CN-121999575-A - Risk identification method, reminding method, wearable terminal, collaborative reminding terminal and system

Abstract

The application discloses a risk identification method, a reminding method, a wearable end, a cooperative reminding end and a system, wherein the method comprises the steps of collecting video stream data containing limb behaviors of a wearer; the method comprises the steps of inputting video stream data into a pre-trained behavior recognition model to recognize risk types and corresponding risk grades of limb behaviors of a wearer, tracking the completion state of sequential risk actions under the condition that the risk types indicate the sequential risk actions, generating a risk recognition result comprising the risk types and the risk grades, and sending the risk recognition result to a collaborative reminding end.

Inventors

• GUO QINGDA
• Deng fuhao
• YANG JUNLONG

Assignees

• 深圳市拓普智造科技有限公司

Dates

Publication Date

20260508

Application Date

20260129

Claims (10)

1. 1. A method for identifying behavioral risks of a wearable terminal, comprising: Collecting video stream data containing the limb behaviors of a wearer; Inputting the video stream data into a pre-trained behavior recognition model to recognize risk types and corresponding risk levels of the limb behaviors of the wearer, wherein the completion state of the sequential risk actions is tracked under the condition that the risk types indicate the sequential risk actions; Generating a risk identification result comprising the risk type and the risk level; And sending the risk identification result to a collaborative reminding end.
2. 2. The behavioral risk identification method of claim 1, wherein the behavioral identification model is a lightweight model.
3. 3. The behavior risk reminding method of the collaborative reminding terminal is characterized by comprising the following steps of: The method comprises the steps of receiving a risk identification result from a wearable end, acquiring video stream data containing the limb behaviors of a wearer by the wearable end, inputting the video stream data into a pre-trained behavior identification model to identify the risk type and the corresponding risk level of the limb behaviors of the wearer, and generating a risk identification result comprising the risk type and the risk level; and executing corresponding risk prompting operation according to the risk identification result.
4. 4. The behavioral risk reminder method according to claim 3, wherein the performing a corresponding risk reminder operation according to the risk identification result includes: Executing a risk prompting operation corresponding to the risk level under the condition that the risk type indicates a single risk action; In case the risk type indicates a sequential risk action, a corresponding risk prompting operation is performed based on a completion status and a risk level of the sequential risk action.
5. 5. The behavioral risk prompting method according to claim 4, wherein the performing a corresponding risk prompting operation based on the completion status and risk level of the sequential risk actions includes: if the completion status indicates that the sequential risk action is not complete, establishing a suspension event corresponding to the sequential risk action, the suspension event for tracking the completion status of the sequential risk action; And executing corresponding risk prompt operation according to the risk level based on the suspension event.
6. 6. A wearable end for behavioral risk identification, comprising: The video acquisition unit is used for acquiring video stream data containing the limb behaviors of a wearer; the behavior recognition unit is used for inputting the video stream data into a pre-trained behavior recognition model so as to recognize the risk type and the corresponding risk level of the limb behavior of the wearer, wherein the completion state of the sequential risk actions is tracked under the condition that the risk type indicates the sequential risk actions; a result generation unit for generating a risk identification result including the risk type and the risk level; The first communication unit is used for sending the risk identification result to the collaborative reminding end.
7. 7. The wearable end of claim 6, further comprising: The device comprises a shell, wherein the video acquisition unit, the behavior recognition unit, the result generation unit and the first communication unit are all arranged in the shell, and an installation piece is arranged on the shell and is used for being fixed on the head or the trunk of a wearer.
8. 8. The wearable end of claim 6, further comprising: The local reminding unit is used for executing risk reminding operation according to the risk identification result under the condition that the first communication unit and the collaborative reminding end are failed in communication.
9. 9. A collaborative reminder for behavioral risk reminders, comprising: The system comprises a wearable end, a second communication unit, a sequential risk action, a first communication unit, a second communication unit and a third communication unit, wherein the second communication unit is used for receiving a risk identification result from the wearable end, the risk identification result is obtained by the wearable end and comprises video stream data of the limb action of the wearer, the video stream data are input into a pre-trained action identification model to identify the risk type and the corresponding risk level of the limb action of the wearer, and a risk identification result comprising the risk type and the risk level is generated; And the prompt execution unit is used for executing corresponding risk prompt operation according to the risk identification result.
10. 10. A behavioral risk identification alert system comprising: a wearable end as claimed in any one of claims 6 to 8; The co-alert terminal of claim 9, in communication with the wearable terminal.

Description

Risk identification method, reminding method, wearable terminal, collaborative reminding terminal and system Technical Field The application relates to the technical field of monitoring equipment, in particular to a risk identification method, a reminding method, a wearable end, a cooperative reminding end and a system. Background The problem of home safety for the elderly is becoming increasingly pronounced, especially in sequential daily activities involving multiple steps, such as fire, medication, and frequent risks due to memory or operator inattention. In the related monitoring technology, dead zones exist in fixed monitoring and privacy is poor, common wearable equipment is focused on physiological index monitoring, specific behavior sequences cannot be identified, and a behavior identification scheme based on cloud analysis is difficult to realize real-time and continuous end-side behavior tracking and state judgment due to limitation of network delay and privacy risks. The monitoring technologies are difficult to realize reliable and noninductive monitoring of household safety risks, and the requirements of users are difficult to meet. Disclosure of Invention The application aims to provide a risk identification method, a reminding method, a wearable end, a collaborative reminding end and a system, which can realize reliable and noninductive monitoring of household safety risks. In a first aspect, an embodiment of the present application provides a method for identifying a risk of behavior of a wearable terminal, including: Collecting video stream data containing the limb behaviors of a wearer; Inputting the video stream data into a pre-trained behavior recognition model to recognize risk types and corresponding risk levels of the limb behaviors of the wearer, wherein the completion state of the sequential risk actions is tracked under the condition that the risk types indicate the sequential risk actions; Generating a risk identification result comprising the risk type and the risk level; And sending the risk identification result to a collaborative reminding end. According to some embodiments of the application, the behavior recognition model is a lightweight model. In a second aspect, an embodiment of the present application provides a behavioral risk reminding method for a collaborative reminder, including: The method comprises the steps of receiving a risk identification result from a wearable end, acquiring video stream data containing the limb behaviors of a wearer by the wearable end, inputting the video stream data into a pre-trained behavior identification model to identify the risk type and the corresponding risk level of the limb behaviors of the wearer, and generating a risk identification result comprising the risk type and the risk level; and executing corresponding risk prompting operation according to the risk identification result. According to some embodiments of the application, the executing a corresponding risk prompting operation according to the risk identification result includes: Executing a risk prompting operation corresponding to the risk level under the condition that the risk type indicates a single risk action; In case the risk type indicates a sequential risk action, a corresponding risk prompting operation is performed based on a completion status and a risk level of the sequential risk action. According to some embodiments of the application, the performing a corresponding risk prompting operation based on the completion status and the risk level of the sequential risk actions includes: if the completion status indicates that the sequential risk action is not complete, establishing a suspension event corresponding to the sequential risk action, the suspension event for tracking the completion status of the sequential risk action; And executing corresponding risk prompt operation according to the risk level based on the suspension event. In a third aspect, an embodiment of the present application provides a wearable end for behavior risk identification, including: The video acquisition unit is used for acquiring video stream data containing the limb behaviors of a wearer; the behavior recognition unit is used for inputting the video stream data into a pre-trained behavior recognition model so as to recognize the risk type and the corresponding risk level of the limb behavior of the wearer, wherein the completion state of the sequential risk actions is tracked under the condition that the risk type indicates the sequential risk actions; a result generation unit for generating a risk identification result including the risk type and the risk level; The first communication unit is used for sending the risk identification result to the collaborative reminding end. According to some embodiments of the application, further comprising: The device comprises a shell, wherein the video acquisition unit, the behavior recognition unit, the result generation unit and the first