An object tracking method and apparatus, and an electronic device

Abstract

The application discloses an object tracking method and device and electronic equipment, and the method comprises the following steps: acquiring picture sets corresponding to each monitoring video, wherein each picture set comprises multiple pictures; performing feature category identification on objects in each picture in each picture set, and screening pictures in which objects are identified and only one object is included according to the identification result; determining a second feature value of a feature category according to first feature values corresponding to different pictures in each screened picture set and belonging to the feature category; for the same feature category, determining a feature similarity value corresponding to the feature category according to a similarity value of the second feature values corresponding to any two different picture sets; and determining a probability of each monitoring video in the tracking area being the same object based on a preset weight corresponding to each feature category and the feature similarity value. In this way, the problem that the traditional personnel tracking technology is affected by various factors and it is difficult to guarantee the accuracy and robustness of tracking is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of object tracking, and in particular to an object tracking method and device and electronic equipment. BACKGROUND

[0002] At present, personnel tracking technology has a wide range of applications in the fields of security monitoring, intelligent transportation, etc., but the traditional personnel tracking technology is affected by various factors such as light, angle of view, etc., and it is difficult to guarantee the accuracy and robustness of tracking. SUMMARY

[0003] The purpose of the present application is to provide an object tracking method and device and electronic equipment. To solve the problem that the traditional personnel tracking technology is affected by various factors, and it is difficult to guarantee the accuracy and robustness of tracking.

[0004] In a first aspect, an object tracking method is provided, which comprises:

[0005] From the monitoring video of each camera in the tracking area, a picture set comprising multiple pictures corresponding to each monitoring video is obtained;

[0006] According to the preset feature categories corresponding to different parts of the object, the object in each picture in each picture set is identified, and the pictures including only one object are filtered out according to the identification result;

[0007] For each picture filtered out, the first feature value of each feature category of the identified object is determined;

[0008] According to the first feature values of different pictures in the filtered picture set belonging to the same feature category, the second feature value of the feature category is determined;

[0009] For the same feature category, the feature similarity value corresponding to the feature category is determined according to the similarity value of the second feature value corresponding to any two different picture sets;

[0010] Based on the preset weight and feature similarity value corresponding to each feature category, the probability of the same object in each monitoring video in the tracking area is determined.

[0011] In some possible embodiments, the feature categories corresponding to the parts include: head tilt feature, height feature, moving speed feature, arm swing amplitude feature, aspect ratio feature and step length feature.

[0012] In some possible embodiments, the first feature value of each feature category of the identified object is determined for each picture filtered out, comprising:

[0013] For each picture in the selected picture set, when the feature category is the head tilt feature, the first feature value is the tilt angle between the head center point of the object and the center point of the object;

[0014] When the feature category is the height feature, the first feature value is the ratio of the first distance to the height of the object, wherein the first distance is the distance between the head center point of the object and the center point of the object;

[0015] When the feature category is the moving speed feature, the first feature value is the ratio of the second distance to the frame extraction time, wherein the second distance is the distance of the center coordinate movement of the object in the adjacent two pictures, and the frame extraction time is the frequency time of extracting pictures from the monitoring video;

[0016] When the feature category is the arm swing amplitude feature, the first feature value is the average value of the swing amplitude of each hand of the object, wherein the hand swing amplitude is determined by the vertical distance between the center point of the hand and the center of the object;

[0017] When the feature category is the aspect ratio feature, the first feature value is the ratio of the width of the rectangular frame of the object to the height of the rectangular frame, wherein the rectangular frame is a frame body selected according to the feature category corresponding to each part of the object in each picture in each picture set according to the preset feature category corresponding to each part of the object;

[0018] When the feature category is the step feature, the first feature value is the average value of the swing amplitude of each foot of the object, wherein the foot swing amplitude is determined by the vertical distance between the center point of the foot and the center of the object.

[0019] In some possible embodiments, the second feature value of the feature category is determined according to the first feature values corresponding to different pictures belonging to the same feature category in each selected picture set, comprising:

[0020] The first average value is obtained by averaging all the first feature values corresponding to different pictures belonging to the same feature category in each picture set, and the second feature value is the first average value.

[0021] In some possible embodiments, the similarity value of the second feature values corresponding between any two different picture sets is determined by the following way:

[0022] The first difference value of the second feature value of the first picture set and the second feature value of the second picture set is determined;

[0023] The second difference value of the preset value and the first difference value is determined;

[0024] determining a ratio of the second difference value and a second feature value of the first picture set as a similarity value of a corresponding second feature value between the first picture set and the second picture set;

[0025] The second picture set is a previous picture set of the first picture set.

[0026] In some possible embodiments, the determining, for a same feature category, a feature similarity value corresponding to the feature category according to the similarity values of the corresponding second feature values between any two different picture sets, comprises:

[0027] The similarity values of the corresponding second feature values between any two different picture sets for a same feature category are determined.

[0028] The average of the similarity values of all the second feature values is determined as the feature similarity value corresponding to the feature category.

[0029] In some possible embodiments, the determining, based on the preset weight corresponding to each feature category and the feature similarity value, a probability of being the same object in each monitoring video of the tracking area, comprises:

[0030] The first product of the weight corresponding to each feature category and the feature similarity value corresponding to the feature category is determined.

[0031] The sum of the first products corresponding to all the feature categories is determined.

[0032] The sum is the probability of being the same object in each monitoring video of the tracking area.

[0033] In a second aspect, an embodiment of the present application provides an object tracking device, which comprises:

[0034] A picture set acquisition module is configured to acquire, from each monitoring video of each camera in a tracking area, a picture set comprising multiple pictures corresponding to each monitoring video;

[0035] A screening module is configured to perform feature category identification on objects in each picture in each picture set according to a preset feature category corresponding to different parts of an object, and screen out pictures in which an object is identified and only one object is included according to a result of the identification.

[0036] A first feature value determination module is configured to determine, for each screened picture, a first feature value of each feature category of an identified object.

[0037] A second feature value determination module is configured to determine, according to the first feature values corresponding to different pictures in each screened picture set and belonging to a same feature category, a second feature value of the feature category.

[0038] determine a feature similarity value corresponding to each feature category according to a similarity value of the second feature values corresponding to any two different picture sets;

[0039] determine a probability of the same object in each monitoring video of the tracking area based on the preset weight corresponding to each feature category and the feature similarity value.

[0040] In a third aspect, an embodiment of the present application provides an electronic device, including at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the object tracking method in the first aspect.

[0041] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program, and the computer program is used to enable a computer to perform the object tracking method in the first aspect.

[0042] In the embodiments of the present application, in order to solve the problem that the traditional personnel tracking technology is affected by multiple factors and it is difficult to guarantee the accuracy and robustness of tracking, the embodiments of the present application realize accurate tracking of objects by comprehensively extracting and comparing multiple feature categories, under the premise that the camera placement positions are basically consistent, and improve the accuracy and reliability of object recognition and tracking. In addition, the feature extraction method of the present application has good applicability in multiple monitoring video scenes, can cope with object tracking tasks in different scenes, and the feature weight distribution method adopted can dynamically adjust the weight according to actual data, so that the algorithm has better flexibility and adaptability. The object in the present application is not limited to people, and is widely used in video monitoring, security, personnel management and other fields, and has high practical value.

[0043] Other features and advantages of the present application will be described in the following description, and some will become apparent from the description, or will be learned by practice of the present application. The purposes and other advantages of the present application can be achieved and obtained by the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly introduced as follows. Obviously, the drawings introduced below are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0045] Figure 1 Flowchart of an object tracking method according to an embodiment of the present application;

[0046] Figure 2 Flowchart of an object tracking method according to an embodiment of the present application;

[0047] Figure 3 Structure diagram of an object tracking device according to an embodiment of the present application;

[0048] Figure 4 Structure diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings. In the description of the embodiments of the present application, unless otherwise specified, " / " represents the meaning of or, for example, A / B can represent A or B; the "and / or" in the text only represents a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which means that there are three cases of A alone, A and B together, and B alone. In addition, in the description of the embodiments of the present application, "multiple" means two or more than two, and other quantifiers similar thereto should be understood as the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application, and the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0050] In the description of the embodiments of the present application, unless otherwise specified, the term "multiple" means two or more than two, and other quantifiers similar thereto should be understood as the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application, and the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0051] To further illustrate the technical solutions provided by the embodiments of the present application, the following will be described in detail in conjunction with the drawings and the specific embodiments. Although the embodiments of the present application provide the method operation steps as described in the following embodiments or drawings, more or fewer operation steps can be included in the method based on conventional or non-creative labor. The execution order of these steps is not limited to the execution order provided by the embodiments of the present application in the steps that do not have necessary causal relationship in logic. The method can be executed in sequence or in parallel when the method is executed in actual processing process or by control device according to the method order shown in the embodiments or drawings.

[0052] Other features and advantages of the present application will be described in the following specification, and some of them will become apparent from the specification, or will be understood by those skilled in the art. The purpose and other advantages of the present application can be achieved and obtained by the structure specifically pointed out in the written specification, claims, and drawings.

[0053] The object tracking method in the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0054] Referring to Figure 1 A flowchart of an object tracking method provided by an embodiment of the present application is shown, which includes the following steps:

[0055] Step 101: Obtain, from the monitoring video of each camera in the tracking area, a picture set including multiple pictures corresponding to each monitoring video.

[0056] Specifically, the tracking area includes multiple cameras, and the video formed by the area monitored by each camera in real time is a monitoring video. Each monitoring video is frame-extracted at a preset time interval to obtain multiple pictures arranged in time sequence, and the multiple pictures corresponding to the monitoring video of each camera form a picture set. For example, there are N cameras in the tracking area, and after frame extraction, N picture sets are obtained.

[0057] Step 102: According to the preset feature categories corresponding to different parts of the object, perform feature category recognition on the object in each picture in each picture set, and filter out the pictures in which the object is recognized and only one object is included according to the recognition result.

[0058] Specifically, the object is the object to be tracked, which can be a person or an animal, and is not limited here. The object to be tracked is determined according to actual tracking needs. The feature categories corresponding to different parts of the object refer to the feature categories used to represent the features of the object. As an optional implementation, the feature categories corresponding to different parts of the object include head inclination feature, height feature, moving speed feature, arm swing amplitude feature, width-height ratio feature, and step feature. Each picture in each picture set is subjected to target feature category recognition to detect the feature object representing the feature category of the object in each picture. For example, for the first picture in the picture set of the monitoring video of the first camera, the head inclination feature, height feature, moving speed feature, arm swing amplitude feature, width-height ratio feature, and step feature of the person included in the picture are framed out by a rectangular frame. The rectangular frame is a frame body framed out according to the preset feature categories corresponding to different parts of the object for each part of the object in each picture in each picture set. After the framing of all pictures in all picture sets is completed, the pictures in which the object exists and only the related features of one object are included are filtered out. For example, picture 1 includes the head of A and the arm of B, so picture 1 does not meet the filtering requirement and includes two objects; picture 2 includes the head of C, the foot of C, so picture 2 meets the filtering requirement; and picture 3 does not capture any feature category, so picture 3 does not meet the filtering requirement and does not include the object.

[0059] Step 103: Determine the first feature value of each feature category of the recognized object for each filtered picture.

[0060] Specifically, first, the first feature value of each feature category of each picture is calculated.

[0061] For the above-mentioned six feature categories, the first feature value calculation method corresponding to each feature category is as follows. It should be noted that the feature categories in the present application are not limited to the above-mentioned six feature categories.

[0062] I. When the feature category is the head inclination feature, the first feature value is the inclination angle of the head center point of the object and the center point of the object.

[0063] For example, when the tracked object is Zhang San, assuming that the head and body of Zhang San appear in the first picture in the picture set corresponding to the monitoring video of the first camera, then in this picture, the first feature value corresponding to the head inclination feature of Zhang San is the inclination angle of the head center point of Zhang San and the center point of Zhang San.

[0064] II. When the feature category is the height feature, the first feature value is the ratio of the first distance and the height of the object, wherein the first distance is the distance from the head center point of the object to the center point of the object.

[0065] For example, when the tracked object is Zhang San, assuming that the head and body of Zhang San appear in the third picture in the picture set corresponding to the monitoring video of the first camera, then in this picture, the first feature value corresponding to the height feature of Zhang San is the ratio of the distance from the head center point of Zhang San to the center point of Zhang San and the height of Zhang San.

[0066] III. When the feature category is the moving speed feature, the first feature value is the ratio of the second distance and the frame extraction time, wherein the second distance is the distance moved by the center coordinates of the object in the adjacent two pictures, and the frame extraction time is the frequency time of extracting pictures from the monitoring video.

[0067] For example, when the tracked object is Zhang San, assuming that the body of Zhang San appears in the fifth picture and the sixth picture (adjacent pictures) in the picture set corresponding to the monitoring video of the first camera, then the first feature value corresponding to the moving speed feature of Zhang San in the sixth picture is the ratio of the distance moved by the center coordinates of Zhang San in the fifth picture and the sixth picture and the frame extraction time.

[0068] IV. When the feature category is the arm swing amplitude feature, the first feature value is the average of the swing amplitude of each hand of the object, wherein the hand swing amplitude is determined by the vertical distance between the center point of the hand and the center of the object.

[0069] For example, when the tracked object is Zhang San, and Zhang San's arms and body appear in the eighth picture in the picture set corresponding to the monitoring video of the first camera, the first feature value corresponding to the arm swing amplitude feature of Zhang San in the eighth picture is the average of the swing amplitude of each hand of Zhang San.

[0070] V. When the feature category is the width-height ratio feature, the first feature value is the ratio of the width of the rectangular frame of the object to the height of the rectangular frame.

[0071] For example, when the tracked object is Zhang San, and Zhang San's body appears in the tenth picture in the picture set corresponding to the monitoring video of the first camera, the first feature value corresponding to the width-height ratio feature of Zhang San in the tenth picture is the ratio of the width of the rectangular frame of Zhang San to the height of the rectangular frame.

[0072] VI. When the feature category is the step feature, the first feature value is the average of the swing amplitude of each foot of the object, wherein the foot swing amplitude is determined by the vertical distance between the center point of the foot and the center of the object.

[0073] For example, when the tracked object is Zhang San, and Zhang San's feet and body appear in the twelfth picture in the picture set corresponding to the monitoring video of the first camera, the first feature value corresponding to the step feature of Zhang San in the twelfth picture is the average of the swing amplitude of each foot of Zhang San.

[0074] Step 104: determining the second feature value of the feature category according to the first feature values corresponding to different pictures in each picture set and belonging to the same feature category.

[0075] Specifically, the first feature values of the feature categories included in each picture in each picture set are calculated in step 103 above. For a picture set, each picture can include the first feature value of one feature category, or the first feature values of multiple feature categories.

[0076] For the feature value of any feature category, as an optional implementation, all the first feature values corresponding to different pictures in each picture set and belonging to the same feature category are averaged to obtain a first average value, and the second feature value is the first average value.

[0077] Specifically, the second feature value is the feature value of each picture set for a feature category. For example, the first camera's monitoring video corresponds to a picture set including 20 pictures after frame extraction. For the head tilt feature, the first, third, fourth, and sixth pictures include the head tilt feature and the first feature values of the respective corresponding pictures are calculated. Then, the head tilt feature value of the first camera's monitoring video corresponding picture set is the first average value of the first feature values of the head tilt features of the first, third, fourth, and sixth pictures.

[0078] Step 105: For the same feature category, the feature similarity value corresponding to the feature category is determined according to the similarity value of the second feature values between any two different picture sets.

[0079] Specifically, after the second feature values of each feature category of the picture set corresponding to the monitoring video of each camera are obtained by step 104, the similarity value of the second feature values between any two different picture sets in the tracking area needs to be calculated.

[0080] As an optional implementation, the similarity value of the second feature values between any two different picture sets is determined by the following method: determining the first difference value of the second feature value of the first picture set and the second feature value of the second picture set; determining the second difference value of the preset value and the first difference value; determining the ratio of the second difference value and the second feature value of the first picture set as the similarity value of the second feature values between the first picture set and the second picture set; wherein the second picture set is the previous picture set of the first picture set.

[0081] Specifically, the preset value in the present application is 1. For any feature category, the similarity value between any two different picture sets is calculated as 1-(the second feature value of the first picture set-the second feature value of the second picture set) / the second feature value of the first picture set.

[0082] Since there are multiple cameras in the tracking area, the monitoring videos of the multiple cameras correspond to multiple picture sets, and any two different picture sets have multiple combinations, for example, there are camera 1, camera 2, camera 3 and camera 4 in the tracking area, and the corresponding picture set 1, picture set 2, picture set 3 and picture set 4, any two picture sets include 6 combinations, including picture set 1 and picture set 2, picture set 1 and picture set 3, picture set 1 and picture set 4, picture set 2 and picture set 3, picture set 2 and picture set 4, and picture set 3 and picture set 4. As an optional implementation, for the same feature category, the similarity value of the second feature value corresponding to any two different picture sets is determined; and the average of the similarity values of all the second feature values is determined as the feature similarity value corresponding to the feature category. Specifically, for example, for the feature of human head inclination, the feature similarity value calculated by picture set 1 and picture set 2 is a, the feature similarity value calculated by picture set 1 and picture set 3 is b, the feature similarity value calculated by picture set 1 and picture set 4 is c, the feature similarity value calculated by picture set 2 and picture set 3 is d, the feature similarity value calculated by picture set 2 and picture set 4 is e, and the feature similarity value calculated by picture set 3 and picture set 4 is f, then the feature similarity value of the human head inclination corresponding to the picture set of the tracking area is the average of a, b, c, d, e and f.

[0083] Step 106: Based on the preset weight corresponding to each feature category and the feature similarity value, the probability of being the same object in each monitoring video in the tracking area is determined. As an optional implementation, the probability is specifically determined by the following manner: determining the first product of the weight corresponding to each feature category and the feature similarity value corresponding to the feature category; determining the sum of the first products corresponding to all feature categories; and the sum is the probability of being the same object in each monitoring video in the tracking area. Specifically, when the feature similarity value corresponding to each feature category in all cameras is calculated, based on the preset weight, the first product of the feature similarity value and the weight corresponding to the feature category is calculated, and when the sum tends to 1, it is inferred that the possibility of the object appearing in each camera in the tracking area being the same object is greater.

[0084] The object tracking technology based on object gait recognition provided in the present application realizes accurate tracking of objects, improves the accuracy and reliability of object recognition and tracking by comprehensively extracting and comparing multiple feature categories under the premise that the camera placement positions are basically consistent; in addition, the feature extraction method of the present application has good applicability in multiple monitoring video scenes and can cope with object tracking tasks in different scenes, and the feature weight distribution method adopted can dynamically adjust the weight according to actual data, so that the algorithm has better flexibility and adaptability; the object in the present application is not limited to people and is widely used in video monitoring, security, personnel management and other fields, and has high practical value.

[0085] Referring to Figure 2 The overall detailed flowchart of the object tracking method provided by the present application is shown.

[0086] Step 201: Obtain, from the monitoring video of each camera in the tracking area, a picture set including multiple pictures corresponding to each monitoring video.

[0087] Step 202: According to the preset feature categories corresponding to different parts of the object, perform feature category recognition on the object in each picture in each picture set, and filter out pictures in which the object is recognized and only one object is included according to the recognition result.

[0088] Step 203: For each picture filtered out, determine the first feature value of each feature category of the recognized object.

[0089] Step 204: Average all first feature values corresponding to different pictures in the same feature category in each picture set to obtain a first average value, and the second feature value is the first average value.

[0090] Step 205: For the same feature category, determine the first difference value between the second feature value of the first picture set and the second feature value of the second picture set.

[0091] Step 206: Determine the second difference value between the preset value and the first difference value.

[0092] Step 207: Determine the ratio of the second difference value to the second feature value of the first picture set as the similarity value of the corresponding second feature value between the first picture set and the second picture set.

[0093] Step 208: Determine the average value of the similarity values of all second feature values as the feature similarity value corresponding to the feature category.

[0094] Step 209: Determine the first product of the weight corresponding to each feature category and the feature similarity value corresponding to the feature category.

[0095] Step 210: Determine the sum of the first products corresponding to all feature categories.

[0096] Step 211: The closer the sum is to 1, the greater the possibility that the object appearing in each camera in the tracking area is the same object.

[0097] Embodiment 2

[0098] Based on the same inventive concept, the present application also provides an object tracking device, as shown in the figure, which comprises: Figure 3

[0099] ​The picture set acquisition module 301 is configured to acquire, from the monitoring video of each camera of the tracking area, a picture set including multiple pictures corresponding to each monitoring video;

[0100] The screening module 302 is configured to perform feature category identification on the object in each picture in each picture set according to a preset feature category corresponding to different parts of the object, and screen out pictures in which the object is identified and only one object is included according to the identification result;

[0101] The first feature value determination module 303 is configured to determine a first feature value of each feature category of the identified object for each screened picture.

[0102] The second feature value determination module 304 is configured to determine a second feature value of a feature category according to the first feature values of different pictures in each screened picture set that belong to the same feature category.

[0103] The feature similarity value determination module 305 is configured to determine, for the same feature category, a feature similarity value corresponding to the feature category according to a similarity value of the second feature values between any two different picture sets.

[0104] The probability determination module 306 is configured to determine a probability of the same object in each monitoring video of the tracking area based on a preset weight corresponding to each feature category and the feature similarity value.

[0105] Optionally, the feature category corresponding to each part includes a human head inclination feature, a height feature, a moving speed feature, an arm swing amplitude feature, an aspect ratio feature, and a step length feature.

[0106] Optionally, the first feature value determination module 303 is specifically configured to:

[0107] For each screened picture, when the feature category is the human head inclination feature, the first feature value is an inclination angle of a head center point of the object and a center point of the object.

[0108] When the feature category is the height feature, the first feature value is a ratio of a first distance and a height of the object, where the first distance is a distance from the head center point of the object to the center point of the object.

[0109] When the feature category is the moving speed feature, the first feature value is a ratio of a second distance and a frame extraction time, where the second distance is a distance of a center coordinate movement of the object in the adjacent two pictures, and the frame extraction time is a frequency time of extracting pictures from the monitoring video.

[0110] When the feature category is the swing arm amplitude feature, the first feature value is an average value of swing amplitudes of each hand of the object, wherein the hand swing amplitude is determined by a vertical distance between a center point of the hand and a center of the object;

[0111] When the feature category is the width-height ratio feature, the first feature value is a ratio of a width of a rectangular frame of the object to a height of the rectangular frame, wherein the rectangular frame is a frame body selected according to a feature category frame corresponding to each part of the object according to a preset feature category corresponding to different parts of the object;

[0112] When the feature category is the step amplitude feature, the first feature value is an average value of swing amplitudes of each foot of the object, wherein the foot swing amplitude is determined by a vertical distance between a center point of the foot and a center of the object.

[0113] Optionally, the determining second feature value module 304 is specifically configured to: average all first feature values corresponding to the same feature category in different pictures in each picture set to obtain a first average value, and the second feature value is the first average value.

[0114] Optionally, the determining feature similarity value module 305 is specifically configured to:

[0115] determine a first difference value between the second feature value of the first picture set and the second feature value of the second picture set;

[0116] determine a second difference value between the preset value and the first difference value;

[0117] determine a ratio of the second difference value to the second feature value of the first picture set as a similarity value of the corresponding second feature value between the first picture set and the second picture set;

[0118] The second picture set is a previous picture set of the first picture set.

[0119] Optionally, the determining feature similarity value module 305 is specifically configured to: for the same feature category, determine a similarity value of the corresponding second feature value between any two different picture sets; and determine an average value of the similarity values of all the second feature values as a feature similarity value corresponding to the feature category.

[0120] Optionally, the determining probability module 306 is specifically configured to: determine a first product of a weight corresponding to each feature category and a feature similarity value corresponding to the feature category; determine a sum of the first products corresponding to all the feature categories; and the sum is a probability of the same object in each monitoring video in the tracking area.

[0121] After introducing the object tracking method and device of the exemplary embodiments of the present application, next, the electronic device according to another exemplary embodiment of the present application is introduced.

[0122] Those skilled in the art can understand that various aspects of the present application can be implemented as a system, a method or a program product. Therefore, various aspects of the present application can be embodied as a whole hardware embodiment, a whole software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuitry", "module" or "system" herein.

[0123] In some possible embodiments, the electronic device according to the present application can at least include at least one processor, and at least one memory. Wherein the memory stores program code, when the program code is executed by the processor, the processor executes the steps in the object tracking method according to various exemplary embodiments of the present application described above in the specification.

[0124] The electronic device 130 according to this embodiment of the present application will be described below with reference to Figure 4 , which is an object tracking device. Figure 4 The electronic device 130 shown is merely an example, and should not impose any limitation on the function and scope of use of the embodiments of the present application.

[0125] As shown in Figure 4 , the electronic device 130 is in the form of a general electronic device. The components of the electronic device 130 can include, but are not limited to, the at least one processor 131 described above, the at least one memory 132 described above, and the bus 133 connecting different system components, including the memory 132 and the processor 131.

[0126] The bus 133 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a processor or local bus using any of a variety of bus structures.

[0127] The memory 132 can include a readable medium in the form of volatile memory, such as a random access memory (RAM) 1321 and / or a cache memory 1322, and can further include a read-only memory (ROM) 1323.

[0128] The memory 132 can also include program / utility 1325 having a set of (at least one) program modules 1324, including but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or a combination thereof can include implementation of a network environment.

[0129] The electronic device 130 can also communicate with one or more external devices 134 such as a keyboard or a pointing device, through an input / output (I / O) interface 135. And, the electronic device 130 can communicate with one or more devices that enable user interaction with the electronic device 130, and / or one or more devices that enable communication of the electronic device 130 with one or more other electronic devices. This communication can be via the I / O interface 135. Further, the electronic device 130 can communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or the public network, such as the Internet) through a network adapter 136. As depicted, the network adapter 136 is in communication with the other modules of the electronic device 130 through the bus 133. It should be appreciated that other hardware and / or software modules can be used in conjunction with the electronic device 130, such as, but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.

[0130] In some possible embodiments, each aspect of the object tracking method provided by the present application can also be implemented in the form of a program product, which includes program codes for causing a computer device to perform the steps of the object tracking method according to various exemplary embodiments of the present application described above in the specification when the program product is run on the computer device.

[0131] The program product can employ any combination of one or more computer-readable media. The computer-readable media can be a computer-readable storage medium or a computer-readable signal medium. The computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0132] The program product for monitoring of the embodiments of the present application can employ a portable compact disc read-only memory (CD-ROM) and include program codes, and can be run on an electronic device. However, the program product of the present application is not limited thereto, and in this document, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0133] A readable signal medium can be any medium that can be read by a machine (e.g., a computer) and can contain various kinds of information. Examples of a readable signal medium include, but are not limited to, a floppy disk, a flexible disk, a hard disk, a magnetic tape, a magnetic disk, an optical disk, a compact disc (CD), a compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a Blu-ray disc, a memory stick, a memory card, a memory stick, a random access memory (RAM), a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), a flash memory device, a solid state drive (SSD), a storage device, a storage medium, a storage, a memory device, or a machine readable medium, etc. In some embodiments, a readable signal medium can be a computer readable medium.

[0134] The program code included in a readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical, RF, etc., or any suitable combination of the foregoing.

[0135] The program code for implementing the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, etc., and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's electronic device, partly on the user's electronic device, as a stand-alone software package, partly on the user's electronic device and partly on a remote electronic device or entirely on the remote electronic device or server. In the latter scenario, the remote electronic device can be connected to the user's electronic device through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external electronic device (for example, through the Internet using an Internet Service Provider). The present application can be implemented in a computing system that includes a back-end component, or a middleware component, or a front-end component, or any combination thereof. The components of the present application can be implemented in a computing system that includes users' electronic device with a display device coupled to it, e.g., a CRT (cathode ray tube), LCD (liquid crystal display) monitor, or touch screen with a keyboard or a pointing device, e.g., a mouse, etc. The computing system can include peripherals coupled to it, including a printing device, a scanner, a fax machine, a disc reader or a writer, or a drive, etc. The computing system can also include a computer- readable medium, e.g., a storage device.

[0136] It should be noted that although the above detailed description refers to several units or sub-units of a device, such division into units or sub-units is merely illustrative and not mandatory. In fact, according to an embodiment of the present application, features and functions of two or more units described above can be embodied in one unit. Conversely, a feature or function of one unit described above can be divided into several sub-units to be embodied by several units.

[0137] Furthermore, although the operations of the method according to the present application are described in a particular, sequential order, this order is not necessarily a mandatory order, and merely indicates one feasible ordering of the operations. One step can have been performed in a different order or omitted. The description similarly applies to workings and the like.

[0138] Those skilled in the art will appreciate that embodiments of the application can be devised for a method, a system, or a computer program product. Accordingly, the present application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0139] The present application is described in reference to the flow diagrams and block diagrams of the method, apparatus (system), and computer program product according to embodiments of the application. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device, 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 flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.

[0140] These computer program instructions can 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 instructions which implement the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.

[0141] These computer program instructions can 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 flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.

[0142] While preferred embodiments of the application have been described, modifications and variations can be apparent to those skilled in the art once aware of the general underlying concepts. Accordingly, the appended claims intend to embrace all such modifications and variations as fall within the scope of the application.

[0143] Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. An object tracking method, characterized in that, The method includes: From the surveillance video of each camera in the tracking area, obtain an image set containing multiple images corresponding to each surveillance video; wherein, the image set corresponding to each surveillance video is obtained in the following way: frames are extracted from each surveillance video at preset time intervals to obtain multiple images arranged in chronological order, and the multiple images corresponding to each camera's surveillance video constitute an image set; Based on the preset feature categories corresponding to different parts of the object, the feature categories of the objects in each image in each image set are identified, and the images that identify the object and contain only one object are selected based on the identification results. For each selected image, determine the first feature value of each feature category of the identified object; Based on the first feature value corresponding to the same feature category for different images in each selected image set, determine the second feature value of the feature category; For the same feature category, the feature similarity value corresponding to that feature category is determined based on the similarity value of the second feature values ​​corresponding to any two different image sets; wherein, the similarity value of the second feature values ​​corresponding to any two different image sets is determined by the following method: determining a first difference between the second feature values ​​of the first image set and the second feature values ​​of the second image set; determining a second difference between a preset value and the first difference; determining the ratio of the second difference to the second feature values ​​of the first image set as the similarity value of the second feature values ​​corresponding to the first image set and the second image set; wherein, the second image set is the image set preceding the first image set; Based on the preset weights and feature similarity values ​​corresponding to each feature category, the probability of the same object being found in each surveillance video within the tracking area is determined.

2. The method according to claim 1, characterized in that, The feature categories corresponding to the aforementioned body parts include: head tilt feature, height feature, movement speed feature, arm swing amplitude feature, width-to-height ratio feature, and stride feature.

3. The method according to claim 1, characterized in that, For each of the selected images, determining the first feature value of each feature category of the identified object includes: For each of the selected images, when the feature category is head tilt feature, the first feature value is the tilt angle between the center point of the object's head and the center point of the object. When the feature category is height feature, the first feature value is the ratio of the first distance to the height of the object, wherein the first distance is the distance between the center point of the object's head and the center point of the object. When the feature category is the movement speed feature, the first feature value is the ratio of the second distance to the frame extraction time, wherein the second distance is the distance the center coordinates of the object move between two adjacent images, and the frame extraction time is the frequency of extracting images from the surveillance video. When the feature category is arm swing amplitude feature, the first feature value is the average value of the arm swing amplitude of each hand of the object, wherein the arm swing amplitude is determined by the vertical distance between the center point of the hand and the center of the object; When the feature category is aspect ratio, the first feature value is the ratio of the width of the rectangle of the object to the height of the rectangle. The rectangle is a frame selected according to the feature categories corresponding to different parts of the object in each image of each image set. When the feature category is stride feature, the first feature value is the average value of the swing amplitude of each foot of the object, wherein the foot swing amplitude is determined by the vertical distance between the center point of the foot and the center of the object.

4. The method according to claim 1, characterized in that, The step of determining the second feature value of the feature category based on the first feature value corresponding to different images in each selected image set that belong to the same feature category includes: The first average value is obtained by averaging all the first feature values ​​belonging to the same feature category for different images in each image set, and the second feature value is the first average value.

5. The method according to claim 1, characterized in that, For the same feature category, determining the feature similarity value corresponding to that feature category based on the similarity value of the second feature values ​​between any two different image sets includes: For the same feature category, determine the similarity value of the corresponding second feature values ​​between any two different image sets; The average of the similarity values ​​of all the second feature values ​​is determined as the feature similarity value corresponding to the feature category.

6. The method according to claim 1, characterized in that, The method of determining the probability of the same object in each surveillance video within the tracking area based on the preset weights and feature similarity values ​​corresponding to each feature category includes: Determine the first product of the weight corresponding to each feature category and the feature similarity value corresponding to that feature category; Determine the sum of the first products corresponding to all feature categories; The sum represents the probability of the same object appearing in each surveillance video within the tracking area.

7. An object tracking device, characterized in that, The device includes: The image set acquisition module is used to acquire an image set, which includes multiple images, from the surveillance video of each camera in the tracking area. The image set corresponding to each surveillance video is acquired in the following way: frames are extracted from each surveillance video at preset time intervals to obtain multiple images arranged in chronological order. The multiple images corresponding to the surveillance video of each camera constitute an image set. The filtering module is used to identify the feature categories of objects in each image in each image set according to the preset feature categories corresponding to different parts of the object, and to filter out the images that identify an object and contain only one object based on the recognition results. The module for determining the first feature value is used to determine the first feature value of each feature category of the identified object for each of the selected images; The module for determining the second feature value is used to determine the second feature value of the feature category based on the first feature value corresponding to different images in each of the selected image sets that belong to the same feature category; The feature similarity value determination module is used to determine the feature similarity value corresponding to the feature category for the same feature category based on the similarity value of the corresponding second feature values ​​between any two different image sets; wherein, the similarity value of the corresponding second feature values ​​between any two different image sets is determined by the following method: determining a first difference between the second feature values ​​of the first image set and the second feature values ​​of the second image set; determining a second difference between a preset value and the first difference; determining the ratio of the second difference to the second feature values ​​of the first image set as the similarity value of the corresponding second feature values ​​between the first image set and the second image set; wherein, the second image set is the image set preceding the first image set; The probability determination module is used to determine the probability that the object is the same in each surveillance video in the tracking area, based on the preset weight and feature similarity value corresponding to each feature category.

8. An electronic device, characterized in that, The method includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The computer storage medium stores a computer program that enables the computer to perform the method as described in any one of claims 1-6.

Images