CN112966652A - Trajectory convergence method and device, computer equipment and storage medium - Google Patents
Trajectory convergence method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN112966652A CN112966652A CN202110324686.XA CN202110324686A CN112966652A CN 112966652 A CN112966652 A CN 112966652A CN 202110324686 A CN202110324686 A CN 202110324686A CN 112966652 A CN112966652 A CN 112966652A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- image
- target
- track
- representative
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 85
- 238000013507 mapping Methods 0.000 claims description 61
- 230000002776 aggregation Effects 0.000 claims description 32
- 238000004220 aggregation Methods 0.000 claims description 32
- 238000004590 computer program Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 abstract description 31
- 238000012545 processing Methods 0.000 abstract description 23
- 230000009471 action Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
Abstract
The embodiment of the disclosure provides a track convergence method, a track convergence device, computer equipment and a storage medium, wherein the method comprises the following steps: determining representative images respectively corresponding to a plurality of pedestrian image cluster clusters, and determining a representative image in the pedestrian feature ratio of corresponding pedestrian features in the representative images to a pedestrian feature ratio of a target pedestrian as a target representative image corresponding to the target pedestrian; and combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergent track of the target pedestrian. In the image comparison process, the representative image of the pedestrian image cluster is adopted, so that the interference of the interference image to the comparison process is reduced, the accuracy of the comparison result can be improved, and the representative image of the pedestrian image cluster is adopted in the comparison process, so that the comparison processing of a large number of pedestrian images is not required, the efficiency of the comparison process is improved, and the generation efficiency of the convergence track is further improved.
Description
Technical Field
The present disclosure relates to the field of image processing technologies, and in particular, to a trajectory aggregation method and apparatus, a computer device, and a storage medium.
Background
With the progress of society and the development of technology, people have entered the information age, and pay more and more attention to the importance of information, and with the help of various information related to users, users can be analyzed to know the preferences and the change trend of users, for example, by analyzing the action tracks of users, the behavior hobbies, the travel modes, the common travel places and the like of users can be known.
For the action track of the user, a common statistical method is to collect the image of the user to converge the track for the user, but with the diversification and generalization of the information collection method, the information of the user that can be collected is exponentially increased, for example, in an actual scene, often a pedestrian can be captured to the image by the same camera or different cameras in the same place or a small range area for many times at a certain moment or a certain short time period, the data amount to be processed is large, the processing efficiency is low, and the operation load is large.
Disclosure of Invention
The embodiment of the disclosure at least provides a track convergence method, a track convergence device, computer equipment and a storage medium.
In a first aspect, an embodiment of the present disclosure provides a trajectory aggregation method, where the method includes:
determining representative images respectively corresponding to a plurality of pedestrian image cluster clusters, wherein pedestrians indicated by the plurality of pedestrian images in each pedestrian image cluster are the same person;
determining a representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images to the pedestrian feature ratio of the target pedestrian as a target representative image corresponding to the target pedestrian;
and combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergent track of the target pedestrian.
In this embodiment, the representative images corresponding to the plurality of determined pedestrian image cluster clusters are respectively based on, and the action track of the target pedestrian is determined based on the corresponding acquired information, so that the interference of the interference image on the action track generation process by the target pedestrian can be reduced, the accuracy of generating the action track of the target pedestrian is improved, the data processing amount is effectively reduced, and the data processing time is saved.
In an alternative embodiment, a plurality of pedestrian image clusters are obtained by:
acquiring a plurality of pedestrian images in a target time period and a target area;
and carrying out image clustering on each pedestrian in the multiple pedestrian images to obtain a pedestrian image cluster corresponding to each pedestrian in the multiple pedestrians.
In this optional embodiment, a plurality of pedestrian image cluster clusters corresponding to each pedestrian are obtained by performing image clustering, and preparation is made for subsequently determining a representative image of the representative pedestrian image cluster, so that the representative image and a target representative image corresponding to a target pedestrian are conveniently compared in the subsequent process, the comparison times are reduced, and the comparison accuracy is improved.
In an optional embodiment, the determining representative images corresponding to a plurality of pedestrian image cluster groups respectively includes:
determining the score of each pedestrian image in the pedestrian image cluster under each preset quality scoring dimension aiming at each acquired pedestrian image cluster;
determining an image quality score of each pedestrian image based on the weight of each preset quality scoring dimension and the score of each pedestrian image under each preset quality scoring dimension;
and determining a representative image corresponding to the pedestrian image cluster based on the image quality score of each pedestrian image in the pedestrian image cluster.
In the optional embodiment, the pedestrian image with the best image quality is selected as the representative image of the pedestrian image cluster, so that the interference of the pedestrian image with poor image quality on subsequent processing is reduced.
In an alternative embodiment, the quality scoring dimension comprises at least one of:
the definition dimension, the shooting posture dimension and the face feature matching dimension of the pedestrian image.
In this alternative embodiment, scoring is performed for multiple quality scoring dimensions, and the best representative image can be selected from multiple angles.
In an alternative embodiment, after determining the target representative image, the method further comprises:
adding the identification information of the target pedestrian, the image identification corresponding to the target representative image and the time-space information of the target representative image into an information table in a preset ratio in a database;
wherein the identification information of the target pedestrian includes at least one of:
the identity of the target pedestrian, and the type of the pedestrian.
In the optional embodiment, by establishing the information table in the ratio, information search and storage are performed, which facilitates the search and subsequent acquisition process of information.
In an alternative embodiment, after determining the target representative image, the method further comprises:
acquiring a track information mapping table corresponding to the target pedestrian in the information table in the ratio;
and updating the track information mapping table based on the convergent track of the target pedestrian.
In this optional embodiment, the track information mapping table is updated, so as to ensure the timeliness of the data in the track information mapping table.
In an optional embodiment, the updating the trajectory information mapping table based on the converged trajectory of the target pedestrian includes:
acquiring an information updating timestamp of the track information mapping table;
if the information updating time stamp is before the acquisition time corresponding to the target clustering image, updating the information updating time stamp to the acquisition time;
and adding the identification information of the target pedestrian, the image identification and the space-time information of each pedestrian image in the target pedestrian image cluster to the track information mapping table.
In an optional implementation manner, the merging the trajectory of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical trajectory of the target pedestrian to obtain the convergent trajectory of the target pedestrian includes:
and converging to obtain the converging track of the target pedestrian according to the spatio-temporal information recorded in the track information mapping table, wherein the spatio-temporal information recorded in the track information mapping table comprises spatio-temporal information of images of all pedestrians.
In the optional embodiment, the convergence trajectory of the target pedestrian is obtained based on the spatiotemporal information recorded in the trajectory information mapping table, and the convergence efficiency of the trajectory is improved.
In a second aspect, an embodiment of the present disclosure provides a trajectory convergence device, where the trajectory convergence device includes:
the pedestrian image clustering system comprises a first determining module, a second determining module and a judging module, wherein the first determining module is used for determining representative images corresponding to a plurality of pedestrian image clustering clusters respectively, and pedestrians indicated by a plurality of pedestrian images in each pedestrian image clustering cluster are the same person;
the second determination module is used for determining a representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images to the pedestrian feature ratio of the target pedestrian as a target representative image corresponding to the target pedestrian;
and the convergence module is used for combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergence track of the target pedestrian.
In an alternative embodiment, the apparatus further comprises: an acquisition module;
the acquisition module is configured to: acquiring a plurality of pedestrian images in a target time period and a target area; and carrying out image clustering on each pedestrian in the multiple pedestrian images to obtain a pedestrian image cluster corresponding to each pedestrian in the multiple pedestrians.
In an optional implementation manner, the first determining module is specifically configured to:
determining the score of each pedestrian image in the pedestrian image cluster under each preset quality scoring dimension aiming at each acquired pedestrian image cluster;
determining an image quality score of each pedestrian image based on the weight of each preset quality scoring dimension and the score of each pedestrian image under each preset quality scoring dimension;
and determining a representative image corresponding to the pedestrian image cluster based on the image quality score of each pedestrian image in the pedestrian image cluster.
In an alternative embodiment, the quality scoring dimension comprises at least one of:
the definition dimension, the shooting posture dimension and the face feature matching dimension of the pedestrian image.
In an alternative embodiment, the apparatus further comprises: adding a module;
the adding module is used for adding the identification information of the target pedestrian, the image identification corresponding to the target representative image and the space-time information of the target representative image into an information table in a preset ratio in a database;
wherein the identification information of the target pedestrian includes at least one of:
the identity of the target pedestrian, and the type of the pedestrian.
In an alternative embodiment, the apparatus further comprises: an update module;
the update module includes:
the acquisition unit is used for acquiring a track information mapping table corresponding to the target pedestrian in the information table in the ratio;
and the updating unit is used for updating the track information mapping table based on the convergent track of the target pedestrian.
In an optional implementation manner, the updating unit is specifically configured to:
acquiring an information updating timestamp of the track information mapping table;
if the information updating time stamp is before the acquisition time corresponding to the target clustering image, updating the information updating time stamp to the acquisition time;
and adding the identification information of the target pedestrian, the image identification and the space-time information of each pedestrian image in the target pedestrian image cluster to the track information mapping table.
In an optional implementation manner, the convergence module is specifically configured to:
and converging to obtain the converging track of the target pedestrian according to the spatio-temporal information recorded in the track information mapping table, wherein the spatio-temporal information recorded in the track information mapping table comprises spatio-temporal information of images of all pedestrians.
In a third aspect, an embodiment of the present disclosure further provides a computer device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the computer device is running, the machine-readable instructions when executed by the processor performing the steps of the first aspect described above, or any possible implementation of the first aspect.
In a fourth aspect, this disclosed embodiment also provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to perform the steps in the first aspect or any one of the possible implementation manners of the first aspect.
The track convergence method, device, computer equipment and storage medium provided by the embodiment of the disclosure comprise: determining representative images respectively corresponding to a plurality of pedestrian image cluster clusters, wherein pedestrians indicated by the plurality of pedestrian images in each pedestrian image cluster are the same person; determining a representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images to the pedestrian feature ratio of the target pedestrian as a target representative image corresponding to the target pedestrian; and combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergent track of the target pedestrian.
In the embodiment of the disclosure, the representative image of the pedestrian image cluster obtained by clustering is utilized, the representative image is compared with the target pedestrian, the target representative image corresponding to the target pedestrian is determined, and the convergence track of the target pedestrian is determined based on the indication information of the pedestrian image cluster to which the target representative image belongs and the history information of the target pedestrian. Therefore, in the image comparison process, the representative image of the pedestrian image cluster is adopted, so that the interference of the interference image to the comparison process is reduced, and the accuracy of the comparison result can be improved. And the representative images of the pedestrian image cluster are adopted in the comparison, so that the comparison of a large number of pedestrian images is not needed, the efficiency of the comparison process is improved, the generation efficiency of the convergence track is improved, the data processing amount is effectively reduced, and the data processing time is saved.
In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly described below, and the drawings herein incorporated in and forming a part of the specification illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the technical solutions of the present disclosure. It is appreciated that the following drawings depict only certain embodiments of the disclosure and are therefore not to be considered limiting of its scope, for those skilled in the art will be able to derive additional related drawings therefrom without the benefit of the inventive faculty.
Fig. 1 is a flowchart illustrating a trajectory aggregation method provided by an embodiment of the present disclosure;
FIG. 2 is a flow chart of another trajectory aggregation method provided by the embodiments of the present disclosure;
FIG. 3 is a flow chart illustrating updating a track information mapping table in an embodiment of the present disclosure;
fig. 4 is a schematic diagram illustrating a trajectory converging device provided by an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of another trajectory aggregation device provided by an embodiment of the present disclosure;
fig. 6 is a schematic diagram illustrating an update module in another track aggregation device provided in the embodiment of the present disclosure;
fig. 7 shows a schematic diagram of a computer device provided by an embodiment of the present disclosure.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. The components of the embodiments of the present disclosure, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure, presented in the figures, is not intended to limit the scope of the claimed disclosure, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the disclosure without making creative efforts, shall fall within the protection scope of the disclosure.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The term "and/or" herein merely describes an associative relationship, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, and may mean including any one or more elements selected from the group consisting of A, B and C.
Research shows that with the diversification and the generalization of information acquisition modes, the information of the user which can be acquired is exponentially increased, for example, in an actual scene, a pedestrian can be captured to an image by the same camera or different cameras for multiple times in the same place or a small area at a certain moment or a certain short time period, the data amount to be processed is large, the processing efficiency is low, and the operation load is large. In addition, in the course of generating the action trajectory for the target pedestrian, the trajectory information corresponding to the target pedestrian is often determined by using the entire image content captured for the target pedestrian, and since a certain number of interference images exist in the entire image content captured for the target pedestrian, for example, images of other persons are mistaken for the image content of the target pedestrian, there is a possibility that the process of generating the trajectory information corresponding to the target pedestrian is disturbed to some extent, and the generated trajectory information cannot accurately represent the action trajectory of the target pedestrian.
Based on the above research, the embodiments of the present disclosure provide a trajectory aggregation, including: a plurality of pedestrian image clustering clusters are obtained by carrying out image clustering on the collected pedestrian images; determining a representative image of each pedestrian image cluster; screening out a target representative image compared with the character characteristics of the target pedestrian from the acquired representative images; and determining the action track of the target pedestrian based on the acquisition time and the acquisition place of the screened target representative image.
In the embodiment of the disclosure, the representative image of the pedestrian image cluster obtained by clustering is utilized, the representative image is compared with the target pedestrian, the target representative image corresponding to the target pedestrian is determined, and the convergence track of the target pedestrian is determined based on the indication information of the pedestrian image cluster to which the target representative image belongs and the history information of the target pedestrian. Therefore, in the image comparison process, the representative image of the pedestrian image cluster is adopted, so that the interference of the interference image to the comparison process is reduced, and the accuracy of the comparison result can be improved. And the representative images of the pedestrian image cluster are adopted in the comparison, so that a large number of pedestrian images do not need to be compared, the efficiency of the comparison process is improved, and the generation efficiency of the convergence track is further improved.
To facilitate understanding of the present embodiment, first, a trajectory aggregation method disclosed in the embodiments of the present disclosure is described in detail, where an execution subject of the trajectory aggregation method provided in the embodiments of the present disclosure is generally a computer device with certain computing capability, and the computer device includes, for example: a terminal device, which may be a User Equipment (UE), a mobile device, a User terminal, a cellular phone, a cordless phone, a Personal Digital Assistant (PDA), a handheld device, a computing device, a vehicle mounted device, a wearable device, or a server or other processing device. In some possible implementations, the trajectory aggregation method may be implemented by a processor invoking computer readable instructions stored in a memory.
Referring to fig. 1, a flowchart of a trajectory aggregation method provided in the embodiment of the present disclosure is shown, where the aggregation method includes steps S101 to S103, where:
s101, representative images corresponding to a plurality of pedestrian image cluster clusters are determined, wherein pedestrians indicated by the pedestrian images in each pedestrian image cluster are the same person.
In this step, a plurality of pedestrian images may be obtained first, and a plurality of pedestrian image cluster clusters are obtained by clustering the plurality of pedestrian images.
The pedestrian images are acquired by shooting through a plurality of cameras arranged in a plurality of places, and can be stored in storage positions corresponding to the cameras or stored in a designated image database. Correspondingly, the pedestrian image can be directly obtained from the storage position corresponding to the arranged camera, and can also be obtained from an image database, wherein the image database is used for storing the shot picture by the arranged camera.
In the embodiment of the disclosure, after the acquired multiple pedestrian images are acquired, clustering processing may be performed on the multiple pedestrian images. Specifically, a plurality of pedestrian image cluster clusters obtained by performing image clustering on a plurality of collected pedestrian images include:
acquiring a plurality of pedestrian images in a target time period and a target area;
and carrying out image clustering on each pedestrian in the multiple pedestrian images to obtain a pedestrian image cluster corresponding to each pedestrian in the multiple pedestrians.
In an embodiment of the disclosure, the image clustering method may be to perform similarity clustering on a plurality of pedestrian images, that is, perform image clustering on the plurality of pedestrian images based on a preset image similarity threshold value to obtain a plurality of pedestrian image clustering clusters; and the image similarity between any two pedestrian images in each pedestrian image cluster is greater than a preset image similarity threshold value.
For example, a corresponding target time period and target area may be determined for the target demand, for example: and pedestrian images in city A of 11 months A in 2020, thereby reducing the number of processes for corresponding pedestrian images and improving the processing efficiency.
After acquiring a plurality of pedestrian images in a target area acquired in a target time period and determining a preset image similarity threshold, image clustering can be performed on the plurality of pedestrian images according to the image similarity threshold. Specifically, based on an image clustering mechanism, after a corresponding image similarity threshold is specified, a plurality of pedestrian images meeting the image similarity threshold can be clustered into a group to obtain a plurality of pedestrian image clustering clusters, wherein the image similarity between any two pedestrian images in each pedestrian image clustering cluster is greater than a preset image similarity threshold.
For example, if it is determined that the image similarity threshold is 90%, based on one pedestrian image, the pedestrian images with the similarity of 90% between the remaining pedestrian images and the pedestrian image are grouped into one group with the image, so that a plurality of pedestrian image cluster groups can be obtained, wherein the image similarity between any two pedestrian images in each pedestrian image cluster group is greater than the preset image similarity threshold, namely 90%.
Therefore, a plurality of pedestrian image cluster clusters corresponding to each pedestrian are obtained by carrying out image clustering on the pedestrian images, preparation is made for subsequently determining the representative images of the representative pedestrian image cluster clusters, the representative images and the target representative images corresponding to the target pedestrians are conveniently compared in the subsequent process, the comparison times are reduced, and the comparison accuracy is improved.
After the pedestrian image cluster corresponding to each pedestrian in the pedestrians is determined, the representative image corresponding to each pedestrian image cluster can be selected from each pedestrian image cluster. Because each pedestrian image cluster comprises a plurality of pedestrian images for the same pedestrian, in order to carry out follow-up by utilizing the pedestrian image most representing the pedestrian, and reduce the follow-up processing amount, one representative image representing the pedestrian image cluster can be selected.
In an embodiment of the present disclosure, a method for determining representative images corresponding to a plurality of pedestrian image cluster clusters respectively includes:
determining the score of each pedestrian image in the pedestrian image cluster under each preset quality scoring dimension aiming at each acquired pedestrian image cluster;
determining an image quality score of each pedestrian image based on the weight of each preset quality scoring dimension and the score of each pedestrian image under each preset quality scoring dimension;
and determining a representative image corresponding to the pedestrian image cluster based on the image quality score of each pedestrian image in the pedestrian image cluster.
In this step, a representative image corresponding to each pedestrian image cluster can be determined according to the image quality score corresponding to each pedestrian image in each pedestrian image cluster, and therefore, the pedestrian image with the best image quality can be used for subsequent processing. Because the pedestrian image with the best image quality is used for subsequent processing, the interference of the pedestrian image with poor image quality on the processing process is reduced, and the accuracy and the efficiency of the subsequent processing are improved. In addition, the processing process only aims at the selected pedestrian image with the best image quality, so that the processing amount of the subsequent processing process can be reduced, and the calculation load can be reduced.
Specifically, the image quality score of each pedestrian image in the pedestrian image cluster can be determined through the following steps:
determining the score of each pedestrian image in the pedestrian image cluster under each preset quality score dimension;
and determining the image quality score of each pedestrian image based on the weight proportion of each preset quality score dimension and the score of each pedestrian image under each preset quality score dimension.
Wherein the quality scoring dimension comprises at least one of:
the definition dimension, the shooting posture dimension and the face feature matching dimension of the pedestrian image.
For example, for the case that the quality score dimension includes a definition dimension of a pedestrian image, the method for determining the score of each pedestrian image in the cluster of pedestrian images in the definition dimension of the pedestrian image includes:
and acquiring a pixel value corresponding to each pedestrian image, wherein the higher the pixel value is, the higher the score of the definition dimension of the pedestrian image corresponding to the pedestrian image is.
For the condition that the quality scoring dimension comprises a shooting attitude dimension, the method for determining the score of each pedestrian image in the pedestrian image cluster under the shooting attitude dimension comprises the following steps:
and acquiring a shooting angle corresponding to each pedestrian image, wherein the more the shooting angle is close to the numerical values of 0 degree, 90 degrees or 180 degrees and the like, the higher the score of the corresponding shooting attitude dimension of the pedestrian image is.
The method for determining the score of each pedestrian image in the pedestrian image cluster under the face feature matching dimension comprises the following steps of:
and inputting each pedestrian image into a trained face feature recognition model, and acquiring the probability numerical value that the pedestrian image of the face feature recognition model data comprises a face. The higher the probability value of the pedestrian image including the face is, the higher the score of the corresponding face feature matching dimension of the pedestrian image is.
After the score of each pedestrian image in the pedestrian image cluster under each preset quality scoring dimension is determined, the weight proportion of each preset quality scoring dimension can be determined based on the influence weight of each preset quality scoring dimension on the image quality.
Illustratively, the shooting attitude has less influence on the image quality relative to the definition dimension of the pedestrian image, and therefore the weight ratio of the shooting attitude relative to the definition dimension of the pedestrian image is lower.
After the score of each pedestrian image in the pedestrian image cluster under each preset quality scoring dimension and the weight proportion of each preset quality scoring dimension are determined, the image quality score of each pedestrian image can be determined.
Specifically, the calculation formula of the image quality score is as follows:
the definition dimension score of the pedestrian image, the definition dimension weight of the pedestrian image, the shooting posture dimension score, the shooting posture dimension weight, the face feature matching dimension score, the face feature matching dimension weight and the image quality score.
Illustratively, for a pedestrian image, the score of the pedestrian image in the definition dimension is 80, the score of the pedestrian image in the shooting posture dimension is 90, and the score of the human face feature matching dimension is 80, and correspondingly, the weight ratio of the definition dimension of the pedestrian image is 40%, the weight ratio of the shooting posture dimension is 20%, and the weight ratio of the human face feature matching dimension is 40%. Therefore, the image quality score of the pedestrian image is:
80*40％+90*20％+80*40％＝82；
that is, the final image quality score of the pedestrian image was 82 points.
Therefore, the pedestrian image with the best image quality can be selected as the representative image of the pedestrian image cluster, and the interference of the pedestrian image with poor image quality on subsequent processing is reduced. Meanwhile, scoring is carried out according to the multi-quality scoring dimensionality, and the best representative image can be selected from multiple angles.
S102, determining the representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images to the pedestrian feature ratio of the target pedestrian as the target representative image corresponding to the target pedestrian.
In this step, after a plurality of representative images are acquired, a target representative image that is compared with the character characteristics of the target pedestrian may be screened out by the comparison system.
The method for determining the target representative image corresponding to the target pedestrian can comprise the following steps:
comparing the character features of each representative image with the character features of the predetermined target pedestrian;
and determining the representative image in the comparison of the corresponding characteristic of the representative image and the character characteristic of the target pedestrian as the target representative image.
In this step, the extracted character features, such as the features of five sense organs and the height feature, of each representative image may be compared with the character features of the target pedestrian pre-stored in the comparison system based on the preset target feature category.
When the character features of the representative image are matched with the character features of the target pedestrian, the corresponding representative image compared with the character features of the target pedestrian can be determined as the target representative image.
S103, combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergent track of the target pedestrian.
In this step, after the representative image compared with the human features of the target pedestrian is determined to be the target representative image, the latest trajectory of the target pedestrian is generated by synthesis based on the current trajectory of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs and the history trajectory stored in the history.
Specifically, because the pedestrian image has information such as corresponding time, place, scene, and collection camera number during collection, the current track of the target pedestrian can be determined based on the collection time and collection place indicated by the pedestrian image cluster to which the target representative image belongs, and the instant empty information, and the latest complete track of the target pedestrian can be obtained by combining the historical track.
Specifically, each pedestrian image cluster corresponding to each target representative image has corresponding shooting time and corresponding shooting equipment, and each shooting equipment is also corresponding to a fixed installation position, so that after a target representative image is determined, the occurrence time and the occurrence place of a target pedestrian corresponding to the pedestrian image cluster corresponding to the target representative image can be obtained, the current track of the target pedestrian can be determined, and after the historical track of the target pedestrian is obtained, the latest complete track of the target pedestrian can be connected and converged.
Referring to fig. 2, fig. 2 is a flowchart of another trajectory aggregation method according to an embodiment of the present disclosure. As shown in fig. 2, the convergence method includes:
s201, representative images corresponding to a plurality of pedestrian image cluster clusters are determined, wherein pedestrians indicated by the plurality of pedestrian images in each pedestrian image cluster are the same person.
S202, determining the representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images and the pedestrian feature ratio of the target pedestrian as the target representative image corresponding to the target pedestrian.
The descriptions of step S201 to step S202 may refer to the descriptions of step S101 to step S102, and the same technical effects may be achieved, which are not described herein again.
S203, adding the identification information of the target pedestrian, the image identification corresponding to the target representative image and the space-time information of the target representative image into an information table in a preset ratio in a database.
In this step, after the target representative image compared with the character features of the target pedestrian is screened out, in order to facilitate the subsequent information query and acquisition process, the identification information of the target pedestrian, the image identification corresponding to the target representative image, and the time-space information of the target representative image may be added to an information table in a preset ratio in the database.
Wherein the identification information of the target pedestrian includes at least one of: the identity of the target pedestrian, and the type of the pedestrian. Specifically, the identification of the target pedestrian may be identification information of the pedestrian, such as an identification number, a residence code, and the like, and the pedestrian type may refer to a social role, a level, and the like of the pedestrian, and represents data of characteristics of the pedestrian identification type. The image identifier corresponding to the target representative image may be an identity identifier of the image, which is convenient for subsequent viewing, for example, when querying data information of the target pedestrian, the image identifier is called.
In the embodiment of the present disclosure, the information is added to the information table in the ratio, and information is searched and stored, which is convenient for the search and subsequent acquisition process of the information.
And S204, acquiring a track information mapping table corresponding to the target pedestrian in the information table in the ratio.
In this step, in order to facilitate the subsequent information search process for the target pedestrian, a trajectory information mapping table for each target pedestrian may be generated based on the information of the target pedestrian in the information table in the ratio. Thereby, the trajectory for each target pedestrian can be generated based on the trajectory information mapping table for each target pedestrian.
Specifically, when history data exists in the trajectory information mapping table of the target pedestrian, the target representative image corresponding to the target pedestrian can be used as a 'cover' file of the trajectory information mapping table of the target pedestrian, namely, the figure image and the feature information which can represent the target pedestrian most, a plurality of pedestrian images in the pedestrian image cluster corresponding to the target representative image, and the corresponding acquisition time and acquisition place of the pedestrian images are packaged into a file and added into the trajectory information mapping table of the target pedestrian, so that the subsequent source searching and tracing process is facilitated. The track information mapping table comprises a plurality of cover files and corresponding packaging files which are arranged according to a time sequence.
When the track information mapping table of the target pedestrian does not have historical data, the target representative image corresponding to the target pedestrian can be used as a 'cover' file of the track information mapping table of the target pedestrian, and the multiple pedestrian images in the pedestrian image cluster corresponding to the target representative image, the corresponding acquisition time and the corresponding acquisition place of the multiple pedestrian images are packaged into a file and stored in the track information mapping table of the target pedestrian.
S205, updating the track information mapping table based on the convergence track of the target pedestrian.
In this step, in order to ensure the timeliness of the trajectory information mapping table, the trajectory information mapping table may be updated based on the convergent trajectory of the target pedestrian.
Specifically, as shown in fig. 3, fig. 3 is a flowchart for updating the track information mapping table in the embodiment of the present disclosure. As shown in fig. 3, the updating the trajectory information mapping table based on the convergent trajectory of the target pedestrian includes:
s301, acquiring an information updating time stamp of the track information mapping table;
s302, if the information updating time stamp is before the acquisition time corresponding to the target clustering image, updating the information updating time stamp to the acquisition time;
s303, adding the identification information of the target pedestrian, the image identification and the spatiotemporal information of the images of the pedestrians in the target pedestrian image cluster to the track information mapping table.
In steps S301 to S303, the data update time stamp corresponding to the last added data or updated data of the trajectory information mapping table of the target pedestrian may be acquired. And if the update time stamp of the track information mapping table is before the acquisition time corresponding to the target pedestrian image, updating the data update time stamp of the track information mapping table of the target pedestrian to be the acquisition time corresponding to the target pedestrian image.
In this step, if the update time stamp of the trajectory information mapping table is before the acquisition time corresponding to the target pedestrian image, it is verified that the target pedestrian image is not added to the data of the trajectory information mapping table of the target pedestrian, and after the target pedestrian image is added to the monitoring database, the data update time stamp of the trajectory information mapping table of the target pedestrian can also be updated to the acquisition time corresponding to the target pedestrian image, so as to represent to update or add the target pedestrian image. And adding the identification information of the target pedestrian, the image identification and the time-space information of each pedestrian image in the target pedestrian image cluster to the track information mapping table so as to ensure the timeliness of the information in the track information mapping table.
And S206, converging to obtain the converging track of the target pedestrian according to the spatiotemporal information recorded in the track information mapping table, wherein the spatiotemporal information recorded in the track information mapping table comprises spatiotemporal information of images of the pedestrians.
In this step, the convergent track of the target pedestrian can be obtained by convergence according to the spatiotemporal information corresponding to the target pedestrian image in the track information mapping table of the target pedestrian, wherein the spatiotemporal information mainly comprises the acquisition time and the acquisition place.
Specifically, when the target pedestrian track is converged, the acquisition time and the acquisition place of the target pedestrian image can be called from the track information mapping table of the target pedestrian, and the track of the target pedestrian is converged by link connection based on the historical information of the target pedestrian.
In another embodiment of the present disclosure, when there is no history data in the trajectory information mapping table of the target pedestrian, the action trajectory of the target pedestrian is generated based on a plurality of collection times and collection places corresponding to the target pedestrian. When historical data exists in the track information mapping table of the target pedestrian, the track of the target pedestrian is updated based on at least one piece of space-time information corresponding to the target pedestrian and the historical space-time information.
In the embodiment of the disclosure, the representative image of the pedestrian image cluster obtained by clustering is utilized, the representative image is compared with the target pedestrian, the target representative image corresponding to the target pedestrian is determined, and the convergence track of the target pedestrian is determined based on the indication information of the pedestrian image cluster to which the target representative image belongs and the history information of the target pedestrian. Therefore, in the image comparison process, the representative image of the pedestrian image cluster is adopted, so that the interference of the interference image to the comparison process is reduced, and the accuracy of the comparison result can be improved. And the representative images of the pedestrian image cluster are adopted in the comparison, so that a large number of pedestrian images do not need to be compared, the efficiency of the comparison process is improved, and the generation efficiency of the convergence track is further improved.
It will be understood by those skilled in the art that in the method of the present invention, the order of writing the steps does not imply a strict order of execution and any limitations on the implementation, and the specific order of execution of the steps should be determined by their function and possible inherent logic.
Based on the same inventive concept, the embodiment of the present disclosure further provides a trajectory aggregation device corresponding to the trajectory aggregation method, and as the principle of solving the problem of the device in the embodiment of the present disclosure is similar to the trajectory aggregation method in the embodiment of the present disclosure, the implementation of the device may refer to the implementation of the method, and repeated details are not repeated.
Referring to fig. 4, 5, and 6, fig. 4 is a schematic diagram of a trajectory aggregation device according to an embodiment of the disclosure, fig. 5 is a schematic diagram of another trajectory aggregation device according to an embodiment of the disclosure, and fig. 6 is a schematic diagram of an update module in another trajectory aggregation device according to an embodiment of the disclosure. As shown in fig. 4, the apparatus includes: a first determining module 410, a second determining module 420, and a converging module 430; wherein:
the first determining module 410 is configured to determine representative images corresponding to a plurality of pedestrian image cluster clusters, where pedestrians indicated by a plurality of pedestrian images in each pedestrian image cluster are the same person;
a second determining module 420, configured to determine a representative image in a pedestrian feature ratio of a corresponding pedestrian feature to a target pedestrian in the plurality of representative images as a target representative image corresponding to the target pedestrian;
the converging module 430 is configured to merge the current trajectory of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical trajectory of the target pedestrian to obtain a converging trajectory of the target pedestrian.
In an alternative embodiment, as shown in fig. 5, the apparatus further comprises: an acquisition module 440;
the obtaining module 440 is configured to: acquiring a plurality of pedestrian images in a target time period and a target area; and carrying out image clustering on each pedestrian in the multiple pedestrian images to obtain a pedestrian image cluster corresponding to each pedestrian in the multiple pedestrians.
In an optional implementation manner, the first determining module 410 is specifically configured to:
determining the score of each pedestrian image in the pedestrian image cluster under each preset quality scoring dimension aiming at each acquired pedestrian image cluster;
determining an image quality score of each pedestrian image based on the weight of each preset quality scoring dimension and the score of each pedestrian image under each preset quality scoring dimension;
and determining a representative image corresponding to the pedestrian image cluster based on the image quality score of each pedestrian image in the pedestrian image cluster.
In an alternative embodiment, the quality scoring dimension comprises at least one of:
the definition dimension, the shooting posture dimension and the face feature matching dimension of the pedestrian image.
In an alternative embodiment, as shown in fig. 5, the apparatus further comprises: an add module 450;
the adding module 450 is configured to add the identification information of the target pedestrian, the image identification corresponding to the target representative image, and the spatiotemporal information of the target representative image to an information table in a ratio preset in a database;
wherein the identification information of the target pedestrian includes at least one of:
the identity of the target pedestrian, and the type of the pedestrian.
In an alternative embodiment, as shown in fig. 5, the apparatus further comprises: an update module 460;
as shown in fig. 6, the update module 460 includes:
an obtaining unit 461, configured to obtain a track information mapping table corresponding to the target pedestrian in the information table in the ratio;
an updating unit 462, configured to update the trajectory information mapping table based on the convergent trajectory of the target pedestrian.
In an optional implementation manner, the updating unit 462 is specifically configured to:
acquiring an information updating timestamp of the track information mapping table;
if the information updating time stamp is before the acquisition time corresponding to the target clustering image, updating the information updating time stamp to the acquisition time;
and adding the identification information of the target pedestrian, the image identification and the space-time information of each pedestrian image in the target pedestrian image cluster to the track information mapping table.
In an optional implementation manner, the aggregation module 430 is specifically configured to:
and converging to obtain the converging track of the target pedestrian according to the spatio-temporal information recorded in the track information mapping table, wherein the spatio-temporal information recorded in the track information mapping table comprises spatio-temporal information of images of all pedestrians.
The description of the processing flow of each module in the device and the interaction flow between the modules may refer to the related description in the above method embodiments, and will not be described in detail here.
In the embodiment of the disclosure, the representative image of the pedestrian image cluster obtained by clustering is utilized, the representative image is compared with the target pedestrian, the target representative image corresponding to the target pedestrian is determined, and the convergence track of the target pedestrian is determined based on the indication information of the pedestrian image cluster to which the target representative image belongs and the history information of the target pedestrian. Therefore, in the image comparison process, the representative image of the pedestrian image cluster is adopted, so that the interference of the interference image to the comparison process is reduced, and the accuracy of the comparison result can be improved. And the representative images of the pedestrian image cluster are adopted in the comparison, so that a large number of pedestrian images do not need to be compared, the efficiency of the comparison process is improved, and the generation efficiency of the convergence track is further improved.
Corresponding to the trajectory aggregation method in fig. 1, an embodiment of the present disclosure further provides a computer device, and as shown in fig. 7, a schematic structural diagram of the computer device provided in the embodiment of the present disclosure includes:
a processor 701, a memory 702, and a bus 703; the memory 702 is used for storing execution instructions and includes a memory 7021 and an external memory 7022; the memory 7021 is also referred to as an internal memory, and is configured to temporarily store operation data in the processor 701 and data exchanged with an external memory 7022 such as a hard disk, the processor 701 exchanges data with the external memory 7022 through the memory 7021, and when the computer apparatus operates, the processor 701 and the memory 702 communicate with each other through a bus 703, so that the processor 701 executes the following instructions:
determining representative images respectively corresponding to a plurality of pedestrian image cluster clusters, wherein pedestrians indicated by the plurality of pedestrian images in each pedestrian image cluster are the same person;
determining a representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images to the pedestrian feature ratio of the target pedestrian as a target representative image corresponding to the target pedestrian;
and combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergent track of the target pedestrian.
The embodiments of the present disclosure also provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the trajectory aggregation method described in the above method embodiments. The storage medium may be a volatile or non-volatile computer-readable storage medium.
The embodiments of the present disclosure also provide a computer program product, where the computer program product carries a program code, and instructions included in the program code may be used to execute the steps of the trajectory aggregation method in the foregoing method embodiments, which may be referred to specifically in the foregoing method embodiments, and are not described herein again.
The computer program product may be implemented by hardware, software or a combination thereof. In an alternative embodiment, the computer program product is embodied in a computer storage medium, and in another alternative embodiment, the computer program product is embodied in a Software product, such as a Software Development Kit (SDK), or the like.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are merely specific embodiments of the present disclosure, which are used for illustrating the technical solutions of the present disclosure and not for limiting the same, and the scope of the present disclosure is not limited thereto, and although the present disclosure is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive of the technical solutions described in the foregoing embodiments or equivalent technical features thereof within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present disclosure, and should be construed as being included therein. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (11)
1. A method for trajectory convergence, the method comprising:
determining representative images respectively corresponding to a plurality of pedestrian image cluster clusters, wherein pedestrians indicated by the plurality of pedestrian images in each pedestrian image cluster are the same person;
determining a representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images to the pedestrian feature ratio of the target pedestrian as a target representative image corresponding to the target pedestrian;
and combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergent track of the target pedestrian.
2. The trajectory aggregation method of claim 1, wherein the plurality of pedestrian image clusters are obtained by:
acquiring a plurality of pedestrian images in a target time period and a target area;
and carrying out image clustering on each pedestrian in the multiple pedestrian images to obtain a pedestrian image cluster corresponding to each pedestrian in the multiple pedestrians.
3. The trajectory aggregation method according to claim 1, wherein the determining representative images respectively corresponding to the pedestrian image cluster comprises:
determining the score of each pedestrian image in the pedestrian image cluster under each preset quality scoring dimension aiming at each acquired pedestrian image cluster;
determining an image quality score of each pedestrian image based on the weight of each preset quality scoring dimension and the score of each pedestrian image under each preset quality scoring dimension;
and determining a representative image corresponding to the pedestrian image cluster based on the image quality score of each pedestrian image in the pedestrian image cluster.
4. The trajectory aggregation method of claim 3, wherein the quality scoring dimension comprises at least one of:
the definition dimension, the shooting posture dimension and the face feature matching dimension of the pedestrian image;
5. the trajectory aggregation method of claim 1, wherein after determining the target representative image, the method further comprises:
adding the identification information of the target pedestrian, the image identification corresponding to the target representative image and the time-space information of the target representative image into an information table in a preset ratio in a database;
wherein the identification information of the target pedestrian includes at least one of:
the identity of the target pedestrian, and the type of the pedestrian.
6. The trajectory aggregation method of claim 5, wherein after determining the target representative image, the method further comprises:
acquiring a track information mapping table corresponding to the target pedestrian in the information table in the ratio;
and updating the track information mapping table based on the convergent track of the target pedestrian.
7. The trajectory aggregation method according to claim 6, wherein the updating the trajectory information mapping table based on the aggregated trajectory of the target pedestrian comprises:
acquiring an information updating timestamp of the track information mapping table;
if the information updating time stamp is before the acquisition time corresponding to the target clustering image, updating the information updating time stamp to the acquisition time;
and adding the identification information of the target pedestrian, the image identification and the space-time information of each pedestrian image in the target pedestrian image cluster to the track information mapping table.
8. The trajectory aggregation method according to claim 6 or 7, wherein the merging the trajectory of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical trajectory of the target pedestrian to obtain the aggregated trajectory of the target pedestrian comprises:
and converging to obtain the converging track of the target pedestrian according to the spatio-temporal information recorded in the track information mapping table, wherein the spatio-temporal information recorded in the track information mapping table comprises spatio-temporal information of images of all pedestrians.
9. A trajectory converging device, characterized in that said device comprises:
the pedestrian image clustering system comprises a first determining module, a second determining module and a judging module, wherein the first determining module is used for determining representative images corresponding to a plurality of pedestrian image clustering clusters respectively, and pedestrians indicated by a plurality of pedestrian images in each pedestrian image clustering cluster are the same person;
the second determination module is used for determining a representative image in the pedestrian feature ratio of the corresponding pedestrian features in the representative images to the pedestrian feature ratio of the target pedestrian as a target representative image corresponding to the target pedestrian;
and the convergence module is used for combining the current track of the target pedestrian indicated by the pedestrian image cluster to which the target representative image belongs with the historical track of the target pedestrian to obtain the convergence track of the target pedestrian.
10. A computer device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when a computer device is running, the machine-readable instructions when executed by the processor performing the steps of the trajectory aggregation method according to any one of claims 1 to 8.
11. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the trajectory aggregation method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110324686.XA CN112966652A (en) | 2021-03-26 | 2021-03-26 | Trajectory convergence method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110324686.XA CN112966652A (en) | 2021-03-26 | 2021-03-26 | Trajectory convergence method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112966652A true CN112966652A (en) | 2021-06-15 |
Family
ID=76278530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110324686.XA Pending CN112966652A (en) | 2021-03-26 | 2021-03-26 | Trajectory convergence method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112966652A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113343060A (en) * | 2021-06-23 | 2021-09-03 | 北京市商汤科技开发有限公司 | Object detection method and device, electronic equipment and storage medium |
| CN113807223A (en) * | 2021-09-07 | 2021-12-17 | 南京中兴力维软件有限公司 | Face cluster subclass merging method, device and equipment |
| KR20230007141A (en) * | 2021-07-05 | 2023-01-12 | 서울과학기술대학교 산학협력단 | Method and apparatus of the same person identification among video sequences from multiple cameras, recording medium and device for performing the same |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170286781A1 (en) * | 2016-04-05 | 2017-10-05 | Omni Al, Inc. | Trajectory cluster model for learning trajectory patterns in videos data |
| CN112132103A (en) * | 2020-09-30 | 2020-12-25 | 新华智云科技有限公司 | Video face detection and recognition method and system |
| CN112232148A (en) * | 2020-09-28 | 2021-01-15 | 浙江大华技术股份有限公司 | Image clustering method, target track tracking method, electronic device and storage medium |
-
2021
- 2021-03-26 CN CN202110324686.XA patent/CN112966652A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170286781A1 (en) * | 2016-04-05 | 2017-10-05 | Omni Al, Inc. | Trajectory cluster model for learning trajectory patterns in videos data |
| CN112232148A (en) * | 2020-09-28 | 2021-01-15 | 浙江大华技术股份有限公司 | Image clustering method, target track tracking method, electronic device and storage medium |
| CN112132103A (en) * | 2020-09-30 | 2020-12-25 | 新华智云科技有限公司 | Video face detection and recognition method and system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113343060A (en) * | 2021-06-23 | 2021-09-03 | 北京市商汤科技开发有限公司 | Object detection method and device, electronic equipment and storage medium |
| KR20230007141A (en) * | 2021-07-05 | 2023-01-12 | 서울과학기술대학교 산학협력단 | Method and apparatus of the same person identification among video sequences from multiple cameras, recording medium and device for performing the same |
| KR102594694B1 (en) * | 2021-07-05 | 2023-10-26 | 서울과학기술대학교 산학협력단 | Method and apparatus of the same person identification among video sequences from multiple cameras, recording medium and device for performing the same |
| CN113807223A (en) * | 2021-09-07 | 2021-12-17 | 南京中兴力维软件有限公司 | Face cluster subclass merging method, device and equipment |
| CN113807223B (en) * | 2021-09-07 | 2024-04-09 | 南京中兴力维软件有限公司 | Face cluster subclass merging method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10755086B2 (en) | Picture ranking method, and terminal | |
| CN112966652A (en) | Trajectory convergence method and device, computer equipment and storage medium | |
| ES2946807T3 (en) | Media Drive Recovery and Related Processes | |
| CN110956547B (en) | Method and system for identifying fraudulent party in real time based on search engine | |
| US20210027085A1 (en) | Universal object recognition | |
| CN109634698B (en) | Menu display method and device, computer equipment and storage medium | |
| CN106933867B (en) | Image query method and device | |
| JP2021518944A (en) | License plate recognition method and its system | |
| JP5963609B2 (en) | Image processing apparatus and image processing method | |
| CN111783882B (en) | Key point detection method and device, electronic equipment and storage medium | |
| CN115862088A (en) | Identity recognition method and device | |
| CN112818149A (en) | Face clustering method and device based on space-time trajectory data and storage medium | |
| WO2023142551A1 (en) | Model training and image recognition methods and apparatuses, device, storage medium and computer program product | |
| CN109902681B (en) | User group relation determining method, device, equipment and storage medium | |
| US20170214756A1 (en) | Dynamic rule allocation for visitor identification | |
| CN110689046A (en) | Image recognition method, image recognition device, computer device, and storage medium | |
| CN110751027A (en) | Pedestrian re-identification method based on deep multi-instance learning | |
| CN110909817B (en) | Distributed clustering method and system, processor, electronic device and storage medium | |
| CN112329660A (en) | Scene recognition method and device, intelligent equipment and storage medium | |
| CN109447112B (en) | Portrait clustering method, electronic equipment and storage medium | |
| CN112925899B (en) | Ordering model establishment method, case clue recommendation method, device and medium | |
| KR20220027000A (en) | Method and device for extracting spatial relationship of geographic location points | |
| CN115082999A (en) | Group photo image person analysis method and device, computer equipment and storage medium | |
| CN112333182B (en) | File processing method, device, server and storage medium | |
| CN112232890B (en) | Data processing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210615 |
|
| RJ01 | Rejection of invention patent application after publication |