CN-115620229-B - Method, device, system and medium for detecting helmet

Abstract

The embodiment of the invention provides a method, a device, a storage medium and a system for detecting a helmet, wherein the method comprises the steps of obtaining a sample image set of a target object, carrying out attribute identification on the sample image set to obtain first label information and second label information, carrying out model training operation on an initial neural network model based on the first label, first label confidence coefficient, second label and second label confidence coefficient to obtain the target neural network model, carrying out label confidence coefficient analysis on the target image information through the target neural network model to obtain label confidence coefficient information of the target image, and sending an attitude adjustment instruction to the target object when the label confidence coefficient information comprises the first label and the first confidence coefficient is larger than a first threshold value, and carrying out alarm operation when the label confidence coefficient information comprises the second label and the second confidence coefficient is larger than the first threshold value. The invention solves the problem of low recognition precision of the helmet, thereby achieving the effect of improving the recognition precision of the helmet.

Inventors

• YANG FENG
• ZHOU YANGBO
• WANG LIUWEI

Assignees

• 湖南喜宝达信息科技有限公司

Dates

Publication Date

20260508

Application Date

20221013

Claims (10)

1. 1. A method of detecting a person's helmet, for use with a sharing device comprising a helmet device, comprising: Acquiring a sample image set of a target object; Performing attribute identification on the sample image set to obtain first tag information and second tag information, wherein the first tag information comprises a first tag of which the target object comprises helmet equipment information and first confidence coefficient of the first tag, and the second tag information comprises a second tag of which the target object does not comprise helmet equipment information and second confidence coefficient of the second tag; Inputting the first label, the first label confidence coefficient, the second label and the second label confidence coefficient into an initial neural network model to perform model training operation on the initial neural network model and obtain a target neural network model; performing label confidence analysis on target image information through the target neural network model to obtain label confidence information of the target image, wherein the label confidence information comprises at least any one of the first label, the first label confidence, the second label and the second label confidence; When the label confidence information comprises the first label and the first confidence is larger than a first threshold value, confirming that the helmet of the target object is in a first state, and sending an attitude adjustment instruction to the target object; and if the label confidence information comprises the second label and the second confidence is larger than a first threshold value, confirming that the helmet of the target object is in a second state, and executing alarm operation.
2. 2. The method of claim 1, wherein performing attribute identification on the sample image set to obtain first tag information and second tag information comprises: Dividing the sample image set through a preset dividing model to obtain initial helmet information; And carrying out tag attribute identification on the initial helmet information through a preset initial neural network model so as to obtain the first tag information and the second tag information.
3. 3. The method of claim 1, wherein, in the event that the tag confidence information includes the first tag and the first confidence is greater than a first threshold, the method further comprises: Acquiring initial position information and target position information of the helmet equipment, wherein the initial position information comprises initial coordinate information and initial height information of the helmet equipment in target time, and the target position information comprises final coordinate information and final height information of the helmet equipment in target time; determining coordinate track information of the helmet equipment based on the initial coordinate information and the final coordinate information; determining altitude trajectory information of the helmet equipment based on the initial altitude information and the final altitude information; and feeding back equipment starting information to an external control center under the condition that the coordinate track information meets the second condition and/or the altitude track information meets the third condition.
4. 4. A method according to claim 3, wherein said determining height trajectory information of the helmet device based on the initial height information and final height information comprises: determining a detection reference plane based on the initial height information; determining a trajectory plane of the helmet device based on the initial height information and the final height information; and determining a movement track included angle according to the track plane and the detection reference plane, and taking the movement track included angle as the height track information.
5. 5. A method according to claim 3, wherein said determining height trajectory information of the helmet device based on the initial height information and final height information comprises: And determining a movement height difference of the helmet equipment in a target time based on the initial height information and the final height information, and taking the movement height difference as the height track information.
6. 6. The method according to claim 1, wherein in case it is determined that the attribute identification result satisfies the first condition, the method further comprises: under the condition that the first timing meets a preset period, a tracking instruction is sent to the helmet equipment so as to instruct the helmet equipment to feed back response information; Upon receiving the response information of the helmet device, a communication link is established with the helmet device.
7. 7. A headgear detection apparatus for use with a shared device comprising a headgear device, comprising: the initial image acquisition module is used for acquiring a sample image set of the target object; The attribute identification module is used for carrying out attribute identification on the sample image set to obtain first tag information and second tag information, wherein the first tag information comprises a first tag of which the target object comprises helmet equipment information and first confidence coefficient of the first tag, and the second tag information comprises a second tag of which the target object does not comprise helmet equipment information and second confidence coefficient of the second tag; The model training module is used for inputting the first label, the first label confidence coefficient, the second label and the second label confidence coefficient into an initial neural network model so as to perform model training operation on the initial neural network model and obtain a target neural network model; The information analysis module is used for carrying out label confidence analysis on the target image information through the target neural network model so as to obtain label confidence information of the target image, wherein the label confidence information comprises at least any one of the first label, the first label confidence, the second label and the second label confidence; The instruction sending module is used for confirming that the helmet of the target object is in a first state and sending an attitude adjustment instruction to the target object when the label confidence information contains the first label and the first confidence is larger than a first threshold; and the alarm module is used for confirming that the helmet of the target object is in a second state and executing alarm operation under the condition that the label confidence information contains the second label and the second confidence is larger than a first threshold value.
8. 8. The apparatus of claim 7, wherein the attribute identification module comprises: The image segmentation unit is used for carrying out segmentation processing on the sample image set through a preset segmentation model so as to obtain initial helmet information; And the attribute identification unit is used for carrying out tag attribute identification on the initial helmet information through a preset initial neural network model so as to obtain the first tag information and the second tag information.
9. 9. A headgear detection system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program to perform the steps of the headgear detection method of any one of claims 1 to 6.
10. 10. A storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method for detecting helmets according to any one of claims 1 to 6.

Description

Method, device, system and medium for detecting helmet Technical Field The present invention relates to the field of image detection technologies, and in particular, to a method, an apparatus, a system, and a medium for detecting a helmet. Background In order to standardize the riding of the shared electric vehicle, the special regulation of the traffic violation of the shared electric vehicle is started at present, and the traffic violation of drivers and passengers without wearing safety helmets is treated in important road sections of urban areas. Therefore, before riding the sharing electric vehicle, a step of correctly wearing the helmet is added to unlock the sharing electric vehicle. In the prior art, the means for detecting whether the helmet is worn correctly is rough, and in a normal case, a user can judge that the helmet is worn correctly as long as the helmet exists in the detection range of the camera, and the detection mode is difficult to avoid the nonstandard situations such as that the helmet is not worn or the helmet is not worn on the top of the head of the user, and the like, and in the actual vehicle using process, the nonstandard situations can be detected as correctly, so that the actual helmet is not in normal. Based on this, there is a need to propose a method, device, system and medium for detecting helmets, which solve or at least alleviate the above-mentioned drawbacks. Disclosure of Invention The application mainly aims to provide a method, a device, a system and a medium for detecting a helmet, which are used for solving the problem that the actual helmet is not standard because the prior art is difficult to avoid the non-standard condition that the helmet is in a state of not being worn or the helmet is not worn on the top of a user and the like passes through a helmet detection link. According to one embodiment of the invention, a method, a device, a system and a medium for detecting a helmet are provided, which are applied to a sharing device comprising a helmet device, and include: Acquiring a sample image set of a target object; Performing attribute identification on the sample image set to obtain first tag information and second tag information, wherein the first tag information comprises a first tag of which the target object comprises helmet equipment information and first confidence coefficient of the first tag, and the second tag information comprises a second tag of which the target object does not comprise helmet equipment information and second confidence coefficient of the second tag; Inputting the first label, the first label confidence coefficient, the second label and the second label confidence coefficient into an initial neural network model to perform model training operation on the initial neural network model and obtain a target neural network model; performing label confidence analysis on target image information through the target neural network model to obtain label confidence information of the target image, wherein the label confidence information comprises at least any one of the first label, the first label confidence, the second label and the second label confidence; When the label confidence information comprises the first label and the first confidence is larger than a first threshold value, confirming that the helmet of the target object is in a first state, and sending an attitude adjustment instruction to the target object; and if the label confidence information comprises the second label and the second confidence is larger than a first threshold value, confirming that the helmet of the target object is in a second state, and executing alarm operation. In an exemplary embodiment, the performing attribute identification on the sample image set to obtain the first tag information and the second tag information includes: Dividing the sample image set through a preset dividing model to obtain initial helmet information; And carrying out tag attribute identification on the initial helmet information through a preset initial neural network model so as to obtain the first tag information and the second tag information. In one exemplary embodiment, in the case that the tag confidence information includes the first tag and the first confidence is greater than a first threshold, the method further comprises: Acquiring initial position information and target position information of the helmet equipment, wherein the initial position information comprises initial coordinate information and initial height information of the helmet equipment in target time, and the target position information comprises final coordinate information and final height information of the helmet equipment in target time; determining coordinate track information of the helmet equipment based on the initial coordinate information and the final coordinate information; determining altitude trajectory information of the helmet equipment based on the initial altitude information and th