CN113158953B

CN113158953B - Personnel searching method, device, equipment and medium

Info

Publication number: CN113158953B
Application number: CN202110480029.4A
Authority: CN
Inventors: 谢宇; 徐敏荣; 赵考鹏
Original assignee: Qingdao Hisense Smart Life Technology Co Ltd
Current assignee: Qingdao Hisense Smart Life Technology Co Ltd
Priority date: 2021-04-30
Filing date: 2021-04-30
Publication date: 2022-11-25
Anticipated expiration: 2041-04-30
Also published as: CN113158953A

Abstract

The embodiment of the application provides a person searching method, a person searching device, a person searching apparatus and a person searching medium, wherein after a person searching instruction is received, a first candidate person can be tracked within a set first time length to obtain a first motion track of the motion track, an image in the first motion track is matched with a target image of a person to be searched, and position information of the target person is determined.

Description

Personnel searching method, device, equipment and medium

Technical Field

The present application relates to the field of computer vision technologies, and in particular, to a method, an apparatus, a device, and a medium for searching for a person.

Background

With the rapid development of computer vision, in the case of missing people, it is more and more common to implement a method of determining whether an image acquired by an image acquisition device contains the missing people, in the prior art, in the process of searching people, face recognition is performed based on an obtained image currently acquired by the image acquisition device, if the distance between the acquired face and the image acquisition device is too large or a person is not facing the image acquisition device, that is, the face is too large in a facing position deflection angle relative to the image acquisition device, a situation that the currently acquired image is unclear occurs, and therefore, a target person to be searched can not be accurately identified based on a face recognition technology, and user experience is affected.

Disclosure of Invention

The application provides a personnel searching method, a personnel searching device, equipment and a medium, which are used for solving the problem that in the prior art, a target personnel to be searched cannot be accurately identified based on a face recognition technology, and user experience is influenced.

The application provides a personnel searching method, which comprises the following steps:

receiving a personnel searching instruction, wherein the personnel searching instruction carries a first target image of a personnel to be searched;

determining a first motion track comprising a motion track obtained by tracking a first candidate within a set first time length in a historical motion track obtained by tracking, and matching an image in the first motion track with the first target image;

and if the image successfully matched with the first target image exists, determining the position information of the target person according to the acquisition position of the image in the first motion track.

Further, the tracked historical motion trail includes:

if the first image is identified to contain the pedestrian, determining whether the first similarity between the sub-image of the pedestrian contained in the first image and the pre-stored image of the set person is larger than a preset first similarity threshold, and if the first similarity is determined to be larger than the preset first similarity threshold, determining whether the first confidence of the first image is larger than the preset first confidence threshold;

if so, tracking the pedestrian and storing the obtained motion track;

if not, tracking the pedestrian, determining whether a second image with a second confidence degree greater than a preset first confidence degree threshold exists in the tracking process, if so, determining that the second similarity between the second image and the image of the set person is greater than a preset first similarity threshold, continuing to track the pedestrian, and storing the obtained motion track.

Further, if there is a second image that is greater than a preset second confidence threshold, but a second similarity between the second image and the pre-saved image of the set person is not greater than a preset first similarity threshold, the method further includes:

stopping the tracking of the pedestrian and deleting the obtained motion trail.

Further, after saving the obtained motion trajectory, the method further includes:

determining the similarity between the image in the motion track and the image of the set person, and determining and storing the maximum value of the similarity as the target similarity;

and determining the confidence of the image in the motion track, and determining and storing the maximum value of the confidence as the target confidence.

Further, determining that there is an image that successfully matches the first target image comprises:

and for each first motion track, according to a first target similarity corresponding to the image in the first motion track and the first target image, which is saved in advance, if the first target similarity is determined to be greater than a preset second similarity threshold, determining that an image successfully matched with the first target image exists.

Further, before determining the position information of the target person according to the acquisition position of the image in the first motion trajectory after the image successfully matched with the first target image exists, the method further includes:

and if the pre-stored first target confidence corresponding to the first motion track is greater than a preset second confidence threshold, performing subsequent steps of determining the position information of the target person according to the acquisition position of the image in the first motion track.

Further, if the similarity of the first target is not greater than a preset second similarity threshold, or the confidence of the first target is not greater than a preset second confidence threshold, the method further includes:

sequencing second motion tracks with set second time length according to the sequence of the maximum value of the similarity from large to small, wherein the set second time length is larger than the set first time length;

searching a first set number of second motion tracks ranked in the front, and displaying a first candidate image corresponding to the maximum confidence value in the second motion tracks;

and if the first candidate image contains the target person according to the received input information, determining the position information of the target person according to the first candidate image and the second motion trail.

Further, the method further comprises:

if the first candidate image is determined not to contain the target person according to the received input information, searching third motion tracks which are sorted in a second set number except the second motion tracks in the first set number and are sorted in the first set number, and displaying a second candidate image corresponding to the maximum confidence value in the third motion tracks;

and if the second candidate image is determined to contain the target person according to the received input information, determining the position information of the target person according to the second candidate image containing the target person and the second motion track.

Further, the determining the position information of the target person according to the second candidate image containing the target person and the second motion trail includes:

updating the candidate image containing the target person into a second target image;

determining a second target similarity corresponding to the image in the second motion track and the second target image, which is stored in advance, aiming at the second motion track;

sequencing the second motion tracks according to the sequence of the maximum value of the similarity from large to small; searching a third set number of fourth motion tracks sequenced at the front, and sequencing the fourth motion tracks from late to early according to time;

searching a fourth set number of fifth motion tracks ranked in the front, and displaying a third candidate image corresponding to the maximum confidence value in the fifth motion tracks;

and if the third candidate image contains the target person according to the received input information, determining the position information of the target person according to the acquisition position containing the third candidate image.

Further, the method further comprises:

if the third candidate image is determined not to contain the target person according to the received input information, searching sixth motion tracks which are ranked in the first fifth set number except the fifth motion tracks in the fourth set number, and displaying a fourth candidate image corresponding to the maximum confidence value in the sixth motion tracks;

and if the fourth candidate image containing the target person is determined to contain the target person according to the received input information, determining the position information of the target person according to the acquisition position of the fourth candidate image containing the target person.

The application provides a personnel seek device, the device includes:

the searching module is used for receiving a personnel searching instruction, wherein the personnel searching instruction carries a first target image of a personnel to be searched;

the processing module is used for determining a first motion track which comprises a motion track obtained by tracking a first candidate in a set first time length in the tracked historical motion track, and matching an image in the first motion track with the first target image;

and the determining module is used for determining the position information of the target person according to the acquisition position of the image in the first motion track if the image successfully matched with the first target image exists.

Further, the processing module is specifically configured to determine, if it is recognized that a first image includes a pedestrian, whether a first similarity between a sub-image of the pedestrian included in the first image and a pre-stored image of a set person is greater than a preset first similarity threshold, and if it is determined that the first similarity is greater than the preset first similarity threshold, determine whether a first confidence of the first image is greater than a preset first confidence threshold; if so, tracking the pedestrian and storing the obtained motion track; if not, tracking the pedestrian, determining whether a second image with a second confidence degree greater than a preset first confidence degree threshold exists in the tracking process, if so, determining that the second similarity between the second image and the image of the set person is greater than a preset first similarity threshold, continuing to track the pedestrian, and storing the obtained motion track.

Further, the processing module is further configured to, if there is a second image that is greater than a preset second confidence threshold and a second similarity between the second image and the image of the pre-saved set person is not greater than a preset first similarity threshold, stop tracking the pedestrian, and delete the obtained motion trajectory.

Further, the determining module is further configured to determine similarity between the image in the motion trajectory and the image of the set person, and determine and store a maximum value of the similarity as a target similarity; and determining the confidence of the image in the motion track, determining the maximum value of the confidence as the target confidence and storing the target confidence.

Further, the processing module is specifically configured to, for each first motion trajectory, determine, according to a first target similarity corresponding to the first target image and an image in the first motion trajectory that is saved in advance, that an image that is successfully matched with the first target image exists if it is determined that the first target similarity is greater than a preset second similarity threshold.

Further, the determining module is further configured to perform subsequent steps of determining the position information of the target person according to the acquisition position of the image in the first motion trajectory if the pre-stored first target confidence corresponding to the first motion trajectory is greater than a preset second confidence threshold.

Further, the determining module is further configured to sort the second motion trajectories with the set second time length according to a sequence from a maximum similarity value to a minimum similarity value, where the set second time length is greater than the set first time length; searching a first set number of second motion tracks ranked in the front, and displaying a first candidate image corresponding to the maximum confidence value in the second motion tracks; and if the first candidate image contains the target person according to the received input information, determining the position information of the target person according to the first candidate image and the second motion trail.

Further, the determining module is further configured to, if it is determined according to the received input information that the first candidate image does not include the target person, search for a second set number of third motion trajectories, which are ranked in the front, except for the first set number of second motion trajectories, and display a second candidate image corresponding to a maximum confidence value in the third motion trajectories; and if the second candidate image is determined to contain the target person according to the received input information, determining the position information of the target person according to the second candidate image containing the target person and the second motion track.

Further, the determining module is specifically configured to update the candidate image including the target person to be a second target image; determining a second target similarity corresponding to the image in the second motion track and the second target image, which is stored in advance, aiming at the second motion track; sequencing the second motion tracks according to the sequence of the maximum value of the similarity from large to small; searching a third set number of fourth motion tracks sorted at the front, and sorting the fourth motion tracks according to the sequence from the late to the early of the time; searching a fourth set number of fifth motion tracks ranked in the front, and displaying a third candidate image corresponding to the maximum confidence value in the fifth motion tracks; and if the third candidate image contains the target person according to the received input information, determining the position information of the target person according to the acquisition position containing the third candidate image.

Further, the determining module is further configured to, if it is determined that the third candidate image does not include the target person according to the received input information, search for sixth motion trajectories, which are ranked in the top fifth set number, except for the fifth motion trajectories in the fourth set number, and display a fourth candidate image corresponding to a maximum value of confidence in the sixth motion trajectories; and if the fourth candidate image containing the target person is determined to contain the target person according to the received input information, determining the position information of the target person according to the acquisition position of the fourth candidate image containing the target person.

The present application provides an electronic device comprising a processor for implementing the steps of the person finding method as described in any one of the above when executing a computer program stored in a memory.

A computer-readable storage medium storing a computer program executable by a terminal, the program, when run on the terminal, causing the terminal to perform the steps of the person search method of any one of the above.

In the embodiment of the application, a person searching instruction is received, wherein the person searching instruction carries a first target image of a person to be searched, the first target image is matched with an image in a first motion track obtained by tracking a first candidate within a set first time length, and if an image successfully matched with the first target image exists, the position information of the target person is determined according to the acquisition position of the image in the first motion track.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.

Fig. 1 is a process diagram of a person search method according to some embodiments of the present application;

FIG. 2 is a schematic diagram of a process for determining a tracked motion trajectory according to some embodiments of the present application;

FIG. 3 is a flow diagram of person finding provided by some embodiments of the present application;

fig. 4 is a schematic structural diagram of a person searching apparatus according to some embodiments of the present application;

fig. 5 is a schematic structural diagram of an electronic device according to some embodiments of the present application.

Detailed Description

In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be described in further detail with reference to the accompanying drawings, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without making any creative effort belong to the protection scope of the present application.

In order to improve accuracy of person searching and improve user experience, embodiments of the present application provide a person searching method, apparatus, device, and medium.

Example 1:

fig. 1 is a schematic process diagram of a person searching method according to some embodiments of the present application, where the process includes the following steps:

s101: and receiving a person searching instruction, wherein the person searching instruction carries a first target image of a person to be searched.

The person searching method provided by the embodiment of the application is applied to electronic equipment, and the electronic equipment can be image acquisition equipment or equipment such as a server.

In the application, if the requirement of the person to be searched exists, a person searching instruction can be sent to the electronic device, wherein the person searching instruction carries information of the person to be searched, and the information of the person to be searched can be a first target image of the person to be searched. Specifically, images of a plurality of set persons may be pre-stored in the electronic device, and when there is a search requirement, an image of a person to be searched may be selected from the images of the plurality of set persons pre-stored in the electronic device, and the selected image is used as the first target image. Or sending a person searching instruction carrying the first target image of the person to be searched to the electronic device, and after receiving the person searching instruction, the electronic device determines the first target image of the person to be searched carried in the person searching instruction.

S102: determining a first motion track comprising a motion track obtained by tracking a first candidate in a set first time length in the tracked historical motion track, and matching an image in the first motion track with the first target image.

In the application, the electronic device tracks each person contained in the image based on the image acquired by the image acquisition device, and obtains the motion track of each person. After a person searching instruction is received, searching a first motion track containing a motion track obtained by tracking a first candidate person within a set first time length in the tracked motion tracks, wherein the first motion track contains all collected images containing the first candidate person, and the time for collecting each image is also determined. The trajectory tracking algorithm is the prior art, and is not described herein.

Because the personnel appearing in the monitoring range may be the personnel needing to be tracked and may also be the personnel not needing to be tracked, if the monitoring range is a cell, the service for tracking and searching is only opened for the personnel living in the cell, therefore, the electronic equipment can track the personnel living in the cell contained in the image acquired by the image acquisition equipment. Thus, the first candidate may be a person in which the cell resides.

After receiving the person search instruction, since the person to be searched may or may not be within a monitorable range at this time, in order to determine whether the person to be searched is within the monitorable range, in the present application, the first motion trajectory may be a motion trajectory obtained by tracking the first candidate person within a set first time length, where a time period corresponding to the set first time length is a time period before the person search instruction is received, and the time period corresponding to the set first time length is very close to the time of the current person to be searched, that is, the time of acquiring the latest acquired image in the first motion trajectory within the set first time length may default to be the same as the time of currently receiving the person search instruction. The set first time period may be 3 minutes or 5 minutes.

In order to ensure the integrity of the recognition, in the embodiment of the present application, the first motion trajectory includes all the motion trajectories corresponding to the set first time length as the first motion trajectory, that is, if any motion trajectory includes the motion trajectory corresponding to the set first time length, all the motion trajectories are the first motion trajectory. For example, in the process of tracking a pedestrian, an image acquisition device installed at a position a between 8.

The first candidate person is a pedestrian included in an image obtained by the electronic device after receiving the person searching instruction, wherein the first candidate person may or may not include the person to be searched.

In order to determine whether a person to be searched exists in the first candidate person, after determining the first target image, the electronic device matches the first target image with all images including a first motion trajectory of the motion trajectory obtained by tracking the first candidate person within the set first time length.

S103: and if the image successfully matched with the first target image exists, determining the position information of the target person according to the acquisition position of the image in the first motion track.

When the first target image is respectively matched with all the images in the first motion trajectory, if an image successfully matched with the first target image exists, the successfully matched image is determined to be an image of a person to be searched, and since the image acquisition device for acquiring the successfully matched image is known, that is, the acquisition position of the image in the first motion trajectory is also known, the position information of the target person can be determined based on the acquisition position of the image in the first motion trajectory, and specifically, the acquisition position of the image can be determined to be the position of the target person.

According to the embodiment of the application, after a person searching instruction is received, the first image in the first motion track can be matched with the target image of the person to be searched based on the first motion track obtained by tracking the first candidate person within the set first time length, and the position information of the target person can be determined.

Example 2:

in order to track the obtained historical motion trail, on the basis of the above embodiment, in the embodiment of the present application, the tracking the obtained motion trail includes:

if so, tracking the pedestrian and storing the obtained motion trail;

In the application, in order to obtain a motion trajectory with satisfactory definition, the electronic device stores an image of a setting person in advance, and immediately tracks the setting person after the image of the setting person is stored in the electronic device. Specifically, after the electronic device acquires the first image, and recognizes that the first image contains a pedestrian, the electronic device determines a sub-image of the pedestrian contained in the first image, and determines a first similarity between the sub-image containing the pedestrian and an image of a set person, wherein the pedestrian may or may not be the set person.

In order to determine whether the pedestrian is a set person, in the application, a first similarity threshold is preset and used for judging whether the first similarity is greater than the first similarity threshold, and if the first similarity is greater than the first similarity threshold, whether the first confidence of the first image is greater than a preset first confidence threshold is judged. If the first confidence coefficient of the image is greater than a preset first confidence coefficient threshold, the pedestrian is a set person, the pedestrian is tracked, namely a plurality of images of the pedestrian are collected, the collection time corresponding to each image of the pedestrian is recorded, and the motion track is determined and stored according to each image of the pedestrian and the collection time corresponding to each image.

The first confidence coefficient is used for representing the quality of the image, wherein the first confidence coefficient is the highest, and the image is clearer when the quality of the corresponding image is higher. In the process of determining the first confidence level, a target distance between the pedestrian and the image acquisition device may be determined based on a face recognition technology, and the first confidence level may be determined according to the target distance, where the smaller the target distance from the image acquisition device, the higher the confidence level is, and conversely, the larger the target distance from the image acquisition device, the lower the confidence level is. Specifically, a corresponding relationship between a distance range and a confidence level is pre-stored, and after the target distance between the pedestrian and the image acquisition device is identified based on the face recognition technology, a first confidence level corresponding to the target distance is determined according to the pre-stored corresponding relationship between the distance range and the confidence level. For example, if the target distance between the pedestrian and the image capturing device is within 3 meters, the first confidence is 0.98, if the target distance between the pedestrian and the image capturing device is between 3 meters and 4 meters, the first confidence is 0.9, and so on.

If the first confidence degree of the first image is not greater than the preset first confidence degree threshold, the pedestrian is possibly a set person or not, in order to determine whether the person is the set person, the pedestrian is tracked, whether a second image with a second confidence degree greater than the preset first confidence degree threshold exists or not is determined in the tracking process, and whether the pedestrian is the set person or not is determined according to the existence or not of the second image with the second confidence degree greater than the preset first confidence degree threshold.

If a second image larger than a preset first confidence threshold exists and the second similarity of the second image is determined to be larger than a preset first similarity threshold, the pedestrian is a set person, therefore, the pedestrian is tracked, namely, a plurality of images including the pedestrian are collected, the collection time corresponding to each image of the pedestrian is recorded, and the motion track is determined and stored according to the collection time corresponding to the plurality of images of the pedestrian and each image of the pedestrian.

In addition, in the process of determining the motion trail, if the electronic device is an image acquisition device, the electronic device tracks the pedestrian, acquires a plurality of images including the pedestrian, records acquisition time of each image, and determines the motion trail according to the plurality of images and the acquisition time of the plurality of images. If the electronic equipment is a server, the image acquisition equipment acquires a plurality of images containing the pedestrian and records the acquisition time of each image, the image acquisition equipment sends the plurality of images and the acquisition time of the plurality of images to the electronic equipment, and the electronic equipment determines and stores the motion track according to the received images and the acquisition time corresponding to each image.

In order to determine the movement track, on the basis of the foregoing embodiments, in an embodiment of the present application, if there exists a second image that is greater than a preset first confidence threshold, and a second similarity between the second image and the image of the pre-saved set person is not greater than a preset first similarity threshold, the method further includes:

stopping the tracking of the pedestrian, and deleting the obtained motion trail.

If the second image which is larger than the preset first confidence coefficient threshold does not exist, that is, the second image which is larger than the first confidence coefficient threshold is not found in a certain time length, or the second image which is larger than the preset first confidence coefficient threshold exists, but the second similarity of the second image and the image of the pre-stored set person is not larger than the preset first similarity threshold, it is indicated that the pedestrian is not the pre-stored set person, the pedestrian does not need to be tracked continuously, therefore, the tracking is finished, and the motion track is deleted.

On the basis of the foregoing embodiments, in this embodiment of the present application, after tracking the pedestrian, before storing the obtained motion trajectory, the method further includes:

In the embodiment of the application, in the process of tracking the pedestrians, for each pedestrian, because the obtained images including the pedestrian are different, the similarity between each image and the image of the preset person is different, and the confidence degree corresponding to each image is also different. Since the greater the similarity is, the closer the image containing the pedestrian is to the image of the preset person, and the higher the confidence is, the higher the quality of the image corresponding to the image with the greater confidence is, therefore, in order to facilitate the subsequent search of the target person for the motion trajectory, the maximum value of the similarity and the maximum value of the confidence can be saved in the process of tracking the pedestrian. Specifically, when a pedestrian is tracked, the similarity between the image in the motion trail and the image of the set person is determined, and the maximum value of the similarity is determined as the target similarity and stored; and determining the confidence of the image in the motion track, and determining and storing the maximum value of the confidence as the target confidence.

Fig. 2 is a schematic diagram of a process for determining a tracked motion trajectory according to some embodiments of the present application, and will now be described with reference to fig. 2:

after the image of the set person is stored in the electronic device, the electronic device tracks the set person, specifically, the electronic device starts to detect after acquiring the image, if a pedestrian is detected, if the electronic device is assumed to be an image acquisition device, the electronic device acquires the image containing the pedestrian, a first similarity between the image of the pedestrian and the image of the set person stored in advance is determined, and if the first similarity is greater than a first similarity threshold rho, the electronic device tracks the set person, and specifically, the electronic device starts to detect the image, and if the pedestrian is detected, the electronic device acquires the image containing the pedestrian, the first similarity between the image of the pedestrian and the image of the set person is determined, and if the first similarity is greater than a first similarity threshold rho _thre Then it is determined whether the first confidence of the image is greater than a second confidence threshold c _thre If so, continuously tracking, determining the similarity and confidence degree of the image containing the pedestrian and other images acquired in the tracking process, storing the maximum value C and the maximum value rho of the confidence degree, and if the tracking is interrupted in the tracking process, ending the tracking and keeping the motion track.

If the first similarity is larger than a first similarity threshold rho _thre But determining that the first confidence level of the image is not greater than the first confidence threshold c _thre Then continuously tracking, and determining to collect in the tracking processAnd if the second confidence level is higher than the first confidence level threshold c in the tracking process _thre Determining whether the second similarity between the image and the image of the set person is greater than a first similarity threshold value rho _thre If not, ending the tracking and deleting the record. If the second confidence coefficient is larger than the first confidence coefficient threshold c _thre And the second confidence is greater than the first confidence threshold c _thre The second similarity between the image (b) and the pre-stored image of the set person is greater than the first similarity threshold rho _thre And continuously tracking, determining the similarity and the confidence degree of the image containing the pedestrian and other images acquired in the tracking process, storing the maximum value C of the confidence degree and the maximum value rho of the similarity, and ending the tracking and keeping the motion track if the tracking is interrupted in the tracking process.

Example 3:

in order to determine an image in the first motion trajectory, which is successfully matched with the first target image, on the basis of the foregoing embodiments, in an embodiment of the present application, determining that an image successfully matched with the first target image exists includes:

In the present application, in the process of matching the first target image with the image in the first motion trajectory, for each first motion trajectory, because the maximum value of the similarity between the image in the motion trajectory and the image of the set person in the tracking process is stored in advance for each motion trajectory, the target similarity is stored. The greater the similarity of the first target is, the higher the matching degree between the image in the first motion trajectory and the first target image is, the greater the possibility that the target person is found based on the frame of image is.

In order to determine whether an image matched with the first target image exists in the first motion track, in the present application, a second similarity threshold is preset, a first target similarity corresponding to the first motion track is compared with the preset second similarity threshold, if the first target similarity is greater than the preset second similarity threshold, it is determined that an image successfully matched with the first target image exists in the first motion track, that is, the first motion track may be a track corresponding to a target person, and if the first target similarity is less than the preset second similarity threshold, it is determined that an image successfully matched with the first target image does not exist in the first track, that is, the first track may not be a track corresponding to the target person.

Example 4:

in order to screen high-quality images and further improve accuracy of person searching, on the basis of the foregoing embodiments, in this embodiment of the application, before determining the position information of the target person according to the acquisition position of the image in the first motion trajectory after the image successfully matched with the first target image exists, the method further includes:

if the first target confidence corresponding to the first motion track which is preserved in advance is larger than a preset second confidence threshold, performing subsequent steps of determining the position information of the target person according to the acquisition position of the image in the first motion track.

After determining that an image successfully matched with the first target image exists in the first motion trajectory, in order to screen out an image with the highest quality, and thus determine whether the image is an image of a target person based on the image with the highest quality, in the present application, a first target confidence corresponding to the first motion trajectory is pre-stored, so as to determine whether the image in the first motion trajectory has an image with high quality, that is, to determine whether the image in the first motion trajectory has an image with high confidence, in the present application, a second confidence threshold is pre-set, and it is determined whether the first target confidence is greater than the pre-set second confidence threshold. Specifically, the position information of the target person is determined according to the position of the image acquisition device acquiring the target image.

Example 5:

in order to determine the location information of the target person, on the basis of the foregoing embodiments, in this application embodiment, if the first target similarity is not greater than a preset second similarity threshold, or the first target confidence is not greater than a preset second confidence threshold, the method further includes:

In the application, after a person searching instruction is received, if the target person is not searched based on the first motion trajectory with the first time length, the time range may be slightly extended, and the target person is searched based on the second motion trajectory with the second time length.

In order to ensure the integrity of the recognition, in this embodiment of the present application, the second motion trajectory includes the motion trajectory corresponding to the set second time length as the second motion trajectory, that is, if any motion trajectory includes the motion trajectory corresponding to the set second time length, all of the motion trajectories are the second motion trajectory, where the second motion trajectory of the set second time length includes the first motion trajectory within the set first time length, that is, the set second time length is greater than the set first time length.

For example, in the process of tracking a pedestrian, when an image acquisition device installed at a position a between 8.

In order to increase the accuracy of searching for the person and improve the efficiency of searching for the person, in the present application, the second motion trajectories with the second time length set are sorted from large to small according to the maximum value of the similarity, and the second motion trajectories with the first set number that are sorted before are searched for, where the first set number may be 3, 4, 5, and the like, and specifically, the first set number may be set according to a requirement.

After the first set number of second motion trajectories are determined, because multiple images exist in each second motion trajectory, in order to facilitate a user to determine whether the images contain target people or not, and improve user experience, the image with the highest quality in each second motion trajectory may be determined to be displayed, that is, the first candidate image corresponding to the maximum confidence value in the first set number of second motion trajectories is displayed, and because the target confidence values corresponding to the second motion trajectories are stored in advance, the first candidate image corresponding to the target confidence values in each second motion trajectory may be directly displayed. The number of the displayed first candidate images is the same as the number of the selected second motion tracks, and the first candidate images are all the first set number, and after the first set number of the first candidate images are displayed, the first candidate images are provided for the user to confirm.

Since the target person may or may not be included in the first candidate images of the first set number, the first candidate images of the first set number may be displayed on the display page of the electronic device, the user may select the image including the target person based on the first candidate images of the first set number, and the electronic device may receive the input information carrying the identification information of the selected first candidate image.

The electronic equipment receives the input information, determines a first candidate image containing a target person according to identification information of the first candidate image carried in the input information, and determines position information of the target person according to the first candidate image containing the target person and a corresponding second motion track. Specifically, the target image including the target person and closest to the time when the person search instruction is received in the second motion trajectory is determined according to the first candidate image including the target person, and the position information of the target person is determined according to the position of the image acquisition device acquiring the target image.

In order to determine the location information of the target person, on the basis of the foregoing embodiments, in an embodiment of the present application, the method further includes:

If the displayed first candidate image does not contain the target person, the input information does not carry any identification information, and the electronic equipment receives the input information and determines that the first candidate image does not contain the target person. In order to determine the position information of the target person, it may be continued to search whether the images of the other motion trajectories except the second motion trajectory include the target person, and in this application, third motion trajectories except the first set number of second motion trajectories and sorted in the first set number are searched.

For example, there are 9 second motion trajectories, and the second motion trajectories are sequentially ranked as a second motion trajectory 1, a second motion trajectory 2, a second motion trajectory 3, a second motion trajectory 4, a second motion trajectory 5, a second motion trajectory 6, a second motion trajectory 7, a second motion trajectory 8, and a second motion trajectory 9 according to the descending order of the maximum similarity. Assuming that the first set number is 3, and it is determined that the second motion trajectory 1, the second motion trajectory 2, and the image of the second motion trajectory 3 do not include the target person, and assuming that the preset second number is 2, it is determined that the second motion trajectory 4 and the second motion trajectory 5 are the third motion trajectories.

After the second set number of third motion trajectories are determined, because a plurality of images exist in each third motion trajectory, in order to facilitate a user to determine whether the images contain target people or not and improve user experience, the images with the highest quality in each third motion trajectory may be determined to be displayed, that is, sorted according to the maximum confidence values, and second candidate images corresponding to the maximum confidence values in the third motion trajectories are displayed. And the number of the displayed second candidate images is a second set number, and the second candidate images of the second set number are displayed for the user to confirm.

Since the second candidate images of the second set number may or may not include the target person, the second candidate images of the second set number may be displayed on the display page of the electronic device, the user may select the image including the target person based on the second candidate images of the second set number, and the electronic device may receive the input information carrying the identification information of the selected second candidate image.

The electronic equipment receives the input information, determines a second candidate image containing the target person according to the identification information of the second candidate image carried in the input information, and determines the position information of the target person according to the second candidate image containing the target person and the corresponding second motion track. Specifically, the target image which is closest to the time when the person search instruction is received and contains the target person in the second motion trajectory is determined according to the second candidate image containing the target person, the position of the image acquisition device for acquiring the target image is determined, and the position information of the target person is determined.

Example 6:

in order to determine the position information of the target person, on the basis of the foregoing embodiments, in an embodiment of the present application, the determining the position information of the target person according to the selected image including the target person and the second motion trajectory includes:

sequencing the second motion tracks according to the sequence of the maximum value of the similarity from large to small; searching a third set number of fourth motion tracks sorted at the front, and sorting the fourth motion tracks according to the sequence from the late to the early of the time;

After the target person is determined to exist based on the first candidate images in the second motion tracks or the second candidate images in the third motion tracks of the first set number, the target image which contains the target person and is closer to the time of receiving a person searching instruction in the second motion track is determined, and the probability of searching the person to be searched is higher based on the position of the image acquisition equipment for acquiring the target image. For convenience of description, the candidate image including the target person is updated to be a second target image, and in order to determine the target image including the target person closest to the time when the person search instruction is received, in the present application, for all second motion trajectories, a second target similarity between an image in the second motion trajectory and the second target image is determined, where the greater the second target similarity, the higher the matching degree between the image included in the second motion trajectory and the second target image is.

In this application, images similar to the second target image may be screened out from the second motion trajectory, and then, images closest to the current time are screened out from the images similar to the second target image.

After the fourth motion trajectories with the third set number are determined, in order to determine motion trajectories closer to the current time, the fourth motion trajectories are sequenced from late to early according to time, and fifth motion trajectories with the fourth set number that are sequenced before are searched for.

Because a plurality of images exist in each fifth motion trajectory, in order to facilitate a user to confirm whether the image contains a target person or not, and improve user experience, it may be determined that each image with the highest quality of the fifth motion trajectory is displayed, that is, the third candidate image corresponding to the maximum confidence value in the fifth motion trajectories of the fourth set number is displayed, that is, the image corresponding to the maximum confidence value in the fifth motion trajectories of the fourth set number is determined as the third candidate image and displayed. And the number of the displayed third candidate images is a fourth set number, and the third candidate images of the fourth set number are displayed for the user to confirm.

Since the target person may be included in the third candidate images with the fourth set number or may not be included in the third candidate images with the fourth set number, the third candidate images with the fourth set number may be displayed on the display page of the electronic device, the user may select the image including the target person based on the third candidate images with the fourth set number, and the electronic device may receive the input information carrying the identification information of the first candidate image with the alternative selection.

The electronic equipment receives the input information, determines a third candidate image containing a target person according to identification information of the third candidate image carried in the input information, determines position information of the target person according to an acquisition position of the third candidate image containing the target person, namely, determines a target image which is closest to the time of receiving a person searching instruction in a fifth motion track of the third candidate image containing the target person, and determines the position information of the target person according to the position of image acquisition equipment for acquiring the target image.

If the displayed third candidate image does not contain the target person, the input information does not carry any identification information, and the electronic equipment receives the input information and determines that the third candidate image does not contain the target person. In order to determine the position information of the target person, it may be continued to search whether the images of the other motion trajectories except the fifth motion trajectory include the target person, and in this application, the sixth motion trajectories except the fifth motion trajectories of the fourth set number are searched for in the fifth set number that is ranked before.

For example, there are 10 second motion trajectories, which are motion trajectory 1, motion trajectory 2, motion trajectory 3, motion trajectory 4, motion trajectory 5, motion trajectory 6, motion trajectory 7, motion trajectory 8, motion trajectory 9, and motion trajectory 10. Determining a second similarity between the image in the second track and the second target image, and sorting the images into a motion track 2, a motion track 3, a motion track 1, a motion track 7, a motion track 8, a motion track 6, a motion track 5, a motion track 4, a motion track 9 and a motion track 10 according to the maximum similarity from small to large, wherein if the third set number is 6, the fourth motion track is the motion track 2, the motion track 3, the motion track 1, the motion track 7, the motion track 8 and the motion track 6. Then the motion trajectories 1, 3, 2, 6, 8, and 7 are ordered according to time from late to early, and assuming that the fourth set number is 4, the fifth motion trajectory is the motion trajectory 1, 3, 2, and 6.

Therefore, in determining that the image in the fifth motion trajectory does not include the target person, it is determined whether the target person is included in the motion trajectories 7, 8, 5, 4, 9, and 10, and specifically, if the fifth set number is 3, the sixth motion trajectory is the motion trajectory 7, 8, and 5.

After the fifth set number of sixth motion trajectories are determined, because each sixth motion trajectory has multiple images, in order to facilitate the user to determine whether the images include the target person, and improve user experience, it may be determined that each sixth motion trajectory has the highest quality, that is, the images are sorted according to the maximum confidence value, and fourth candidate images corresponding to the maximum confidence values of the sixth motion trajectories of the fifth set number are displayed, where the number of the displayed fourth candidate images is the fifth set number, and after the fourth candidate images of the fifth set number are displayed, the fourth candidate images are provided for the user to determine.

Since the fifth set number of fourth candidate images may or may not include the target person, the fifth set number of fourth candidate images may be displayed on the display page of the electronic device, the user may select an image including the target person based on the fifth set number of fourth candidate images, and the electronic device may receive input information carrying identification information of the selected fourth candidate image.

The electronic device receives the input information, determines a fourth candidate image containing a target person according to identification information of the fourth candidate image carried in the input information, determines position information of the target person according to an acquisition position of the fourth candidate image containing the target person, determines a target image in a sixth motion track where the fourth candidate image containing the target person is located, the time of the fourth candidate image being closest to the time of receiving a person searching instruction, and determines the position information of the target person according to the position of an image acquisition device for acquiring the target image, so as to ensure the accuracy of the confirmation of the position information of the target person.

Fig. 3 is a flow chart of person finding provided by some embodiments of the present application, and is now described with respect to fig. 3:

after searching is initiated, that is, after the electronic device receives a person searching instruction, determining whether the confidence and the similarity of a motion track at the current time are greater than threshold values, that is, determining whether a first motion track containing a motion track obtained by tracking a first candidate person within a set first time length is included, determining whether the maximum value of the similarity between an image in the first motion track and a first target image with the searched person is greater than a preset first similarity threshold value, and whether the maximum value of the confidence is greater than a preset first confidence threshold value, if so, determining that the image in the first motion track contains the target person, and returning to the current picture of the track, that is, returning to the image with the latest time in the first motion track.

If the maximum value of the similarity of the images in the first motion trajectory is not greater than a preset first similarity threshold or the maximum value of the confidence is not greater than a preset first confidence threshold, determining whether a target person exists in the images included in the second motion trajectory within a second time length, specifically, displaying the images with the highest respective effects of the plurality of trajectories with the highest similarity, that is, sorting the second motion trajectories according to the sequence of the similarity from the largest to the smallest, determining to display the first candidate images corresponding to the maximum confidence in the first set number of second motion trajectories, if the user determines that the target exists, selecting the first candidate images including the target person by the user, if the user determines that the target does not exist, determining whether an alternative path exists, if so, returning to display the images with the highest respective effects of the plurality of trajectories with the second highest similarity, that is, displaying the third set number of third motion trajectories with the highest respective effects in the second motion trajectory except the first set number of second motion trajectories, and determining whether the third candidate images corresponding to the maximum confidence in the second motion trajectory correspond to the target person.

If the target person is determined to be included in the second candidate image, determining the image with the highest confidence coefficient in the second motion track closest to the current time in the second motion tracks, so that the respective best images of the tracks with the highest similarity and the closest time to the current time with the second target image are returned, that is, the second motion tracks are ranked according to the order from the maximum value of the similarity to the minimum, the fourth motion tracks with the third set number ranked in the front are searched, the fourth motion tracks with the fourth set number ranked in the front are ranked according to the order from the late to the early, the fifth motion tracks with the fourth set number ranked in the front are searched, and the third candidate image corresponding to the maximum value of the confidence coefficient in the fifth motion tracks is displayed. If the third candidate image contains the target person, that is, the target is determined to be in the third candidate image, the user selects the third candidate image containing the target person, and returns and displays the image of the third candidate image containing the target person closest to the current time, or returns the real-time video corresponding to the third candidate image containing the target person, and may determine the position information of the target person according to the acquisition position containing the third candidate image, that is, may determine the position information of the target person according to the target image closest to the current time in the fifth motion trajectory of the third candidate image containing the target person, and according to the position information of the image acquisition device acquiring the target image.

If the third candidate image does not contain the target person, returning an image with the best effect of each of the tracks with the second highest similarity, namely, searching the sixth motion tracks with the fifth set number, which are sequenced in the front, except the fifth motion tracks with the fourth set number, displaying the fourth candidate image corresponding to the maximum confidence value in the sixth motion tracks, determining that the fourth candidate image contains the target person, and if so, performing the step of selecting the fourth candidate image containing the target person by the user and returning the current fourth candidate image or the real-time video.

Example 7:

fig. 4 is a schematic structural diagram of a person searching apparatus according to some embodiments of the present application, where the apparatus includes:

the searching module 401 is configured to receive a person searching instruction, where the person searching instruction carries a first target image of a person to be searched;

a processing module 402, configured to determine, in a historical tracked motion trajectory, a first motion trajectory including a motion trajectory obtained by tracking a first candidate within a set first time length, and match an image in the first motion trajectory with the first target image;

a determining module 403, configured to determine, if an image successfully matched with the first target image exists, the position information of the target person according to the acquisition position of the image in the first motion trajectory.

In a possible implementation manner, the processing module 402 is specifically configured to determine, if it is identified that a first image includes a pedestrian, whether a first similarity between a sub-image of the pedestrian included in the first image and an image of a preset person is greater than a preset first similarity threshold, and if it is determined that the first similarity is greater than the preset first similarity threshold, determine whether a first confidence of the first image is greater than a preset first confidence threshold; if so, tracking the pedestrian and storing the obtained motion track; if not, the pedestrian is tracked, whether a second image with a second confidence coefficient larger than a preset first confidence coefficient threshold exists in the tracking process is determined, if yes, and the second similarity of the second image and the image of the set person is determined to be larger than the preset first similarity threshold, the pedestrian is continuously tracked, and the obtained motion track is stored.

In a possible implementation manner, the processing module 402 is further configured to, if there is a second image that is greater than a preset second confidence threshold and a second similarity between the second image and the image of the pre-saved set person is not greater than a preset first similarity threshold, stop tracking the pedestrian, and delete the obtained motion trajectory.

In a possible implementation manner, the determining module 403 is further configured to determine similarity between the image in the motion trajectory and the image of the set person, determine a maximum value of the similarity as a target similarity, and store the target similarity; and determining the confidence of the image in the motion track, and determining and storing the maximum value of the confidence as the target confidence.

In a possible implementation manner, the processing module 402 is specifically configured to, for each first motion trajectory, determine that there is an image successfully matched with the first target image if it is determined that the first target similarity is greater than a preset second similarity threshold, according to a first target similarity corresponding to an image in the first motion trajectory and the first target image that is stored in advance.

In a possible implementation manner, the determining module 403 is further configured to, if the first target confidence corresponding to the first motion trajectory stored in advance is greater than a preset second confidence threshold, perform subsequent steps of determining the position information of the target person according to the acquisition position of the image in the first motion trajectory.

In a possible implementation manner, the determining module 403 is further configured to sort the second motion trajectories for a set second time length according to a descending order of the maximum similarity value, where the set second time length is greater than the set first time length; searching a first set number of second motion tracks ranked in the front, and displaying a first candidate image corresponding to the maximum confidence value in the second motion tracks; and if the first candidate image contains the target person according to the received input information, determining the position information of the target person according to the first candidate image and the second motion trail.

In a possible implementation manner, the determining module 403 is further configured to, if it is determined that the first candidate image does not include the target person according to the received input information, search for a second set number of third motion trajectories, which are ranked in the first set number, except for the first set number of second motion trajectories, and display a second candidate image corresponding to a maximum confidence value in the third motion trajectories; and if the second candidate image is determined to contain the target person according to the received input information, determining the position information of the target person according to the second candidate image containing the target person and the second motion track.

In a possible implementation, the determining module 403 is specifically configured to update the candidate image including the target person to be a second target image; determining a second target similarity corresponding to the image in the second motion track and the second target image, which is stored in advance, aiming at the second motion track; sequencing the second motion tracks according to the sequence of the maximum value of the similarity from large to small; searching a third set number of fourth motion tracks sorted at the front, and sorting the fourth motion tracks according to the sequence from the late to the early of the time; searching a fourth set number of fifth motion tracks ranked in the front, and displaying a third candidate image corresponding to the maximum confidence value in the fifth motion tracks; and if the third candidate image contains the target person according to the received input information, determining the position information of the target person according to the acquisition position containing the third candidate image.

In a possible implementation manner, the determining module 403 is further configured to, if it is determined that the third candidate image does not include the target person according to the received input information, search for sixth motion trajectories, which are ranked in the first fifth set number, except for the fifth motion trajectories in the fourth set number, and display a fourth candidate image corresponding to a maximum confidence value in the sixth motion trajectories; and if the fourth candidate image containing the target person is determined to contain the target person according to the received input information, determining the position information of the target person according to the acquisition position of the fourth candidate image containing the target person.

Example 8:

on the basis of the foregoing embodiments, some embodiments of the present application further provide an electronic device, as shown in fig. 5, including: the system comprises a processor 501, a communication interface 502, a memory 503 and a communication bus 504, wherein the processor 501, the communication interface 502 and the memory 503 are communicated with each other through the communication bus 504.

The memory 503 has stored therein a computer program which, when executed by the processor 501, causes the processor 501 to perform the steps of:

Further, the processor 501 is further configured to determine, if the first image includes a pedestrian, whether a first similarity between a sub-image of the pedestrian included in the first image and a pre-stored image of a set person is greater than a preset first similarity threshold, and if the first similarity is greater than the preset first similarity threshold, determine whether a first confidence of the first image is greater than a preset first confidence threshold; if so, tracking the pedestrian and storing the obtained motion trail; if not, tracking the pedestrian, determining whether a second image with a second confidence degree greater than a preset first confidence degree threshold exists in the tracking process, if so, determining that the second similarity between the second image and the image of the set person is greater than a preset first similarity threshold, continuing to track the pedestrian, and storing the obtained motion track.

Further, the processor 501 is further configured to, if there is a second image that is greater than a preset second confidence threshold and a second similarity between the second image and the image of the pre-saved set person is not greater than a preset first similarity threshold, stop tracking the pedestrian, and delete the obtained motion trajectory.

Further, the processor 501 is further configured to determine similarity between the image in the motion trajectory and the image of the set person, and determine and store a maximum value of the similarity as a target similarity; and determining the confidence of the image in the motion track, and determining and storing the maximum value of the confidence as the target confidence.

Further, the processor 501 is further configured to, for each first motion trajectory, according to a first target similarity corresponding to a pre-stored image in the first motion trajectory and the first target image, determine that an image successfully matched with the first target image exists if it is determined that the first target similarity is greater than a preset second similarity threshold.

Further, the processor 501 is further configured to, if the first target confidence corresponding to the first motion trajectory that is stored in advance is greater than a preset second confidence threshold, perform subsequent steps of determining the position information of the target person according to the acquisition position of the image in the first motion trajectory.

Further, the processor 501 is further configured to sort the second motion trajectories with a second set time length according to a descending order of the maximum similarity value if the first target similarity is not greater than a preset second similarity threshold, or the first target confidence is not greater than a preset second confidence threshold, where the second set time length is greater than the first set time length; searching a first set number of second motion tracks ranked in the front, and displaying a first candidate image corresponding to the maximum confidence value in the second motion tracks; and if the first candidate image contains the target person according to the received input information, determining the position information of the target person according to the first candidate image and the second motion trail.

Further, the processor 501 is further configured to, if it is determined that the first candidate image does not include the target person according to the received input information, search a second set number of third motion trajectories, which are ranked in the front, except for the first set number of second motion trajectories, and display a second candidate image corresponding to a maximum confidence value in the third motion trajectories; and if the second candidate image is determined to contain the target person according to the received input information, determining the position information of the target person according to the second candidate image containing the target person and the second motion track.

Further, the processor 501 is further configured to update the candidate image including the target person to be a second target image; aiming at a second motion track, determining a second target similarity corresponding to an image in the second motion track and the second target image, wherein the second target similarity is stored in advance; sequencing the second motion tracks according to the sequence of the maximum value of the similarity from large to small; searching a third set number of fourth motion tracks sorted at the front, and sorting the fourth motion tracks according to the sequence from the late to the early of the time; searching a fourth set number of fifth motion tracks ranked in the front, and displaying a third candidate image corresponding to the maximum confidence value in the fifth motion tracks; and if the third candidate image contains the target person according to the received input information, determining the position information of the target person according to the acquisition position containing the third candidate image.

Further, the processor 501 is further configured to, if it is determined that the third candidate image does not include the target person according to the received input information, search for sixth motion trajectories, which are ranked in the first fifth set number, except for the fifth motion trajectories in the fourth set number, and display a fourth candidate image corresponding to a maximum confidence value in the sixth motion trajectories; and if the fourth candidate image containing the target person is determined to contain the target person according to the received input information, determining the position information of the target person according to the acquisition position of the fourth candidate image containing the target person.

The communication bus mentioned in the above server may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface 502 is used for communication between the above-described electronic apparatus and other apparatuses.

The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Alternatively, the memory may be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, including a central processing unit, a Network Processor (NP), and the like; but may also be a Digital instruction processor (DSP), an application specific integrated circuit, a field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like.

Example 9:

on the basis of the foregoing embodiments, an embodiment of the present application further provides a computer-readable storage medium, where a computer program executable by an electronic device is stored in the computer-readable storage medium, and when the program is run on the electronic device, the electronic device is caused to perform the following steps:

the memory has stored therein a computer program that, when executed by the processor, causes the processor to perform the steps of:

Further, the tracking the obtained historical motion track includes:

if so, tracking the pedestrian and storing the obtained motion trail;

Further, the method further comprises:

and if the second candidate image is determined to contain the target person according to the received input information, determining the position information of the target person according to the second candidate image containing the target person and the second motion trail.

Further, the method further comprises:

and if the fourth candidate image containing the target person is determined according to the received input information, determining the position information of the target person according to the acquisition position of the fourth candidate image containing the target person.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims

1. A people searching method, the method comprising:

if an image successfully matched with the first target image exists, determining the position information of the target person according to the acquisition position of the image in the first motion track;

wherein determining that there is an image that successfully matches the first target image comprises:

for each first motion track, according to a first target similarity corresponding to the image in the first motion track and the first target image, which is stored in advance, if the first target similarity is determined to be greater than a preset second similarity threshold, determining that an image successfully matched with the first target image exists;

wherein, after the image successfully matched with the first target image exists, before the position information of the target person is determined according to the acquisition position of the image in the first motion trajectory, the method further comprises:

if the pre-stored first target confidence corresponding to the first motion track is larger than a preset second confidence threshold, performing subsequent steps of determining the position information of the target personnel according to the acquisition position of the image in the first motion track;

if the first target similarity is not greater than a preset second similarity threshold, or the first target confidence is not greater than a preset second confidence threshold, the method further includes:

2. The method of claim 1, further comprising:

3. The method of claim 2, wherein determining the location information of the target person according to the second candidate image containing the target person and the second motion trajectory comprises:

and if the third candidate image containing the target person is determined according to the received input information, determining the position information of the target person according to the acquisition position of the third candidate image containing the target person.

4. The method of claim 3, further comprising:

5. A people finding apparatus, characterized in that the apparatus comprises:

the determining module is used for determining the position information of the target person according to the acquisition position of the image in the first motion track if the image successfully matched with the first target image exists;

the processing module is specifically configured to determine, for each first motion trajectory, that there is an image successfully matched with the first target image if it is determined that the first target similarity is greater than a preset second similarity threshold, according to a first target similarity corresponding to an image in the first motion trajectory and the first target image, which is stored in advance;

the determining module is further configured to perform subsequent steps of determining the position information of the target person according to the acquisition position of the image in the first motion trajectory if the first target confidence corresponding to the first motion trajectory which is stored in advance is greater than a preset second confidence threshold;

the determining module is further configured to sequence the second motion trajectories with the set second time length according to a descending order of the maximum similarity value, where the set second time length is longer than the set first time length; searching a first set number of second motion tracks ranked in the front, and displaying a first candidate image corresponding to the maximum confidence value in the second motion tracks; and if the first candidate image contains the target person according to the received input information, determining the position information of the target person according to the first candidate image and the second motion trail.

6. An electronic device, characterized in that the electronic device comprises a processor for implementing the steps of the method according to any of claims 1-4 when executing a computer program stored in a memory.

7. A computer-readable storage medium, characterized in that it stores a computer program executable by a terminal, which program, when run on the terminal, causes the terminal to carry out the steps of the method according to any one of claims 1-4.