CN110222561A - Missing crew's facial images match method, apparatus and storage medium - Google Patents

Abstract

The present invention relates to a face image matching method, device and storage medium of a missing person, which is used to solve the problem of low efficiency in finding missing persons in related technologies. The image is subjected to face detection; the face in the target image of the detected face is matched with the facial image of the missing person pre-stored in the smart wearable device, and the difference between the human face and the facial image of the missing person is determined. the similarity between them; when the similarity between the human face and the facial image of the missing person is greater than a threshold, a prompt message is sent on the smart wearable device. The invention can improve the efficiency of finding missing persons.

Description

Face image matching method, device and storage medium of missing persons

technical field

The invention relates to the technical field, in particular to a missing person's face image matching method, device and storage medium.

Background technique

At present, a large number of missing persons appear every year all over the world. The search for missing persons usually requires a lot of manpower and material resources. However, due to the large search area and large population, the probability of successfully finding missing persons is still not high. With the continuous advancement of artificial intelligence technology, information on missing persons gradually appears on the Internet in order to spread the information on missing persons, hoping to use the power of the majority of netizens to help find missing persons. However, human attention is limited after all. Identifying missing persons remains difficult and less efficient.

Contents of the invention

The invention provides a missing person's face image matching method, device and storage medium, which are used to solve the problem of low efficiency in finding the missing person in the related art.

According to the first aspect of the present invention, there is provided a facial image matching method for a missing person, comprising: collecting an image when the smart wearable device is in a working state; performing face detection on the image; Match the face in the face with the facial image of the missing person pre-stored in the smart wearable device to determine the similarity between the face and the facial image of the missing person; When the similarity between people's facial images is greater than a threshold, a prompt message is sent on the smart wearable device.

Optionally, the method further includes: after performing face detection on the image, deleting the image in which no face is detected; determining the similarity between the face and the facial image of the missing person After the degree of similarity between the human face and the facial image of the missing person is not greater than the threshold, the image is deleted.

Optionally, the method further includes: after the smart wearable device captures the image, setting a unique identifier and/or a geographic location tag for the image, where the geographic location tag is used to identify the geographic location where the image was taken.

Optionally, the sending a prompt message on the smart wearable device includes: prompting the unique identifier of the target image and/or the geographic location tag on the smart wearable device.

Optionally, the method further includes: when the similarity between the human face and the facial image of the missing person is greater than a threshold, notifying at least one target device bound to the smart wearable device to present the missing person image and/or text information of the missing person to present the image and/or text information of the missing person on the target device.

Optionally, the method further includes: when the similarity between the human face and the facial image of the missing person is greater than a threshold, sending the target image to at least one target bound to the smart wearable device device to render the target image on the target device.

Optionally, the method further includes: requesting the server for updated missing person information at a preset period; deleting the found missing person information according to the updated missing person information and/or saving the latest The missing person's information is newly added within a preset period, and the missing person's information includes the missing person's facial image and/or text information.

According to the second aspect of the present invention, there is provided a facial image matching device for missing persons, including: an acquisition module, used to collect images when the smart wearable device is in working state; Detection; matching module, for matching the human face in the target image detected with the human face with the facial image of the missing person pre-stored in the smart wearable device, and determining the human face and the facial image of the missing person The similarity between them; a prompt module, configured to send a prompt message on the smart wearable device when the similarity between the human face and the facial image of the missing person is greater than a threshold.

According to a third aspect of the present invention, there is provided a facial image matching device for missing persons, including: a memory and a processor, wherein computer-readable instructions are stored in the memory, and the computer-readable instructions are executed by the processor , causing the processor to execute the steps of the missing person facial image matching method according to the first aspect of the present invention.

According to a fourth aspect of the present invention, there is provided a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor, the processor can perform the operations described in the first aspect of the present invention. The steps of the face image matching method for missing persons.

The face image matching method for missing persons in the embodiment of the present invention utilizes the feature that the smart wearable device can be worn by the user so that it can appear in various occasions with the user, so that the smart wearable device collects images when it is in the working state, and from the collected The target image including the face is screened out from the image, the target image is matched with the facial image of the missing person, and the face image matching the missing person is determined based on the similarity between the two, which realizes the use of smart wearable devices to collect and For the purpose of identifying the facial images of missing persons, this solution does not need to manually collect and match images, which improves the efficiency of finding missing persons.

Description of drawings

The accompanying drawings described here are used to provide a further understanding of the present invention, and constitute a part of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention, and do not constitute improper limitations to the present invention. In the attached picture:

Fig. 1 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment;

Fig. 2 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment;

Fig. 3 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment;

Fig. 4 is a flow chart of a missing person facial image matching method shown according to an exemplary embodiment;

Fig. 5 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment;

Fig. 6 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment;

Fig. 7 is a block diagram of an apparatus for matching facial images of missing persons according to an exemplary embodiment.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

Fig. 1 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment. As shown in Fig. 1, the method includes the following steps:

Step 101: collecting images when the smart wearable device is in working condition;

Smart wearable devices may include, for example, smart watches, smart bracelets, smart glasses, and smart clothing. Micro cameras can be installed on smart wearable devices, for example, cameras can be installed on the dials of smart watches or smart bracelets, cameras can be installed on the frames of smart glasses or on the surface of smart clothing, and when the smart wearable devices are detected to be in working state, the control The camera starts to capture images continuously. For example, the camera can be controlled to take pictures every specified time interval, for example, every 1s to take pictures, or it can be taken continuously. After the smart wearable device is working, the camera can be controlled to take pictures. multiple image frames. In view of the small memory of the smart wearable device, and the captured image can meet the requirements of face recognition, the minimum resolution of the image captured by the smart wearable device can be set to 320*480, so as to occupy as little as possible of the smart wearable device. Memory.

In addition, considering that when the smart wearable device is worn by the user and is in a working state, if the user is still, the images captured by the smart wearable device may be roughly the same for a long period of time. In this case, the captured images cannot capture Therefore, after detecting that the geographical location of the smart wearable device has not changed for a preset period of time, the smart wearable device can be controlled to stop collecting images until the geographical location of the smart wearable device changes After that, control the smart wearable device to continuously collect images.

Step 102: Perform face detection on the image;

In step 102, the smart wearable device can perform face detection on the frame image after each shot to obtain a frame image, to detect whether the image includes a human face and/or the number of human faces included in the image , the position information of the face in the image, the size and posture of the face. In the case that multiple human faces are detected in the image, different identifications can be set for the multiple human faces to distinguish the multiple human faces, or the position information of the multiple human faces in the image can be directly used to identify the multiple human faces. individual faces.

Step 103: Match the face in the target image of the detected face with the facial image of the missing person pre-stored in the smart wearable device, and determine the similarity between the human face and the facial image of the missing person Spend;

In step 103, among the multi-frame images captured by the smart wearable device, the image with detected faces can be called the target image, and the facial image information of multiple missing persons can be pre-stored in the smart wearable device. The multiple missing persons are identified in the images collected by the wearable device, and the faces detected in the target image can be sequentially matched with the facial images of the multiple missing persons pre-stored in the smart wearable device to obtain the missing persons in the target image respectively. The similarity between the detected face and multiple missing person facial images is used to preliminarily judge the possibility that the detected human face in the target image is the same person as one of the multiple missing persons according to the similarity. Among them, to determine the similarity between the face detected in the target image and the facial image of the missing person, the existing facial recognition (also called portrait recognition) technology can be used, and the facial recognition model can be pre-stored in the smart wearable device. The model takes the face image in the target image and the face image of the missing person as the input of the model, and the output is the similarity between the two. The server side can verify the model based on a large amount of face image-related data collected, and continuously improve the recognition accuracy, thereby optimizing the model, and the smart wearable device can regularly obtain the latest model information from the server side for its The local model is updated.

In one possible way, in view of the fact that smart wearable devices tend to move more frequently in a certain area when worn by users, the probability of smart wearable devices collecting facial images of missing persons in this area is relatively high, so the smart The pre-stored missing person information in the wearable device may include information on missing persons within a certain area where the smart wearable device is located, for example, the smart wearable device may include information on all missing persons in the province where the smart wearable device is located , or may include information on all missing persons in the city where the smart wearable device is located; the user may also select/set the pre-stored missing person information in the smart wearable device, for example, the user may select a certain province/ The city/district is used as the target area. According to the target area selected by the user, the smart wearable device requests the server for the information of the missing person in the area, receives and stores the information returned by the server; the user can also choose the missing person concerned. Personnel information, for example, the user can enter the name and other identification information of the missing person in the smart wearable device, and the smart wearable device can request the information of the missing person from the server according to the identification information, receive and store the information of the missing person returned by the server .

Step 104: When the similarity between the human face and the facial image of the missing person is greater than a threshold, send a prompt message on the smart wearable device.

As the disappearance time of the missing person goes on, the appearance of the missing person may change to a certain extent, especially for missing children, whose appearance changes greatly over time, so the above threshold cannot be set too high. Based on this, the above threshold can, for example, be is 85%, when the similarity between the face in the target image and the face image of the missing person is greater than this threshold, it can be considered that the face detected in the target image matches the face image of the missing person successfully, and the face in the target image The person is likely to be the missing person, so the user can be prompted through the smart wearable device, so that the user can further check whether the person in the target image and the missing person are the same person, so as to facilitate the retrieval of the missing person. Wherein, the above-mentioned prompt information may be text information or voice information, for example, the prompt information may include the matching information of the missing person, such as identifying information such as the name or number of the missing person, or may also include the identification of the target image, etc. Information, such as the number or label of the target image, or, when more than one face is detected in the target image, the prompt information may also include the position information of the face in the target image that matches the facial image of the missing person. In addition, the smart wearable device can also remind the user that the image of the missing person is currently matched by ringing or vibrating.

In an achievable way, the smart wearable device can collect images in real time, perform face detection on the collected images in real time, and perform face recognition in real time. The function can be enabled by default, and the user can select this function in the setting interface of the smart wearable device , or the smart wearable device can turn off this function when it detects that the user is sleeping or the user is not using the smart wearable device.

The face image matching method for missing persons in the embodiment of the present invention utilizes the feature that the smart wearable device can be worn by the user so that it can appear in various occasions with the user, so that the smart wearable device collects images when it is in the working state, and from the collected The target image including the face is screened out from the image, the target image is matched with the facial image of the missing person, and the face image matching the missing person is determined based on the similarity between the two, which realizes the use of smart wearable devices to collect and For the purpose of identifying the facial images of missing persons, this solution does not need to manually collect and match images, which improves the efficiency of finding missing persons.

Fig. 2 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment. As shown in Fig. 2 , the method may further include on the basis of the method shown in Fig. 1 : After the face detection is performed on the image, delete the image that does not detect a face. Since the smart wearable device continues to collect images during its work, a large number of images may be stored in the smart wearable device. In view of the memory of the smart wearable device Usually small, so you can delete images where faces are not detected in the captured images. Step 202: After determining the similarity between the human face and the missing person's facial image, delete the image whose similarity between the human face and the missing person's facial image is not greater than the threshold, When the similarity between the face detected in the image and the face image of the missing person is not greater than the threshold, it can be considered that the face detected in the image is not the face image of the missing person. The image is stored in the device, and in order to avoid occupying the memory of the smart wearable device, the image can be deleted from the smart wearable device. In addition, it is also possible to set the smart wearable device to automatically search for and shoot images containing human faces when it is within its shooting range, which ensures that all images captured by the smart wearable device contain human faces and reduces the number of images captured by the smart wearable device. the number of images.

Fig. 3 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment. As shown in Fig. 3 , the method may further include, on the basis of the method described in Fig. After the smart wearable device collects the image, set a unique identifier and/or geographic location tag for the image, the geographic location tag is used to identify the geographic location where the image was taken, wherein the unique identifier of the image is used to uniquely identify the image , setting a unique identifier for the image may, for example, number the image. Among them, step 301 can be executed before step 102, and can also be executed after step 102. Figure 3 only shows a case where step 301 is executed before step 102. When step 301 is executed before step 102, the smart wearable The device can set unique identifiers and/or geographic location tags for all images it collects, and when step 301 is performed after step 102, the smart wearable device can only set unique identifiers and/or geographic location tags for target images. To set a geographic location tag for an image, the GPS (Global Positioning System, Global Positioning System) module in the smart wearable device can obtain the geographic location information of the smart wearable device when the image is taken, and use the geographic location information as the geographic location corresponding to the image Tags, similarly, when the face detected in the captured image is successfully matched with the face of the missing person, the geographic location can be considered as the geographic location where the missing person was found, so that the missing person can be identified based on the geographic location information. to retrieve.

Based on the method shown in FIG. 3 , in an implementable manner, sending a prompt message on the smart wearable device may include: prompting the unique identification and/or of the target image on the smart wearable device or the geolocation tag. For example, the text information of "the missing person y is currently found in the image x" can be displayed on the display screen of the smart wearable device, where x is the unique identifier of the image, and y is the identifier of the missing person. Or a geographic location tag can be included in the text information, for example, "find the missing person in the z image", where z can be the geographic location tag of the image, for example, Beijing-Chaoyang District-Chaoyang Park, or it can be in the The text information includes a unique identifier of the image and a geographic location tag, for example, the text information may be "find the missing person in the x-y image".

Fig. 4 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment. As shown in Fig. 4 , the method may further include, on the basis of the method shown in Fig. When the similarity between the face and the facial image of the missing person is greater than a threshold, at least one target device bound to the smart wearable device is notified to present the image and/or text information of the missing person, so as to The image and/or text information of the missing person is presented on the target device. Wherein, the target device can be a mobile terminal device or a PC, and the image data and text data of the missing person pre-stored in the smart wearable device can be pre-stored in the target device. When receiving the facial image of the missing person, the identification of the missing person can be sent to the target device. After receiving the identification, the target device can retrieve the corresponding image data and text data of the missing person according to the identification. According to the user's operation, it can Present these materials to the user so that the user can obtain detailed information about the missing person; or, the image and/or text data of the missing person can be pre-stored in the smart wearable device, and identified in the image collected by the smart wearable device When the facial image of the missing person is received, the image and/or text data of the missing person can be sent to the target device, so that the image and/or text data can be presented on the target device. It should be noted that step 401 may be performed before step 104, or may be performed after step 104, and FIG. 4 only shows a case where step 401 is performed after step 104.

In an achievable way, in order to facilitate the user to check the face detected in the target image with the facial image of the missing person, the target image can be sent to the target device pre-bound with the smart wearable device, so that the The target image is clearly displayed on the device, for example, the target image can be sent to a mobile terminal device or PC that is pre-bound with the smart wearable device. Fig. 5 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment. As shown in Fig. 5 , the method may further include, on the basis of the method shown in Fig. 1 : When the similarity between the human face and the facial image of the missing person is greater than a threshold, the target image is sent to at least one target device bound to the smart wearable device, so as to present the target device on the target device target image. When multiple faces are identified in the target image, the face that matches the missing person can be identified in the target image sent to the target device. Wherein, step 501 may be performed before step 104, and may also be performed after step 104, and FIG. 5 only shows a case where step 501 is performed after step 104.

Fig. 6 is a flow chart of a method for matching facial images of missing persons according to an exemplary embodiment. As shown in Fig. 6, the method may further include on the basis of the method shown in Fig. Set a periodic request to the server for updated missing person information, for example, the smart wearable device may request the server for the missing person information updated on the server side within one month every other month (an example of a preset cycle), wherein, The information of the missing person updated on the server side may include the information of the missing person within the month, and may also include the information of the missing person recovered within the month; step 602: delete the missing person according to the updated information of the missing person The information of the missing person that has been found and/or the information of the missing person newly added in the last preset period is saved, wherein the information of the missing person includes the facial image and/text information of the missing person. It should be noted that, in the method provided in this embodiment, step 601 and step 602 can be performed before or after any step in step 101 to step 104, and FIG. 6 only shows a possible implementation of this method .

Fig. 7 is a block diagram of a facial image matching device for missing persons according to an exemplary embodiment. As shown in Fig. 7, the device 70 includes: an acquisition module 71, which is used to acquire images when the smart wearable device is in working state; The detection module 72 is used to detect the face of the image; the matching module 73 is used to match the face in the target image with the detected face with the facial image of the missing person pre-stored in the smart wearable device , to determine the similarity between the face and the facial image of the missing person; the prompt module 74 is configured to, when the similarity between the human face and the facial image of the missing person is greater than a threshold, in the A reminder message is issued on the smart wearable device.

In a practicable manner, the above-mentioned device may further include: a first deletion module, configured to delete the image in which no face is detected after performing face detection on the image; a second deletion module, configured to After determining the similarity between the human face and the missing person's facial image, deleting the images whose similarity between the human face and the missing person's facial image is not greater than the threshold.

In a practicable manner, the above-mentioned apparatus may further include: a setting module, configured to set a unique identifier and/or a geographic location tag for the image after the smart wearable device captures the image, and the geographic location tag is used for A geographic location where the image was taken is identified.

In an implementable manner, the prompt module may be configured to prompt the unique identifier of the target image and/or the geographic location tag on the smart wearable device.

In a practicable manner, the above-mentioned apparatus may further include: a notification module, configured to notify the user bound to the smart wearable device when the similarity between the human face and the facial image of the missing person is greater than a threshold At least one target device presents the image and/or textual material of the missing person to present the image and/or textual material of the missing person on the target device.

In a practicable manner, the above device may further include: a sending module, configured to send the target image to the At least one target device bound to the smart wearable device, so as to present the image on the target device.

In a practicable manner, the above device may further include: a requesting module, configured to request the server for updated missing person information at a preset period; an updating module, configured to delete the found missing person information according to the updated missing person information. Missing person information and/or save missing person information newly added in the last preset period, and the missing person information includes missing person's face image and/text information.

Based on the method shown in Figure 1 above, correspondingly, the embodiment of the present invention also provides a storage medium on which a computer program is stored, and when the program is executed by a processor, the face of the missing person shown in Figure 1 above Image matching method.

Based on this understanding, the technical solution of the present invention can be embodied in the form of software products, which can be stored in a non-volatile storage medium (which can be CD-ROM, U disk, mobile hard disk, etc.), including several The instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute the facial image matching method for missing persons described in each implementable manner of the present invention.

Based on the method shown in Figure 1 above and the embodiment of the virtual device shown in Figure 7, in order to achieve the above purpose, the embodiment of the present invention also provides a computer device, which can be a personal computer, a server, a network device, etc. , the entity device includes a storage medium and a processor; the storage medium is used to store a computer program; and the processor is used to execute the computer program to realize the above method for matching facial images of missing persons as shown in FIG. 1 .

Optionally, the computer device may further include a user interface, a network interface, a camera, a radio frequency (Radio Frequency, RF) circuit, a sensor, an audio circuit, a WI-FI module, and the like. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the like, and optional user interfaces may also include a USB interface, a card reader interface, and the like. Optionally, the network interface may include a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface) and the like.

Those skilled in the art can understand that the structure of a computer device provided in this embodiment does not constitute a limitation to the physical device, and may include more or less components, or combine some components, or arrange different components.

The storage medium may also include an operating system and a network communication module. An operating system is a program that manages the hardware and software resources of a computer device and supports the operation of information processing programs and other software and/or programs. The network communication module is used to realize the communication between various components inside the storage medium, and communicate with other hardware and software in the physical device.

Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be realized by means of software plus a necessary general-purpose hardware platform, or by hardware. By applying the technical scheme of the present invention.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily necessary for implementing the present invention. Those skilled in the art can understand that the modules in the devices in the implementation scenario can be distributed among the devices in the implementation scenario according to the description of the implementation scenario, or can be located in one or more devices different from the implementation scenario according to corresponding changes. The modules of the above implementation scenarios can be combined into one module, or can be further split into multiple sub-modules.

The above serial numbers of the present invention are for description only, and do not represent the pros and cons of the implementation scenarios. The above disclosures are only some specific implementation scenarios of the present invention, however, the present invention is not limited thereto, and any changes conceivable by those skilled in the art shall fall within the protection scope of the present invention.

Claims (10)

1. a kind of missing crew's facial images match method characterized by comprising

Image is acquired when intelligent wearable device is in running order；

Face datection is carried out to described image；

It will test the face of pre-stored missing crew in face and the intelligent wearable device in the target image of face Portion's image is matched, and determines the similarity between the face and the face-image of the missing crew；

When similarity between the face and missing crew's face-image is greater than threshold value, in the intelligent wearable device Upper sending prompting message.

2. the method according to claim 1, wherein the method also includes:

After carrying out Face datection to described image, the described image that face is not detected is deleted；

After determining the similarity between the face and the face-image of the missing crew, delete the face with it is described Similarity is not more than the described image of the threshold value between missing crew's face-image.

3. the method according to claim 1, wherein the method also includes:

After the intelligent wearable device acquires image, unique identification and/or geographical location label, institute are set for described image Geographical location label is stated for identifying the geographical location of shooting described image.

4. according to the method described in claim 3, it is characterized in that, described issue in the intelligent wearable device prompts to disappear Breath, comprising:

Prompted in the intelligent wearable device target image the unique identification and/or the geographical location label.

5. the method according to claim 1, wherein the method also includes:

When similarity between the face and missing crew's face-image is greater than the threshold value, notice and the intelligence The image and/or text information of the missing crew is presented at least one target device of wearable device binding, in the mesh The image and/or text information of the missing crew are presented on marking device.

6. the method according to claim 1, wherein the method also includes:

When similarity between the face and missing crew's face-image is greater than the threshold value, by the target image It is sent at least one target device with intelligent wearable device binding, the target to be presented on the target device Image.

7. method according to any one of claims 1 to 6, which is characterized in that the method also includes:

With the information for the missing crew that predetermined period is updated to server request；

The information of the missing crew found according to the information deletion of the missing crew of the update and/or save a nearest institute State in predetermined period increase newly missing crew information, the information of the missing crew include missing crew face-image and/ Text information.

8. a kind of missing crew's facial images match device characterized by comprising

Acquisition module, for acquiring image when intelligent wearable device is in running order；

Detection module, for carrying out Face datection to described image；

Matching module, face in target image for will test face with it is pre-stored in the intelligent wearable device The face-image of missing crew matches, and determines the similarity between the face and the face-image of the missing crew；

Cue module, when being greater than threshold value for the similarity between the face and missing crew's face-image, in institute It states and issues prompting message in intelligent wearable device.

9. a kind of missing crew's facial images match device characterized by comprising

Memory and processor are stored with computer-readable instruction in the memory, and the computer-readable instruction is described When processor executes, so that the processor executes the missing crew face as described in any one of claims 1 to 7 claim The step of portion's image matching method.

10. a kind of non-transitorycomputer readable storage medium, which is characterized in that when the instruction in the storage medium is by handling When device executes, enable a processor to execute according to claim 1 to the face of missing crew described in any one of 7 claims The step of image matching method.