Images

Figure 2026103352000001_ABST
Abstract
Claims
1. In an information processing device comprising at least one processor, the at least one processor provides: Acquire an image of the area to be detected as a hazard, A prompt is generated that includes at least an instruction to detect hazardous content related to the target area from the image, The aforementioned image and the aforementioned prompt are input to the generating AI model. Based on the image and the output from the generating AI model in response to the prompt input, it is determined whether to notify the user of the detected danger. If it is determined that the user should be notified of the detected danger, a notification command is output instructing the user to be notified of the detected danger. Hazard detection methods.
2. The prompt further includes an instruction to calculate the probability of occurrence of the detected hazard, If the aforementioned probability of occurrence satisfies a predetermined threshold condition, the notification command is output. The hazard detection method according to claim 1.
3. The aforementioned prompt further includes content related to on-site safety and health, The hazard detection method according to claim 1.
4. The prompt further includes environmental information relating to the environment contained in the image, The hazard detection method according to claim 1.
5. The aforementioned notification order includes the hazardous content detected from the image, The hazard detection method according to claim 1.
6. The notification command is output to the notification device corresponding to the target area. The hazard detection method according to claim 1.
7. The user input regarding the correctness of the output content from the aforementioned generated AI model is obtained. If the user input indicates that the output content is incorrect, change at least one of the threshold condition and the prompt. The hazard detection method according to claim 2.
8. Equipped with at least one processor, The at least one processor is Acquire an image of the area to be detected as a hazard, A prompt is generated that includes at least an instruction to detect hazardous content related to the target area from the image, The aforementioned image and the aforementioned prompt are input to the generating AI model. Based on the image and the output from the generating AI model in response to the prompt input, it is determined whether to notify the user of the danger detection. If it is determined that the user should be notified of the detected danger, a notification command is output instructing the user to be notified of the detected danger. Hazard detection device.
9. The aforementioned generation AI model further comprises at least one memory on which the AI model is constructed. The hazard detection device according to claim 8.
10. The hazard detection device according to claim 8, The cloud server on which the aforementioned AI model is constructed includes, Hazard detection system.
11. Acquire an image of the area to be detected as a hazard, A prompt is generated that includes at least an instruction to detect hazardous content related to the target area from the image, The aforementioned image and the aforementioned prompt are input to the generating AI model. Based on the image and the output from the generating AI model in response to the prompt input, it is determined whether to notify the user of the danger detection. To instruct the computer to output a notification command instructing it to notify the user of a detected danger, if it determines that a danger has been detected. program.
Citation Information
Patent Citations
Video monitoring system and video monitoring method
JP2021132267A