Cross-terminal target continuous tracking method and tracking system

Abstract

The invention relates to a cross-terminal target continuous tracking method and tracking system, and the method comprises the steps: determining a tracking object in image data sent back by a first terminal according to an obtained instruction; establishing a feature region by taking tracking as a center, and identifying other objects in the feature region to obtain a reference object; judging an association relationship between the reference object and the tracking object; retrieving a reference object in the image data sent back by the second terminal, and when the reference object exists in the image data sent back by the second terminal, determining a retrieval range according to the position of the second terminal; searching the tracking object in the retrieval range; matching the tracking object and the suspected tracking object belonging to the same moving track; and marking the successfully matched suspected tracking object as a tracking object. According to the cross-terminal target continuous tracking method and tracking system disclosed by the invention, continuous tracking of the target is realized in a mode of matching different terminals, and continuous position management of personnel in a coverage range is further realized.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for continuous tracking of a target across terminals. Background Art

[0002] Smart construction site management refers to the integrated management of personnel, equipment, environment, and construction at the construction site through a combination of various means, to achieve centralized display and intelligent analysis of information to ensure project progress, project safety, and risk control.

[0003] In complex collaborative production sites, overall management of personnel is required. One purpose is safety management. For example, during collaborative work, personnel can be quickly identified and evacuated, and it can also play a role in personnel locating.

[0004] The commonly used method at present is to rely on cameras to obtain images, but the coverage of cameras is limited, and there are difficulties in finding people, locating people and continuously tracking them. This is mainly due to factors such as image clarity, difficulty in cross-camera recognition, and lack of basic information on personnel models. Once the camera has a blind spot, it faces the problem of losing the target. Summary of the invention

[0005] The present application provides a method and system for continuous target tracking across terminals, which realizes continuous tracking of targets by matching between different terminals, thereby realizing continuous location management of people within the coverage area.

[0006] The above-mentioned purpose of the present application is achieved through the following technical solutions:

[0007] In a first aspect, the present application provides a method for continuous tracking of a target across terminals, comprising:

[0008] According to the acquired instruction, determining the tracking object in the image data sent back by the first terminal;

[0009] Establish a feature area with tracking as the center and identify other objects in the feature area to obtain a reference object;

[0010] Determine the relationship between the reference object and the tracking object;

[0011] Retrieving the reference object in the image data sent back by the second terminal, and when the reference object exists in the image data sent back by the second terminal, determining the retrieval range according to the position of the second terminal;

[0012] Search for the tracking object within the search range and find the suspected tracking object;

[0013] Using the tracking object and the suspected tracking object to generate a movement trajectory of the tracking object and using the movement trajectory to exclude some of the suspected tracking objects;

[0014] Match the tracking objects and suspected tracking objects that belong to the same moving trajectory;

[0015] The suspected tracking object that is successfully matched is recorded as the tracking object.

[0016] In a possible implementation manner of the first aspect, determining the association relationship between the reference object and the tracking object includes:

[0017] Obtaining the body movements and facial orientation changes of the tracking object and the body movements and facial orientation changes of the reference object;

[0018] Determine the spatial correlation between the body movements or facial orientation changes of the tracked object and the body movements or facial orientation changes of the reference object and count the number of correlations;

[0019] When the number is greater than or equal to the allowed number, it is determined that the reference object has an association relationship with the tracked object.

[0020] In a possible implementation manner of the first aspect, determining the spatial correlation between the body movement or facial orientation change of the tracked object and the body movement or facial orientation change of the reference object includes:

[0021] Drawing a body movement trajectory according to the body movement of the object, the object includes a tracking object and a reference object;

[0022] Drawing a facial orientation change trajectory according to the facial orientation change of the object, the object including the tracking object and the reference object;

[0023] When the body movement trajectories and / or facial orientation change trajectories of any two subjects match, the number of associations is counted;

[0024] The matching of the limb movement trajectories and / or the facial orientation change trajectories of any two objects overlaps in the time dimension and the length of the overlapping area is greater than the required length or greater than the required ratio.

[0025] In a possible implementation manner of the first aspect, determining the search range according to the location of the second terminal includes:

[0026] Using the location of the second terminal to establish a search area;

[0027] The search area is segmented using the moving trajectory of the reference object to obtain multiple search sub-areas;

[0028] According to the determined relative position of the tracking object and the reference object;

[0029] A search sub-region is selected using the relative positions of the tracking object and the reference object, and the selected search sub-region is used as a search range.

[0030] In a possible implementation manner of the first aspect, using the movement trajectory to exclude some suspected tracking objects includes:

[0031] Assign coordinate points to the tracked objects and suspected tracked objects within the spatial range;

[0032] Connecting the tracking object and the suspected tracking object according to the order of appearance time to obtain a movement trajectory, wherein the movement trajectory includes multiple segments of sub-movement trajectories connected in sequence;

[0033] Calculate the moving speed of the sub-movement trajectory and use the moving speed to exclude some suspected tracking objects;

[0034] Among them, the continuity of multiple suspected tracking objects is determined with reference to the surrounding environment, and the continuity is assigned to the movement trajectory.

[0035] In a possible implementation manner of the first aspect, assigning continuity to the movement trajectory includes:

[0036] Determine the surrounding environment characteristics of a suspected tracking object;

[0037] Use the suspected tracking object and the surrounding environment features to form a feature matrix and record the movement trajectory of the feature matrix;

[0038] The movement trajectory of the feature matrix is ​​used to replace the movement trajectory of the suspected tracking object.

[0039] In a possible implementation manner of the first aspect, matching the tracking object and the suspected tracking object belonging to the same moving trajectory includes:

[0040] Transferring the tracking object into the grayscale domain and picking up a first grayscale feature group, the first grayscale feature group including a shape and a size;

[0041] Transferring the suspected tracking object into the grayscale domain and picking up a second grayscale feature group, the second grayscale feature group including a pattern and a size;

[0042] Putting similar graphics in the first grayscale feature group and the second grayscale feature group into a graphic group;

[0043] Determine the change vector of the graphic group with the help of the size and count the dispersion of the change vector;

[0044] The suspected tracking object with a discreteness less than or equal to the allowed value is recorded as the tracking object;

[0045] Among them, any two objects on the moving trajectory need to be matched, and the objects include tracking objects and suspected tracking objects.

[0046] In a second aspect, the present application provides a cross-terminal target continuous tracking device, comprising:

[0047] A first data acquisition unit, configured to determine a tracking object in the image data sent back by the first terminal according to the acquired instruction;

[0048] A second data acquisition unit is used to establish a feature area with the tracking as the center and identify other objects in the feature area to obtain a reference object;

[0049] A first judgment unit, used to judge the association relationship between the reference object and the tracking object;

[0050] A search range processing unit, used to search for a reference object in the image data sent back by the second terminal, and when the reference object exists in the image data sent back by the second terminal, determine the search range according to the position of the second terminal;

[0051] A third data acquisition unit is used to search for the tracking object within the search range and find the suspected tracking object;

[0052] A moving trajectory unit, used to generate a moving trajectory of the tracking object using the tracking object and the suspected tracking object and to exclude part of the suspected tracking object using the moving trajectory;

[0053] A matching unit, used for matching the tracking object and the suspected tracking object belonging to the same moving trajectory;

[0054] The first marking unit is used to record the suspected tracking object that has been successfully matched as the tracking object.

[0055] In a third aspect, the present application provides a cross-terminal target continuous tracking system, the system comprising:

[0056] one or more memories for storing instructions; and

[0057] One or more processors, used to call and run the instructions from the memory to execute the method as described in the first aspect and any possible implementation of the first aspect.

[0058] In a fourth aspect, the present application provides a computer-readable storage medium, the computer-readable storage medium comprising:

[0059] Program, when the program is executed by a processor, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0060] In a fifth aspect, the present application provides a computer program product, comprising program instructions. When the program instructions are executed by a computing device, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0061] In a sixth aspect, the present application provides a chip system, which includes a processor for implementing the functions involved in the above aspects, for example, generating, receiving, sending, or processing the data and / or information involved in the above methods.

[0062] The chip system may be composed of chips, or may include chips and other discrete devices.

[0063] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wired or wireless means, or the processor and the memory can also be coupled on the same device. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a schematic flowchart of the steps of a method for continuous target tracking provided by the present application.

[0065] Figure 2 It is a principle schematic diagram of obtaining a reference object provided by this application.

[0066] Figure 3 It is a schematic diagram of a movement trajectory provided by this application.

[0067] Figure 4 This is a schematic diagram of the principle of judging whether two trajectories are consistent with each other, provided by the present application.

[0068] Figure 5 This is a schematic diagram of the principle of obtaining a search sub-area provided by the present application.

[0069] Figure 6 This is a schematic diagram of obtaining a movement trajectory provided by the present application.

[0070] Figure 7 This is a schematic diagram of screening movement trajectories provided by the present application.

[0071] Figure 8 This is a schematic diagram provided by the present application of projecting a graphic onto a reference plane. DETAILED DESCRIPTION

[0072] The technical solution in this application is further described in detail below in conjunction with the accompanying drawings.

[0073] This application discloses a method for continuous tracking of targets across terminals. Figure 1 In some examples, the cross-terminal target continuous tracking method disclosed in the present application includes the following steps:

[0074] S101, determining a tracking object in the image data sent back by the first terminal according to the acquired instruction;

[0075] S102, establishing a feature area with the tracking as the center and identifying other objects in the feature area to obtain a reference object;

[0076] S103, determining the association relationship between the reference object and the tracking object;

[0077] S104, searching for a reference object in the image data sent back by the second terminal, and when the reference object exists in the image data sent back by the second terminal, determining a search range according to the location of the second terminal;

[0078] S105, searching for a tracking object within the search range to find a suspected tracking object;

[0079] S106, using the tracking object and the suspected tracking object to generate a movement trajectory of the tracking object and using the movement trajectory to exclude some of the suspected tracking objects;

[0080] S107, matching the tracking object and the suspected tracking object belonging to the same moving trajectory;

[0081] S108: Record the suspected tracking object that has been successfully matched as the tracking object.

[0082] First of all, it should be explained that this application is applied to an analysis server, which collects images through cameras deployed within the coverage area (such as a construction site), and then uses the images to continuously track the tracking object. Currently, one of the difficulties in continuous tracking is the discontinuity of the tracking object caused by cross-terminals (cameras), and the information displayed by the tracking object in the image is unstable, which makes it easy to lose track during continuous tracking.

[0083] In the present application, specifically, in step S101, firstly, according to the acquired instruction, the tracking object is determined in the image data sent back by the first terminal, and then a feature area is established with the tracking as the center and other objects in the feature area are identified to obtain a reference object, that is, the content in step S102, such as Figure 2 shown.

[0084] For the feature area, generally speaking, it is a circular area established based on the tracking object. The radius of the circular area is a set value, generally 0.5-1 meter.

[0085] The reference objects here refer to other objects in motion, including people and equipment.

[0086] Then, in step S103, the association relationship between the reference object and the tracking object is determined. The purpose of using the reference object is to expand the feature library of the tracking object.

[0087] It should be understood that the identification of the tracking object is based on the features belonging to the tracking object. When there are fewer features belonging to the tracking object, it is necessary to expand the feature library of the tracking object. In this application, a reference object is used to expand the feature library of the tracking object.

[0088] In step S104, the reference object is searched in the image data sent back by the second terminal. When the reference object exists in the image data sent back by the second terminal, the search range is determined according to the position of the second terminal, that is, the range where the reference object appears, and the tracking object may also appear in this range.

[0089] Then execute step S105, in which the tracking object is searched within the search range to find the suspected tracking object. The suspected tracking object is still found through the feature library mentioned above. For example, if one or two (or refer to the set rules) features in the feature library are found on an object, the object is recorded as a suspected tracking object.

[0090] In step S106, the tracking object and the suspected tracking object are used to generate a moving trajectory of the tracking object, and the moving trajectory is used to exclude some suspected tracking objects. The moving trajectory obtained at this time is as follows: Figure 3 As shown, there are two possibilities: reasonable and unreasonable. By excluding unreasonable movement trajectories, some suspected tracking objects can be excluded.

[0091] Then, in step S107, the tracking objects and suspected tracking objects belonging to the same moving trajectory are matched, and the results are either matched or not matched. At this time, the unmatched suspected tracking objects are excluded. The matched suspected tracking objects are recorded as tracking objects, which is the content of step S108.

[0092] One application scenario of the above method is to specify a tracking object in the acquired video, and then the analysis server can give the movement trajectory of the tracking object. After obtaining the movement trajectory of the tracking object, analysis can be performed based on the movement trajectory of the tracking object to obtain the required results.

[0093] The application scenarios of this application are mainly multi-person scenarios. Multi-person scenarios have interference factors such as position changes and mutual occlusion. The technical solution in this application just uses the information provided by multiple people to achieve continuous tracking of the tracking object.

[0094] In some examples, the specific method of determining the association relationship between the reference object and the tracking object is:

[0095] S201, obtaining the body movements and facial orientation changes of the tracking object and the body movements and facial orientation changes of the reference object;

[0096] S202, determining the spatial correlation between the body movements or facial orientation changes of the tracked object and the body movements or facial orientation changes of the reference object and counting the number of correlations;

[0097] S203: When the number is greater than or equal to the allowed number, it is determined that the reference object has an association relationship with the tracking object.

[0098] The contents of step S201 to step S203 are to determine the association between the reference object and the tracking object through body movements and facial orientation changes. The specific determination method is based on spatial association.

[0099] When two actions (limb movement and limb movement, facial orientation change and facial orientation change, limb movement and facial orientation change) are considered to be spatially associated, the number of associations between the reference object and the tracking object will be accumulated. When the number is greater than or equal to the allowed number, it is determined that the reference object and the tracking object are associated.

[0100] The method of determining the spatial correlation between the body movement or facial orientation change of the tracked object and the body movement or facial orientation change of the reference object is:

[0101] Drawing a body movement trajectory according to the body movement of the object, the object includes a tracking object and a reference object;

[0102] Drawing a facial orientation change trajectory according to the facial orientation change of the object, the object including the tracking object and the reference object;

[0103] When the body movement trajectories and / or facial orientation change trajectories of any two subjects match, the number of associations is counted;

[0104] The matching of the limb movement trajectories and / or the facial orientation change trajectories of any two objects overlaps in the time dimension and the length of the overlapping area is greater than the required length or greater than the required ratio.

[0105] The above method is to judge through motion trajectory. The body movement draws the body movement trajectory, and the facial orientation change draws the facial orientation change trajectory. The facial orientation change trajectory is an arc, which is drawn with a facial feature (such as eyes, nose, mouth, etc.). Each feature is set with an inherent parameter. For example, taking the eyes as an example, a rotation radius will be assigned to the eyes, and the rotation radius is obtained by taking the average value after measuring multiple people.

[0106] The trajectory of a limb movement is a line segment, a combination of multiple line segments connected end to end, or an arc.

[0107] See also Figure 4When the body movement trajectories and / or facial orientation change trajectories of any two objects match, they must overlap in the time dimension and the length of the overlapping area must be greater than the required length or greater than the required ratio, where the length and ratio are both set values.

[0108] The main consideration here is the reaction time. For example, when the tracked object makes an action, the reaction time of the reference object should be limited. The limitation methods include complete overlap in the time dimension, the length of the overlapping area is greater than the required length, the length of the overlapping area is greater than the required proportion, etc.

[0109] For matching, we place one end of the body movement trajectory and / or the facial orientation change trajectory of any two objects at the same point, and then rotate and proportionally adjust one of the trajectories (body movement trajectory, facial orientation change trajectory) so that the two trajectories can overlap as much as possible. Then calculate the length of the overlapping area. When the length accounts for

[0110] The required length and required ratio are generally limited according to the recognition accuracy. The higher the recognition accuracy, the larger the values ​​of the required length and required ratio. Conversely, the smaller the values ​​of the length and required ratio.

[0111] In some examples, the method of determining the search range according to the location of the second terminal is:

[0112] S301, establishing a search area using the location of the second terminal;

[0113] S302, segmenting the search area using the moving trajectory of the reference object to obtain a plurality of search sub-areas;

[0114] S303, according to the determined relative position of the tracking object and the reference object;

[0115] S304: select a search sub-region using the relative position of the tracking object and the reference object, and use the selected search sub-region as a search range.

[0116] In the above method, please refer to Figure 5 , the moving trajectory of the reference object will be used to divide the search area into multiple search sub-areas. Generally, the number of search sub-areas is two. Here, the distance of the reference object in the horizontal direction is used, and the number of search sub-areas is two.

[0117] Then, the search sub-area is selected according to the relative position of the tracking object and the reference object. For example, if the tracking object is located on the left side of the reference object, the search sub-area on the left side is selected as the search range. The purpose of this method is to reduce the area of ​​the search range. At this time, the probability of the tracking object existing in the search range is high.

[0118] However, when it is finally determined that there is no tracking object in the search range, the search range needs to be expanded.

[0119] In some examples, the steps to use motion trajectories to exclude some suspected tracking objects are as follows:

[0120] S401, assigning coordinate points to the tracking object and the suspected tracking object within the space range;

[0121] S402, connecting the tracking object and the suspected tracking object according to the order of appearance time to obtain a movement trajectory, where the movement trajectory includes multiple sub-movement trajectories connected in sequence;

[0122] S403, calculating the moving speed of the sub-movement trajectory, and using the moving speed to exclude some suspected tracking objects;

[0123] Among them, the continuity of multiple suspected tracking objects is determined with reference to the surrounding environment, and the continuity is assigned to the movement trajectory.

[0124] See also Figure 6 In the above method, the suspected tracking objects are sorted in time order, and then connected in sequence to obtain a movement trajectory. At this time, the position of the suspected tracking object can be regarded as a point. The movement trajectory includes multiple sub-movement trajectories connected in sequence, and each sub-movement trajectory includes only two points.

[0125] Then the moving speed of the sub-moving trajectory can be calculated, and the moving speed can be used to eliminate some suspected tracking objects. Regarding the moving speed, a reference value (given manually) or the actual moving speed of the tracking object in the above method can be used. When there is no actual moving speed, a reference value (given manually) is used.

[0126] The above method can be used to exclude some sub-movement trajectories. Figure 6 and Figure 7 .

[0127] When excluding, only the suspected tracking object at the end of the sub-movement trajectory is deleted, because the above method can only determine that the suspected tracking object at the end of the sub-movement trajectory is wrong.

[0128] Then repeat the above process to obtain one or more moving trajectories.

[0129] In addition, the continuity of multiple suspected tracking objects can be determined by referring to the surrounding environment, and the continuity can be assigned to the movement trajectory. For example, a suspected tracking object appears at different time points, and there are other objects in its surrounding environment. These objects can determine the continuity of the suspected tracking object.

[0130] At this time, continuity is given to the moving trajectory, that is, this part of the moving trajectory is directly fixed, rather than being obtained by connecting according to the order of appearance time.

[0131] The specific way to give continuity to the movement track is as follows:

[0132] Determine the surrounding environment characteristics of a suspected tracking object;

[0133] Use the suspected tracking object and the surrounding environment features to form a feature matrix and record the movement trajectory of the feature matrix;

[0134] The movement trajectory of the feature matrix is ​​used to replace the movement trajectory of the suspected tracking object.

[0135] When multiple movement tracks are obtained, the tracking objects and suspected tracking objects belonging to the same movement track are matched for processing. This is because even if the movement track is processed by the speed screening method, there is still a possibility that the tracking object and the suspected tracking object belong to two or more objects.

[0136] In some examples, the steps for matching tracked objects and suspected tracked objects belonging to the same motion trajectory are as follows:

[0137] S501, transferring the tracking object to the grayscale domain and picking up a first grayscale feature group, the first grayscale feature group including a pattern and a size;

[0138] S502, transferring the suspected tracking object to the grayscale domain and picking up a second grayscale feature group, where the second grayscale feature group includes a pattern and a size;

[0139] S503, putting similar graphics in the first grayscale feature group and the second grayscale feature group into one graphic group;

[0140] S504, determining the change vector of the graphic group by means of the size and counting the discreteness of the change vector;

[0141] S505, recording the suspected tracking object whose discreteness is less than or equal to the allowed value as the tracking object;

[0142] Among them, any two objects on the moving trajectory need to be matched, and the objects include tracking objects and suspected tracking objects.

[0143] In step S501 to step S505, the tracking object is first transferred to the grayscale domain, that is, the tracking object is grayscale processed, and then the tracking object after grayscale processing is picked up to obtain the first grayscale feature group. The first grayscale feature group includes graphics and size, and the size is the size of the graphics. The same processing method is performed on the suspected tracking object.

[0144] Next, similar graphics in the first grayscale feature group and the second grayscale feature group are placed into a graphic group, and then the change vector of the graphic group is determined by using the size. The change vector is obtained by comparing the specific change direction when one graphic in the graphic group changes to another graphic. A specific method is to establish a reference plane, place the first graphic in the graphic group on the reference plane, and move the second graphic in space and project it on the reference plane, such as Figure 8 shown.

[0145] When the projected figure matches the first figure located on the reference plane, a direction vector is generated according to the movement of the second figure in space.

[0146] For the change vector, a quantitative expression is to assign different directions to the six degrees of freedom (X-axis movement, Y-axis movement, Z-axis movement, X-axis rotation, Y-axis rotation, and Z-axis movement), and then use the movement amount (angle, distance) as the length in each direction. That is, decompose the movement of the second figure in space into six basic movements.

[0147] At this point, multiple change vectors will be obtained, and then the discreteness of the change vectors will be counted. The discreteness of the change vector refers to the similarity of two change vectors. One evaluation method is to put the two change vectors into a coordinate system so that their starting points or end points coincide with each other, and then move one of the change vectors to make the two change vectors coincide as much as possible. Then, the proportion of the length of the overlapping part is counted to get a proportion value.

[0148] When counting the length of the overlapping portion, if the maximum straight-line distance of a portion of two change vectors is within the allowable range (set value), the portion is also considered to overlap.

[0149] After processing each group (two) of change vectors, a proportion value is obtained, and then the discreteness of these proportion values ​​(change vectors) is counted. The statistical method of discreteness is to remove the maximum value and the minimum value in the proportion value, and then obtain an interval, which has a length.

[0150] Finally, the suspected tracking objects whose discreteness is less than or equal to the allowed value are recorded as tracking objects. The allowed value here is a set value related to the recognition accuracy.

[0151] In the above process, any two objects on the moving trajectory need to be matched, including the tracking object and the suspected tracking object.

[0152] The present application also provides a cross-terminal target continuous tracking device, comprising:

[0153] A first data acquisition unit, configured to determine a tracking object in the image data sent back by the first terminal according to the acquired instruction;

[0154] A second data acquisition unit is used to establish a feature area with the tracking as the center and identify other objects in the feature area to obtain a reference object;

[0155] A first judgment unit, used to judge the association relationship between the reference object and the tracking object;

[0156] A search range processing unit, used to search for a reference object in the image data sent back by the second terminal, and when the reference object exists in the image data sent back by the second terminal, determine the search range according to the position of the second terminal;

[0157] A third data acquisition unit is used to search for the tracking object within the search range and find the suspected tracking object;

[0158] A moving trajectory unit, used to generate a moving trajectory of the tracking object using the tracking object and the suspected tracking object and to exclude part of the suspected tracking object using the moving trajectory;

[0159] A matching unit, used for matching the tracking object and the suspected tracking object belonging to the same moving trajectory;

[0160] The first marking unit is used to record the suspected tracking object that has been successfully matched as the tracking object.

[0161] Furthermore, it also includes:

[0162] A fourth data acquisition unit, used to acquire the body movements and facial orientation changes of the tracking object and the body movements and facial orientation changes of the reference object;

[0163] A second determination unit is used to determine the spatial correlation between the body movement or facial orientation change of the tracked object and the body movement or facial orientation change of the reference object and to count the number of correlations;

[0164] The first determining unit is configured to determine that an association relationship exists between the reference object and the tracking object when the number of times is greater than or equal to the allowed number of times.

[0165] Furthermore, it also includes:

[0166] A first trajectory drawing unit, used for drawing a body movement trajectory according to the body movement of the object, the object including a tracking object and a reference object;

[0167] A second trajectory drawing unit, used for drawing a facial orientation change trajectory according to the facial orientation change of the object, the object including the tracking object and the reference object;

[0168] A counting unit, used for counting the number of associations when the body movement trajectories and / or facial orientation change trajectories of any two objects match;

[0169] The matching of the limb movement trajectories and / or the facial orientation change trajectories of any two objects overlaps in the time dimension and the length of the overlapping area is greater than the required length or greater than the required ratio.

[0170] Furthermore, it also includes:

[0171] An area establishing unit, used to establish a search area using the location of the second terminal;

[0172] A region segmentation unit, used to segment the search region using the moving trajectory of the reference object to obtain a plurality of search sub-regions;

[0173] a position determination unit, configured to determine a relative position between the tracked object and the reference object;

[0174] The region selection unit is used to select a search sub-region using the relative position of the tracking object and the reference object, and use the selected search sub-region as a search range.

[0175] Furthermore, it also includes:

[0176] A coordinate assigning unit, used for assigning coordinate points to the tracking object and the suspected tracking object within a spatial range;

[0177] A first movement trajectory processing unit is used to connect the tracking object and the suspected tracking object according to the order of appearance time to obtain a movement trajectory, wherein the movement trajectory includes a plurality of sub-movement trajectories connected in sequence;

[0178] A speed calculation unit, used to calculate the moving speed of the sub-movement trajectory, and use the moving speed to exclude some suspected tracking objects;

[0179] Among them, the continuity of multiple suspected tracking objects is determined with reference to the surrounding environment, and the continuity is assigned to the movement trajectory.

[0180] Furthermore, it also includes:

[0181] A second determination unit, used to determine the surrounding environment characteristics of a suspected tracking object;

[0182] A feature matrix processing unit, used to form a feature matrix using features of a suspected tracking object and surrounding environment and record a moving trajectory of the feature matrix;

[0183] The second movement trajectory processing unit is used to replace the movement trajectory of the suspected tracking object with the movement trajectory of the feature matrix.

[0184] Furthermore, it also includes:

[0185] A first processing unit, configured to transfer the tracking object into a grayscale domain and pick up a first grayscale feature group, wherein the first grayscale feature group includes a pattern and a size;

[0186] A second processing unit is used to transfer the suspected tracking object into the grayscale domain and pick up a second grayscale feature group, wherein the second grayscale feature group includes a pattern and a size;

[0187] A grouping unit, used for putting similar graphics in the first grayscale feature group and the second grayscale feature group into one graphic group;

[0188] A third processing unit, for determining a change vector of the graphic group by means of the size and counting the discreteness of the change vector;

[0189] A second marking unit, used for marking a suspected tracking object whose discreteness is less than or equal to an allowed value as a tracking object;

[0190] Among them, any two objects on the moving trajectory need to be matched, and the objects include tracking objects and suspected tracking objects.

[0191] In one example, the unit in any of the above devices can be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASICs), or, one or more digital signal processors (DSPs), or, one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0192] For another example, when the units in the device can be implemented in the form of a processing element scheduling program, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0193] Various objects such as various messages / information / equipment / network elements / systems / devices / actions / operations / processes / concepts that may appear in this application are named. It can be understood that these specific names do not constitute a limitation on the relevant objects. The names assigned may change with factors such as scenarios, contexts or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from the functions and technical effects embodied / executed in the technical scheme.

[0194] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0195] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0196] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0197] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0198] It should also be understood that in various embodiments of the present application, the first, second, etc. are only used to indicate that multiple objects are different. For example, the first time window and the second time window are only used to indicate different time windows. They should not have any impact on the time window itself, and the first, second, etc. mentioned above should not impose any limitations on the embodiments of the present application.

[0199] It should also be understood that in the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other, and the technical features in different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0200] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a computer-readable storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned computer-readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0201] The present application also provides a cross-terminal target continuous tracking system, the system comprising:

[0202] one or more memories for storing instructions; and

[0203] One or more processors are used to call and run the instructions from the memory to execute the method as described above.

[0204] The present application also provides a computer program product, which includes instructions. When the instructions are executed, the target continuous tracking system (terminal device and network device) performs operations of the target continuous tracking system (terminal device and network device) corresponding to the above method.

[0205] The present application also provides a chip system, which includes a processor for implementing the functions involved in the above content, such as generating, receiving, sending, or processing the data and / or information involved in the above method.

[0206] The chip system may be composed of chips, or may include chips and other discrete devices.

[0207] The processor mentioned in any of the above places can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for executing programs for controlling the above-mentioned feedback information transmission method.

[0208] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wire or wireless means to support the chip system to implement various functions in the above embodiments. Alternatively, the processor and the memory can also be coupled on the same device.

[0209] Optionally, the computer instructions are stored in a memory.

[0210] Optionally, the memory is a storage unit within the chip, such as a register, a cache, etc. The memory can also be a storage unit within the terminal located outside the chip, such as a ROM or other types of static storage devices that can store static information and instructions, RAM, etc.

[0211] It can be understood that the memory in the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0212] The non-volatile memory may be a ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory.

[0213] The volatile memory may be a RAM, which is used as an external cache. There are many different types of RAM, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct memory bus random access memory.

[0214] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for continuous target tracking across terminals. It is characterized in that include: According to the acquired instruction, determining the tracking object in the image data sent back by the first terminal; Establish a feature area with tracking as the center and identify other objects in the feature area to obtain a reference object; Determine the relationship between the reference object and the tracking object; Retrieving the reference object in the image data sent back by the second terminal, and when the reference object exists in the image data sent back by the second terminal, determining the retrieval range according to the position of the second terminal; Search for the tracking object within the search range and find the suspected tracking object; Using the tracking object and the suspected tracking object to generate a movement trajectory of the tracking object and using the movement trajectory to exclude some of the suspected tracking objects; Match the tracking objects and suspected tracking objects that belong to the same moving trajectory; The suspected tracking object that is successfully matched is recorded as the tracking object.

2. The method for continuous target tracking across terminals according to claim 1, It is characterized in that Determining the relationship between the reference object and the tracking object includes: Obtaining the body movements and facial orientation changes of the tracking object and the body movements and facial orientation changes of the reference object; Determine the spatial correlation between the body movements or facial orientation changes of the tracked object and the body movements or facial orientation changes of the reference object and count the number of correlations; When the number is greater than or equal to the allowed number, it is determined that the reference object has an association relationship with the tracked object.

3. The method for continuous target tracking across terminals according to claim 2, It is characterized in that Determining the spatial correlation between the body movement or facial orientation change of the tracked object and the body movement or facial orientation change of the reference object includes: Drawing a body movement trajectory according to the body movement of the object, the object includes a tracking object and a reference object; Drawing a facial orientation change trajectory according to the facial orientation change of the object, the object including the tracking object and the reference object; When the body movement trajectories and / or facial orientation change trajectories of any two subjects match, the number of associations is counted; The matching of the limb movement trajectories and / or the facial orientation change trajectories of any two objects overlaps in the time dimension and the length of the overlapping area is greater than the required length or greater than the required ratio.

4. The method for continuous target tracking across terminals according to any one of claims 1 to 3, It is characterized in that Determining the search scope according to the location of the second terminal includes: Using the location of the second terminal to establish a search area; The search area is segmented using the moving trajectory of the reference object to obtain multiple search sub-areas; According to the determined relative position of the tracking object and the reference object; A search sub-region is selected using the relative positions of the tracking object and the reference object, and the selected search sub-region is used as a search range.

5. The method for continuous target tracking across terminals according to claim 1, It is characterized in that Using mobile tracks to exclude some suspected tracking objects include: Assign coordinate points to the tracked objects and suspected tracked objects within the spatial range; Connecting the tracking object and the suspected tracking object according to the order of appearance time to obtain a movement trajectory, wherein the movement trajectory includes multiple segments of sub-movement trajectories connected in sequence; Calculate the moving speed of the sub-movement trajectory and use the moving speed to exclude some suspected tracking objects; Among them, the continuity of multiple suspected tracking objects is determined with reference to the surrounding environment, and the continuity is assigned to the movement trajectory.

6. The method for continuous target tracking across terminals according to claim 5, It is characterized in that Assigning continuity to a moving trajectory involves: Determine the surrounding environment characteristics of a suspected tracking object; Use the suspected tracking object and the surrounding environment features to form a feature matrix and record the movement trajectory of the feature matrix; The movement trajectory of the feature matrix is ​​used to replace the movement trajectory of the suspected tracking object.

7. The method for continuous target tracking across terminals according to claim 1, 5 or 6, It is characterized in that Tracking objects and suspected tracking objects that match the same moving trajectory include: Transferring the tracking object into the grayscale domain and picking up a first grayscale feature group, the first grayscale feature group including a pattern and a size; Transferring the suspected tracking object into the grayscale domain and picking up a second grayscale feature group, the second grayscale feature group including a pattern and a size; Putting similar graphics in the first grayscale feature group and the second grayscale feature group into a graphic group; Determine the change vector of the graphic group with the help of the size and count the dispersion of the change vector; The suspected tracking object with a discreteness less than or equal to the allowed value is recorded as the tracking object; Among them, any two objects on the moving trajectory need to be matched, and the objects include tracking objects and suspected tracking objects.

8. A cross-terminal target continuous tracking device, It is characterized in that include: A first data acquisition unit, configured to determine a tracking object in the image data sent back by the first terminal according to the acquired instruction; A second data acquisition unit is used to establish a feature area with the tracking as the center and identify other objects in the feature area to obtain a reference object; A first determination unit, used to determine the association relationship between the reference object and the tracking object; A search range processing unit, used to search for a reference object in the image data sent back by the second terminal, and when the reference object exists in the image data sent back by the second terminal, determine the search range according to the position of the second terminal; A third data acquisition unit is used to search for the tracking object within the search range and find the suspected tracking object; A moving trajectory unit, used to generate a moving trajectory of the tracking object using the tracking object and the suspected tracking object and to exclude part of the suspected tracking object using the moving trajectory; A matching unit, used for matching the tracking object and the suspected tracking object belonging to the same moving trajectory; The first marking unit is used to record the suspected tracking object that has been successfully matched as the tracking object.

9. A cross-terminal target continuous tracking system, It is characterized in that The system comprises: one or more memories for storing instructions; and One or more processors, configured to call and execute the instructions from the memory to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, It is characterized in that The computer readable storage medium comprises: The program, when the program is executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images