AI-based public place emergency medical person searching method
Technical Field
The invention relates to an AI-based software detection technology, in particular to an AI-based public place emergency medical people searching method.
Background
Aiming at the 2019 novel coronavirus epidemic situation prevention and control situation, although all parties act emergently, the epidemic situation outbreak is rapid, and great challenges are brought to epidemic prevention work. In order to avoid the unknown virus spreading of more potential patients, how to quickly and accurately find the patients and the pedestrians becomes a very delicate problem. On one hand, each media channel provides various functions and services such as a 'fever clinic' map, novel pneumonia patient peer query service and the like, and people can search by themselves; on the other hand, governments and epidemic prevention related units also strive to search for the same patient and the same pedestrian from a plurality of channels, wherein the channels comprise a skynet, a public transport video, a public place video and the like, but the resolution ratio of part of videos is low, a large number of citizens wear masks to go out, the face identification is difficult to identify, and great difficulty is caused to quick positioning personnel.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the existing problems, the AI-based public place emergency medical person searching method is provided, and rapid tracing of target persons is realized by using AI-based face extraction, identification and comparison technologies.
The invention provides an AI-based public place emergency medical person searching method, which comprises the following steps:
step 1, acquiring a face image of a diagnostician A offline, and storing the face image in a face base library of the diagnostician;
step 2, obtaining the video information of the resident time period of the diagnostician A;
step 3, comparing and retrieving the video information obtained in the step 2 by utilizing a face bottom library of the diagnosticians based on the face extraction, identification and comparison technology of AI, and automatically capturing the picture containing the diagnostician A;
step 4, extracting a face image of the risk person B from the picture containing the diagnostician A according to the judgment condition of the risk person;
and 5, comparing and analyzing the face image of the risk person B through a resident population database, and confirming the identity information of the risk person B.
Further, the face image of the diagnostician a acquired under the line in step 1 includes a front, side and whole body photograph.
Further, step 3 comprises the following substeps:
step 3.1, extracting video frames from the video information obtained in the step 2;
step 3.2, judging whether a face exists in the video frame through a face extraction technology based on AI;
3.3, extracting the face image of the diagnostician A from the video frame with the face by using an AI-based face recognition technology;
and 3.4, comparing the face image of the diagnostician A extracted in the step 3.3 with the face base library of the diagnostician by using an AI-based face comparison technology, and automatically capturing a picture containing the diagnostician A after the comparison is successful.
In another embodiment, before step 3.4 is executed, the image quality of the face image of the diagnostician a extracted in step 3.3 is determined, and the sharpness of the face image of the diagnostician a with insufficient sharpness is improved through a super-resolution restoration technique.
Further, step 4 comprises the following substeps:
step 4.1, finding the time point and scene of the confirmed diagnostician A in the picture containing the confirmed diagnostician A;
and 4.2, extracting the face image of the risk person B according to the judgment condition of the risk person in the scene where the confirmed person A is located based on the time point of the confirmed person A.
Preferably, the determination conditions of the risk person include a residence time, an action range, and a trajectory of the diagnostician a.
In another embodiment, before step 5 is executed, the image quality of the face image of the risk person B extracted in step 4 is determined, and the definition of the face image of the risk person B with insufficient definition is improved through a super-resolution reduction technology.
In another embodiment, the method further comprises: and carrying out information management on the confirmed diagnostician A and a risk person B associated with the confirmed diagnostician A.
Further, the method further comprises: and (4) searching, inquiring and statistically analyzing the information of the confirmed diagnostician A and the risk personnel B according to the needs.
The invention also provides an AI-based public place emergency medical people-searching system, which comprises:
the first acquisition module is used for acquiring the face image of the diagnostician A offline and storing the face image in a diagnostician face base;
the second acquisition module is used for acquiring the video information of the resident time period of the confirmed person A;
the first face processing module is used for comparing and retrieving the video information acquired by the second acquisition module by utilizing the face bottom library of the confirmed person based on the face extraction, identification and comparison technology of AI, and automatically capturing the picture containing the confirmed person A;
the second face processing module is used for extracting a face image of the risk person B from the picture containing the confirmed person A according to the judgment condition of the risk person;
the identity confirmation module is used for comparing and analyzing the face image of the risk person B through the resident population database to confirm the identity information of the risk person B;
and the storage module is used for storing various data.
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:
1. the invention realizes the rapid tracing of the target personnel through the face extraction, identification and comparison technology based on AI.
2. The invention enhances the definition of the low-resolution image by the super-resolution reduction technology before the face comparison, and can better realize the face recognition under the low resolution so as to improve the accuracy.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a block flow diagram of an AI-based public place emergency medical locating method according to an embodiment of the present invention.
FIG. 2 is a block diagram of a flow chart of an AI-based method for locating persons in emergency medical treatment in public places according to another embodiment of the invention
Fig. 3 is a block diagram illustrating an AI-based public emergency medical locating system according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the detailed description and specific examples, while indicating the preferred embodiment of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
The features and properties of the present invention are described in further detail below with reference to examples.
Example 1
The AI-based public place emergency medical person searching method comprises the following five steps:
step 1, acquiring a face image of a diagnostician A offline, and storing the face image in a face base library of the diagnostician;
the confirmed person A is a known object, and a database of face images of the confirmed person is firstly established for subsequent face recognition and comparison. Preferably, the face image of the diagnostician a acquired offline includes a frontal, lateral and whole-body photograph, so as to obtain a more complete and accurate face image of the diagnostician.
Step 2, obtaining the video information of the resident time period of the diagnostician A;
according to the self-describing movement track of the confirmed person A, the related functional departments coordinate the units to which the monitoring resources of the image data of the confirmed person A belong, and the video of the resident time period of the confirmed person A is provided in a matching manner.
As can be seen from the above steps, the present embodiment is established under the complete basic conditions of various video monitoring devices in the current city, such as the callable eye system, traffic monitoring, and anti-theft monitoring installed in buildings and pavements, and can be used as the source of the video providing the resident time period of the diagnostician a.
Step 3, comparing and retrieving the video information obtained in the step 2 by utilizing a face bottom library of the diagnosticians based on the face extraction, identification and comparison technology of AI, and automatically capturing the picture containing the diagnostician A;
specifically, the method comprises the following steps:
step 3.1, extracting video frames from the video information obtained in the step 2;
step 3.2, judging whether a face exists in the video frames through a face extraction technology based on AI, removing the video frames without the face, and only carrying out subsequent processing on the video frames with the face, so that the data processing amount can be reduced, and the operation speed and efficiency can be improved;
3.3, extracting the face image of the diagnostician A from the video frame with the face by using an AI-based face recognition technology;
and 3.4, comparing the face image of the diagnostician A extracted in the step 3.3 with the face base library of the diagnostician by using an AI-based face comparison technology, and automatically capturing a picture containing the diagnostician A after the comparison is successful.
In another embodiment, before step 3.4 is executed, the image quality of the face image of the diagnostician a extracted in step 3.3 is determined, and the definition of the face image of the diagnostician a with insufficient definition is improved by the super-resolution restoration technique, so as to improve the accuracy of the face comparison result.
Step 4, extracting a face image of the risk person B from the picture containing the diagnostician A according to the judgment condition of the risk person;
specifically, the method comprises the following steps:
step 4.1, finding the time point and scene of the confirmed diagnostician A in the picture containing the confirmed diagnostician A;
and 4.2, extracting the face image of the risk person B according to the judgment condition of the risk person in the scene where the confirmed person A is located based on the time point of the confirmed person A.
The determination conditions of the risk personnel include elements such as residence time, action range and track of the confirmed person a, for example, in the 2019 novel coronavirus epidemic situation prevention and control, a medical researcher and the like can give a possibly infected scene including close talking or contact, in the same closed space and the like, and according to the elements such as residence time, action range and track of the confirmed person a, the risk personnel B possibly infected by the confirmed person a can be identified in the corresponding time and scene.
In another embodiment, before step 5 is executed, the image quality of the face image of the risk person B extracted in step 4 is determined, and the definition of the face image of the risk person B with insufficient definition is improved through a super-resolution reduction technology.
And 5, comparing and analyzing the face image of the risk person B through a resident population database, and confirming the identity information of the risk person B. The resident population library is coordinately provided by related functional departments, and the risk personnel B with potential questions can be quickly traced through the face recognition technology. Therefore, the method of the embodiment can meet the requirements of quick retrieval, comparison and association on target personnel when a major emergency medical treatment event occurs, realize tracing and tracing of important persons when the major emergency medical treatment event occurs, be favorable for supporting local epidemic situation prevention and control work and assist epidemic prevention departments in making decisions.
In another embodiment, the method further comprises: and carrying out information management on the confirmed diagnostician A and a risk person B associated with the confirmed diagnostician A. The information of the confirmed person A and the related risk person B can be stored by establishing a database so as to facilitate the management of epidemic prevention work. Still further, the method further comprises: and (4) searching, inquiring and statistically analyzing the information of the confirmed diagnostician A and the risk personnel B according to the needs.
In addition, an AI-based public place emergency medical person finding system is also provided, which is used for realizing the AI-based public place emergency medical person finding method, and comprises the following steps:
the first acquisition module is used for acquiring the face image of the diagnostician A offline and storing the face image in a diagnostician face base;
the second acquisition module is used for acquiring the video information of the resident time period of the confirmed person A;
the first face processing module is used for comparing and retrieving the video information acquired by the second acquisition module by utilizing the face bottom library of the confirmed person based on the face extraction, identification and comparison technology of AI, and automatically capturing the picture containing the confirmed person A;
the second face processing module is used for extracting a face image of the risk person B from the picture containing the confirmed person A according to the judgment condition of the risk person;
the identity confirmation module is used for comparing and analyzing the face image of the risk person B through the resident population database to confirm the identity information of the risk person B;
and the storage module is used for storing various data. Such as the confirmed person face bottom library, the input video information, the face image, and the information of the confirmed person A and the risk person B.
The process of using the public place emergency medical people searching system comprises the following steps:
(1) leading the face image of the diagnostician A into the system by a first acquisition module, and storing the face image in a face bottom library of the diagnostician;
(2) importing corresponding video information to the system by a second acquisition module;
(3) the first face processing module is used for extracting, identifying and comparing faces, and a picture containing a diagnostician A is intercepted;
(4) extracting a face image of a risk person B from a picture containing a diagnostician A by a second face processing module;
(5) and the identity confirmation module confirms the identity information of the risk personnel B according to the face image of the risk personnel B.
(6) And the extracted face image of the confirmed person A and the extracted face image of the risk person B can be manually rechecked according to the requirement.
(7) And storing the information of the confirmed diagnostician A and the information of the risk personnel B related to the confirmed diagnostician A, and performing retrieval query and statistical analysis when needed.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.