Target state switching method and device, equipment and storage medium

Abstract

The application relates to a target state switching method and device, equipment and a storage medium, wherein the method comprises the following steps: acquiring a first state of a target; acquiring a state switching indication parameter, wherein the state switching indication parameter comprises classification results obtained by classifying the target at least twice; and according to the state switching indication parameter, the first state of the target is switched to a second state or the target is controlled to keep the first state; wherein when the first state is an unlocked classification label state, the second state is a locked classification label state; and when the first state is a locked classification label state, the second state is an unlocked classification label state. The application aims to solve the technical problem of frequent switching of classification labels in the prior art.

Description

Technical Field

[0001] The present application relates to the field of target tracking technology, and in particular to a target state switching method, device, equipment and storage medium. Background Art

[0002] Millimeter-wave radar, a type of radar, boasts high sensitivity, strong penetration, and a non-contact nature. Therefore, it is widely used in various fields. Currently, millimeter-wave radar technology is widely used in the automotive sector. However, its use in home applications and human recognition is still relatively lacking. This will be an important direction for the development of radar technology, truly bringing benefits to human life.

[0003] In the millimeter-wave radar-based personnel detection provided by the existing technology, since human behavior contains many chaotic and irregular movements, it is difficult to identify chaotic and irregular movements when using millimeter-wave radar for personnel recognition, and it is easy to make mistakes, resulting in frequent switching of classification labels. Summary of the Invention

[0004] The present application provides a target state switching method, device, equipment and storage medium to solve the technical problem of frequent switching of classification labels in the prior art.

[0005] In a first aspect, an embodiment of the present application provides a method for switching the state of a target, including:

[0006] Obtaining a current first state of the target;

[0007] Acquiring a state switching indication parameter, wherein the state switching indication parameter includes: classification results obtained by classifying the target at least twice;

[0008] According to the state switching indication parameter, controlling the target to switch from the first state to the second state, or controlling the target to maintain the first state;

[0009] When the first state is an unlocked classification tag state, the second state is a locked classification tag state; when the first state is a locked classification tag state, the second state is an unlocked classification tag state.

[0010] Optionally, controlling the target to switch from the first state to the second state, or controlling the target to maintain the first state, according to the state switching indication parameter, includes:

[0011] If the target is determined to be classified as a target category according to the classification results, the first state of the target is switched to a second state;

[0012] Otherwise, the first state of the target is maintained.

[0013] Optionally, determining that the target is classified as a target category based on the classification results, and switching the first state of the target to the second state includes:

[0014] When it is determined that among the classification results, there are classification results that exceed a preset number of times continuously, indicating that the classification of the target is all the target category, the first state of the target is switched to the second state.

[0015] Optionally, when the first state is an unlocked classification tag state, the second state is a locked classification tag state, and the target category is a movable object category, the state switching indication parameter further includes: a moving distance of the target; the preset number of times is: a first preset number of times;

[0016] The controlling the switching of the first state of the target to the second state according to the state switching indication parameter includes:

[0017] If it is determined that among the classification results, there are classification results that exceed a first preset number of times continuously, indicating that the classification of the target is all the movable object category, and the moving distance of the target exceeds a preset distance, the target is switched from the unlocked classification label state to the locked classification label state.

[0018] Optionally, the determining that, among the classification results, there are classification results that exceed a first preset number of times continuously, indicating that the classification of the target is all the movable object category, and the movement distance of the target exceeds a preset distance, then switching the target from the unlocked classification label state to the locked classification label state includes:

[0019] Counting the first number of times that the target is continuously classified into the movable object category in each classification result;

[0020] Determining whether the first number is greater than the first preset number;

[0021] If yes, obtaining the position of the target in the last classification result of each classification result of the continuous classification into the movable object category;

[0022] calculating a movement distance between the position and an initial position of the target;

[0023] Determining whether the moving distance is greater than the preset distance;

[0024] If yes, the state of the target is switched from the unlocked classification tag state to the locked classification tag state.

[0025] Optionally, when the target category is an immovable object category, the first state is a locked classification label state, and the second state is an unlocked classification label state; the state switching indication parameter further includes: a time duration for continuously classifying the target as the immovable object category for more than a second preset number of times;

[0026] The controlling the switching of the first state of the target to the second state according to the state switching indication parameter includes:

[0027] If it is determined that among the classification results, there are classification results that exceed a second preset number of times continuously, indicating that the classification of the target is all the immovable object category, and the duration is less than the preset duration, the target is switched from the locked classification label state to the unlocked classification label state.

[0028] Optionally, the determining that, among the classification results, there are a second preset number of consecutive classification results indicating that the target is classified as the immovable object category, and the duration is less than a preset duration, then switching the target from the locked classification label state to the unlocked classification label state includes:

[0029] counting a second number of times, in each of the classification results, the target is continuously classified as the immovable object category;

[0030] Determining whether the second number is greater than or equal to a second preset number;

[0031] If yes, obtaining the duration used for continuously classifying the target as the immovable object for the second consecutive time;

[0032] It is determined whether the duration is less than the preset duration; if so, the target is switched from the locked classification tag state to the unlocked classification tag state.

[0033] In a second aspect, an embodiment of the present application provides a target state switching device, comprising:

[0034] A first acquisition module, configured to acquire a current first state of the target;

[0035] A second acquisition module is configured to acquire a state switching indication parameter, wherein the state switching indication parameter includes: classification results obtained by classifying the target at least twice;

[0036] a control module, configured to control the target to switch from the first state to the second state, or to control the target to maintain the first state, according to the state switching indication parameter;

[0037] When the first state is an unlocked classification tag state, the second state is a locked classification tag state; when the first state is a locked classification tag state, the second state is an unlocked classification tag state.

[0038] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, and a communication bus, wherein the processor and the memory communicate with each other via the communication bus;

[0039] The memory is used to store computer programs;

[0040] The processor is used to execute the program stored in the memory to implement the state switching method of the target described in the first aspect.

[0041] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the state switching method for the target described in the first aspect.

[0042] The above-mentioned technical solution provided by the embodiment of the present application has the following advantages over the prior art: the method provided by the embodiment of the present application determines whether to switch the target from the first state to the second state based on the state switching indication parameter, wherein the state switching indication parameter includes the classification results obtained by classifying the target at least twice; instead of directly switching the classification label of the target based on the single classification result of the classifier as in the prior art, it avoids the situation of frequently switching the classification label of the target and can improve the stability and accuracy of target recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0045] Figure 1 A flowchart of a method for switching target states provided in an embodiment of the present application;

[0046] Figure 2 A schematic diagram of the overall detection process of the target state switching method provided in an embodiment of the present application;

[0047] Figure 3 A flowchart of a target lock detection method provided in an embodiment of the present application;

[0048] Figure 4 A flowchart of a target unlocking detection method provided in an embodiment of the present application;

[0049] Figure 5 A schematic diagram of a target state switching device provided in an embodiment of the present application;

[0050] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0052] In the prior art, during the process of target monitoring (for example, target tracking, target detection or identification, and target counting), targets are continuously classified. However, when the target is a person, there is a possibility of misidentification as a non-person because human behaviors contain many chaotic and irregular movements. Moreover, due to performance issues of the classifier, there is also a possibility of misclassifying a person as a non-person, resulting in frequent switching of classification labels, unstable classification results, and the inability to accurately classify the target.

[0053] In order to solve the above technical problems, the embodiment of the present application provides a method for switching the state of a target, such as Figure 1 As shown, the specific steps include:

[0054] Step 101, obtaining the current first state of the target;

[0055] In a specific implementation, the current first state of the target can be obtained during the process of tracking the target, wherein the target can be tracked by radar monitoring or video monitoring.

[0056] Taking radar monitoring as an example, millimeter-wave radar processes the received echo signal to obtain point cloud data. The point cloud data information includes: 1) the distance between the target point and the radar; 2) the azimuth of the target point relative to the radar; 3) the speed of the target point relative to the radar; 4) the signal-to-noise ratio of the echo signal, etc.

[0057] Typically, the point cloud data obtained from the echo signal of a target (e.g., a person) contains multiple target points, that is, multiple target points are described as one target. Typically, multiple point cloud data are clustered to obtain point cloud data of the same class. Point cloud data of the same class belong to the same target. At this point, the target is detected and then tracked. Target tracking can use the extended Kalman filter tracking algorithm, of course, other target tracking algorithms can also be used.

[0058] The target may be a person or other object. The first state may be an unlocked classification label state or a locked classification label state.

[0059] In the embodiment of the present application, locking means fixing the classification label of the target. Because people's movements change frequently, there is a possibility of misidentification as non-human. When the target is judged to be a human, the classification label of the target is locked to prevent subsequent misidentification as non-human. Even if the classifier continues to classify the target, the classification label of the target is not modified; unlocking means not fixing the classification label of the target. If the two classification results are different, the classification label of the target will change in real time according to the classification result of the classifier.

[0060] Step 102: Acquire a state switching indication parameter, wherein the state switching indication parameter includes: classification results obtained by classifying the target at least twice;

[0061] In the process of tracking a target, a pre-trained classifier is used to classify the target, for example, the pre-trained classifier is used to determine whether the target is a person or a non-person.

[0062] The state switching indication parameter includes the above classification results. In addition, the state switching indication parameter may also include: the distance the target moves, the duration of continuous classification into the target category, and the like.

[0063] Step 103: Control the target to switch from the first state to the second state according to the state switching indication parameter, or control the target to maintain the first state;

[0064] When the first state is an unlocked classification tag state, the second state is a locked classification tag state; when the first state is a locked classification tag state, the second state is an unlocked classification tag state.

[0065] In an embodiment of the present application, whether to switch the target from the first state to the second state is determined based on a state switching indication parameter, wherein the state switching indication parameter includes classification results obtained by classifying the target at least twice; instead of directly switching the classification label of the target based on the single classification result of the classifier as in the prior art, the frequent switching of the classification label of the target is avoided, and the stability and accuracy of target recognition can be improved.

[0066] In a specific embodiment, when the state switching indication parameter includes: each classification result obtained by classifying the target at least twice, if the target is determined to be classified as a target category based on each classification result, the first state of the target is switched to the second state; otherwise, the first state of the target is maintained.

[0067] When the first state is the unlocked classification label state and the second state is the locked classification label state, the target category is a movable object category, such as a person or a car. When the target is determined to belong to a movable object category, such as a person, based on the classification results, if the target's current state is the unlocked classification label state, the target's state is switched to the locked classification label state to fix the target's classification label, thereby reducing the problem of frequent classification label switching during continuous target recognition due to recognition errors.

[0068] When the first state is the locked classification label state and the second state is the unlocked classification label state, the target category is an immovable object category, such as a stationary object, such as a tree, a scarecrow, etc. When the target is determined to be an immovable object through each classification result, if the current state of the target is locked, the state of the target is switched to the unlocked classification label state. For a locked target, if it is determined that the classification of the target is wrong based on each classification result, that is, the currently locked classification label is wrong, the locked classification label state of the target is unlocked, the locked classification label state is changed to the unlocked classification label state, and the target is continued to be classified to correct the classification of the target and obtain the correct classification result.

[0069] In a specific embodiment, due to the changes in the target's movements and the influence of the classification performance of the classifier, not all classification results are necessarily classified into the same category. At this time, it can be determined that in each classification result, there are classification results that indicate that the target is classified as the target category for more than a preset number of times, and then the first state of the target is switched to the second state.

[0070] When the first state is an unlocked classification label state and the second state is a locked classification label state, the target category is a movable object category. If the target is classified as a movable object category for more than a preset number of times in a row, the target is switched from the unlocked classification label state to the locked classification label state to fix the classification label of the target, thereby avoiding frequent switching of labels during incorrect identification.

[0071] When the first state is a locked classification label state and the second state is an unlocked classification label state, the target category is an immovable object category. If the target is classified as an immovable object category for more than a preset number of times in a row, the target is switched from the locked classification label state to the unlocked classification label state to correct the classification error.

[0072] In specific implementation, the preset number of times can be set according to actual needs, for example, the preset number of times is set to 2. For ease of understanding, an example is given here. For example, when the target is not locked, if the target is continuously classified for more than the preset number of times, for example, 2 times, and the target is classified as a person, the target is switched from the unlocked classification label state to the locked classification label state. For another example, if the target is locked, if the target is continuously classified for more than the preset number of times, for example, 2 times, and the target is classified as a non-person, the target is switched from the locked classification label state to the unlocked classification label state.

[0073] In a specific embodiment, when the first state is an unlocked classification label state, the second state is a locked classification label state, and the target category is a movable object category, the state switching indication parameter includes, in addition to: classification results obtained by classifying the target at least twice, the movement distance of the target;

[0074] Controlling the switching of the first state of the target to the second state according to the state switching indication parameter includes:

[0075] If it is determined that among the classification results, there are classification results indicating that the classification of the target is a movable object category for more than a first preset number of times, and the moving distance of the target exceeds a preset distance, the target is switched from an unlocked classification label state to a locked classification label state.

[0076] The first preset number and the preset distance can be set according to actual needs, for example, the first preset number is 2, and the preset distance can be 1 meter. The target's movement distance can be calculated using point cloud data. Any method for calculating the distance based on point cloud data can be used, and this is not limited here.

[0077] Specifically, the first number of times the target is continuously classified as a movable object category can be counted in each classification result; it can be determined whether the first number is greater than a first preset number; if so, the position of the target in the last classification result of each classification result in which the target is continuously classified as a movable object category is obtained; the moving distance between the position and the initial position of the target is calculated; it can be determined whether the moving distance is greater than a preset distance; if so, the state of the target is switched from an unlocked classification label state to a locked classification label state.

[0078] The initial position of the target may be the position when the target is first classified as a movable object, for example, the position when the target is classified as a person, or the position when the target is first detected.

[0079] An embodiment of the present application provides a target lock detection process. When a target is not locked with a classification label, whether the target should be locked with a classification label is determined based on the number of consecutive classifications as movable objects and the target's moving distance. This reduces the frequent switching of classification labels, makes the target classification results more accurate, and improves the accuracy of target recognition.

[0080] In a specific embodiment, when the target category is an immovable object category, the first state is a locked classification label state, and the second state is an unlocked classification label state; the state switching indication parameter further includes: the length of time it takes to continuously classify the target as an immovable object category for more than a second preset number of times;

[0081] Controlling the switching of the first state of the target to the second state according to the state switching indication parameter includes:

[0082] If it is determined that in each classification result, there are classification results indicating that the classification of the target is an immovable object category for more than a second preset number of times in a row, and the duration is less than the preset duration, the target is switched from the locked classification label state to the unlocked classification label state.

[0083] The second preset number and the preset duration can be set according to actual needs, for example, the second preset number is 10 or 20 times, and of course other numbers can also be used; the preset duration can be. In a specific implementation, a timer can be used to time the process of continuously classifying an object as an immovable object.

[0084] Specifically, count the second number of times the target is continuously classified as an immovable object in each classification result; determine whether the second number is greater than or equal to a second preset number; if it is 5 seconds, obtain the duration used for the second consecutive number to continuously classify the target as an immovable object; determine whether the duration is less than the preset duration, and if so, switch the target from a locked classification label state to an unlocked classification label state.

[0085] An embodiment of the present application also provides a target unlocking detection process. When a target is locked with a classification label, if there are classification results indicating that the target is classified as an immovable object category for more than a second preset number of times in a row, and the duration of continuous classification as an immovable object category is less than a preset duration, it means that the classification label currently locked on the target is wrong. The target is then switched from a locked classification label state to an unlocked classification label state, and the target is then continued to be judged based on the state switching indication parameters to improve the accuracy of target classification.

[0086] In specific implementation, such as Figure 2 As shown in the figure, for the target obtained by target tracking using point cloud data, each time the classifier classifies it, it is determined whether the target has been locked with a classification label;

[0087] If it has been locked, that is, the current state of the target is the locked classification label state, then the target is unlocked and detected to determine whether the current classification is correct. If it is correct, the classification label state is kept locked. If it is not correct, the classification label of the target is unlocked, the target is tracked, and the target lock detection process is performed;

[0088] On the contrary, if it is not locked, that is, the current state of the target is an unlocked classification label state, then a target locking detection process is performed on the target to determine whether to lock the classification label of the target.

[0089] In addition, it should be noted that in the embodiment of the present application, the locking classification tag is only for targets belonging to the category of movable objects, such as people; the unlocking detection process is only for targets that have locked classification tags, and targets that do not have locked classification tags are not involved in the unlocking detection process.

[0090] In order to facilitate understanding of the target lock detection method provided in the embodiment of the present application, the following Figure 3 The target locking detection method provided in the embodiment of the present application is further described. In this embodiment of the present application, the movable object category is a person.

[0091] like Figure 3 As shown, each classification result performs the following target lock detection method process:

[0092] Step 301, classify the target using a classifier;

[0093] Step 302: determine whether the target has been currently locked to a classification tag; if not, proceed to step 303;

[0094] If it is locked, the lock detection process is no longer performed, and the unlock detection process is performed; if it is not locked, the subsequent process continues;

[0095] In the embodiment of the present application, all processes are executed in a loop, meaning that the entire process is executed once for each classification process, thereby updating the target tracking and classification status in real time. During step 302, a lock identifier is added to each target. If locked, the identifier is set to 1; if not, the identifier is set to 0. Of course, other identifiers are also possible. Whether the target is locked is determined based on the target's lock identifier. If not, the target lock detection process continues. If locked, the following process is not repeated.

[0096] Step 303: Determine whether the target is classified as a person based on the current classification result; if yes, execute steps 304a and 305; if not, execute steps 304b and 305;

[0097] Step 304a, the continuous classification counter is incremented by 1;

[0098] The continuous classification counter is used to indicate the number of times the target is continuously classified as a person;

[0099] Step 304b, the continuous classification counter is reset to zero;

[0100] Step 305, determine whether the continuous classification counter is equal to zero; if so, execute step 306; otherwise, execute steps 307 to 312;

[0101] Step 306: The current position of the target is used as the initial position corresponding to the next classification result, and the maximum distance moved by the target is reset to zero, and then the lock detection process for the next classification result is performed;

[0102] Step 307, calculating the moving distance between the current position of the target and the initial position of the target;

[0103] The initial position can be set to the location of the target when it is first detected, such as the location when the target is first detected by radar or video. It can also be set to the first frame of data when the target is first classified as a person, such as the location of the target indicated in a video or image. For example, if a classification process corresponds to multiple frames of data, such as four frames, and the classification result is a person at the end of the fourth frame, the initial position of the target can be set to the location of the target indicated in the first frame of the four frames.

[0104] Step 308, determine whether the moving distance is greater than the currently stored maximum distance; if so, execute step 309, otherwise, execute steps 310-311;

[0105] Step 309: Update the currently stored maximum distance using the moving distance, and execute steps 310 to 312;

[0106] Step 310, determining whether the continuous classification counter is greater than a first preset number; if yes, executing step 311; otherwise, executing step 313;

[0107] Step 311, determine whether the currently stored maximum distance is greater than the preset distance; if so, execute step 312; otherwise, execute step 313;

[0108] Step 312: Lock the target, determine that the target is classified as a human, mark the target's current state as a locked classification tag state, reset the continuous classification counter to zero, and then perform the unlock detection process;

[0109] In specific implementation, a lock identifier can be set for the target, for example: 1 represents that the target's classification label is locked, and 0 represents that it is unlocked; 1 can also be set to represent that the target's classification label is unlocked, and 0 represents that it is locked; in addition, other types of lock identifiers can also be set, which are not limited here, as long as the locked state and the unlocked state can be distinguished.

[0110] Step 313: Unlock the target, and then perform the lock detection process for the next classification result.

[0111] In an embodiment of the present application, a classifier alone is used to classify people inaccurately, and there are cases of misidentification. According to the results of each classification, it is judged that among the classification results, there are classification results that exceed the first preset number of times continuously, indicating that the classification of the target is a movable object category, for example: a person, and the moving distance of the target exceeds the preset distance, then the target is switched from the unlocked classification label state to the locked classification label state, and the target is identified as a person on the basis of meeting the above-mentioned locking conditions, and the target is locked to prevent subsequent recognition errors when the recognition accuracy is low.

[0112] The following combination Figure 4 The target unlocking detection method provided in the embodiment of the present application is further explained. In this embodiment of the present application, the immovable object category is a non-human category.

[0113] like Figure 4 As shown, each classification result performs the following target unlocking detection method process:

[0114] Step 401, classify the target using a classifier;

[0115] Step 402: determine whether the target has been currently locked to a classification tag; if so, proceed to step 403;

[0116] For targets without locked classification labels, the unlocking detection process is not involved.

[0117] Step 403, determine whether the current unlock classification counter is equal to zero; if yes, execute steps 404 and 405; if no, execute steps 413 and 414;

[0118] The unlock classification counter is used to indicate the number of times the target is continuously classified as an immovable object.

[0119] Step 404: The unlock timer is set to 1, i.e., no time is counted;

[0120] Step 405, determining whether the classification of the target indicated by the current classification result is human; if yes, executing step 406; otherwise, executing step 409;

[0121] Step 406, determine whether the unlock classification counter is zero; if so, execute steps 407 and 410; otherwise, execute steps 408 and 410;

[0122] Step 407, keeping the first consecutive number of times zero;

[0123] Step 408, decrement the unlock category counter by 1;

[0124] Step 409, increment the unlock category counter by 1;

[0125] Step 410, determining whether the unlock classification counter is greater than a second preset number, if yes, executing step 411; otherwise, executing step 412;

[0126] Step 411: The target's lock category tag status is changed to unlocked, and the unlock category counter is reset to zero; then the lock detection process is repeated for the target.

[0127] Step 412: Maintain the locked classification tag status of the target and perform the unlock detection process of the next latest classification result;

[0128] Step 413, the unlocking timer starts timing;

[0129] Step 414, determine whether the timing duration exceeds the preset duration; if yes, proceed to step 415; otherwise, execute step 405;

[0130] Among them, the timed duration is the duration of continuous classification as non-human use;

[0131] In step 415 , the unlock classification counter is reset to zero, and step 405 is executed.

[0132] When unlocking, it is mainly judged by the unlock classification counter. If the classifier classifies it as non-human, the unlock classification counter will increase by 1. If the unlock classification counter is greater than the unlock threshold, that is, the second preset number of times, the target is unlocked, and the target classification label is re-determined according to the classification result of the classifier.

[0133] The unlock timer counts the unlock category counters. When the unlock category counter is greater than 0, the unlock timer starts counting until the unlock threshold (the preset duration) is reached. During this preset duration, it determines whether the locked target can be unlocked. This means that unlocking is time-limited and must be determined within a certain timeframe. The unlock timer records this timeframe.

[0134] In the embodiments of the present application, the unlock detection method is only used for locked targets. In actual applications, target lock detection may sometimes result in misjudgments. For example, if a target's actual classification label should be non-human, but it meets the conditions and is locked during target lock detection, it is mistakenly identified as a human. If the unlock is not performed, the non-human target will continue to be identified as a human after being locked, and the mistake will continue and cannot be corrected. Therefore, the embodiments of the present application provide a method for unlocking a target, which unlocks the target when the above unlock conditions are met, to solve the problem of misjudgment of lock.

[0135] Based on the same concept, an embodiment of the present application provides a state switching device for a target. The specific implementation of the device can be found in the description of the method embodiment part, and the repeated parts will not be repeated. Figure 5 As shown, the device mainly includes:

[0136] A first acquisition module 501 is used to acquire a current first state of a target;

[0137] The second acquisition module 502 is configured to acquire a state switching indication parameter, wherein the state switching indication parameter includes: classification results obtained by classifying the target at least twice;

[0138] A control module 503 is configured to control the target to switch from the first state to the second state according to the state switching indication parameter, or to control the target to maintain the first state;

[0139] When the first state is an unlocked classification tag state, the second state is a locked classification tag state; when the first state is a locked classification tag state, the second state is an unlocked classification tag state.

[0140] In a specific embodiment, the control module 503 is configured to switch the first state of the target to the second state if it is determined that the target is classified as a target category according to the classification results; otherwise, maintain the first state of the target.

[0141] In a specific embodiment, the control module 503 is configured to switch the first state of the target to the second state when it is determined that, among the classification results, there are classification results that exceed a preset number of times continuously indicating that the classification of the target is all the target category.

[0142] In a specific embodiment, the control module 503 is used to determine that among the classification results, there are classification results that exceed a first preset number of times continuously, indicating that the classification of the target is the movable object category, and the moving distance of the target exceeds the preset distance, and then switch the target from the unlocked classification label state to the locked classification label state.

[0143] In a specific embodiment, the control module 503 is used to count the first number of times that the target is continuously classified as the movable object category in each classification result; determine whether the first number is greater than the first preset number; if so, obtain the position of the target in the last classification result of each classification result in which the target is continuously classified as the movable object category; calculate the moving distance between the position and the initial position of the target; determine whether the moving distance is greater than the preset distance; if so, switch the state of the target from the unlocked classification label state to the locked classification label state.

[0144] In a specific embodiment, the control module 503 is used to determine that among the classification results, there are classification results that exceed a second preset number of times continuously, indicating that the classification of the target is the immovable object category, and the duration is less than the preset duration, then the target is switched from the locked classification label state to the unlocked classification label state.

[0145] In a specific embodiment, the control module 503 is used to count the second number of times that the target is continuously classified as the immovable object category in each classification result; determine whether the second number is greater than or equal to a second preset number; if so, obtain the time duration used for continuously classifying the target as the immovable object for the second consecutive time; determine whether the time duration is less than the preset time duration, and if so, switch the target from the locked classification label state to the unlocked classification label state.

[0146] Based on the same concept, an electronic device is also provided in the embodiment of the present application, such as Figure 6 As shown, the electronic device mainly includes: a processor 601, a memory 602 and a communication bus 603, wherein the processor 601 and the memory 602 communicate with each other via the communication bus 603. The memory 602 stores a program that can be executed by the processor 601, and the processor 601 executes the program stored in the memory 602 to implement the following steps:

[0147] Get the current first state of the target;

[0148] Acquiring a state switching indication parameter, wherein the state switching indication parameter includes: classification results obtained by classifying the target at least twice;

[0149] According to the state switching indication parameter, controlling the target to switch from the first state to the second state, or controlling the target to maintain the first state;

[0150] When the first state is an unlocked classification tag state, the second state is a locked classification tag state; when the first state is a locked classification tag state, the second state is an unlocked classification tag state.

[0151] The communication bus 603 mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus 603 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0152] The memory 602 may include a random access memory (RAM) or a non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor 601.

[0153] The above-mentioned processor 601 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc., and can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0154] In another embodiment of the present application, a computer-readable storage medium is provided, which stores a computer program. When the computer program runs on a computer, the computer executes the state switching method of a target described in the above embodiment.

[0155] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instruction is loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instruction can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instruction is transmitted from a website, computer, server or data center to another website, computer, server or data center by wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (such as a floppy disk, hard disk, tape, etc.), an optical medium (such as a DVD) or a semiconductor medium (such as a solid-state hard disk), etc.

[0156] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0157] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is intended to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for switching the state of a target, characterized in that: include: Obtaining a current first state of the target; Acquiring a state switching indication parameter, wherein the state switching indication parameter includes: classification results obtained by classifying the target at least twice; According to the state switching indication parameter, controlling the target to switch from the first state to the second state, or controlling the target to maintain the first state, wherein, when it is determined that among the classification results, there are classification results that exceed a preset number of times continuously, indicating that the classification of the target is all the target category, then switching the first state of the target to the second state; Wherein, when the first state is an unlocked classification tag state, the second state is a locked classification tag state; when the first state is a locked classification tag state, the second state is an unlocked classification tag state, the locked classification tag state means that the classification tag of the target is fixed, and the unlocked classification tag state means that the classification tag of the target is not fixed; Among them, when the first state is an unlocked classification tag state and the second state is a locked classification tag state, according to the state switching indication parameter, the first state of the target is controlled to be switched to the second state, including: determining that among the classification results, there are classification results that exceed the first preset number of times continuously, indicating that the classification of the target is a movable object category, and the moving distance of the target exceeds the preset distance, then the target is switched from the unlocked classification tag state to the locked classification tag state.

2. The target state switching method according to claim 1, characterized in that: The determining that, among the classification results, there are classification results that exceed a first preset number of times continuously, indicating that the classification of the target is all the movable object category, and the movement distance of the target exceeds a preset distance, then switching the target from the unlocked classification label state to the locked classification label state includes: Counting a first number of times that the target is continuously classified as the movable object category in each classification result; Determining whether the first number is greater than the first preset number; If yes, obtaining the position of the target in the last classification result of each classification result in which the target is continuously classified into the movable object category; calculating a movement distance between the position and an initial position of the target; Determining whether the moving distance is greater than the preset distance; If yes, the state of the target is switched from the unlocked classification tag state to the locked classification tag state.

3. The target state switching method according to claim 1, characterized in that: When the target category is an immovable object category, the first state is a locked classification label state, and the second state is an unlocked classification label state, the state switching indication parameter further includes: a time duration for continuously classifying the target as the immovable object category for more than a second preset number of times; The controlling the switching of the first state of the target to the second state according to the state switching indication parameter includes: If it is determined that among the classification results, there are classification results that exceed a second preset number of times continuously, indicating that the classification of the target is all the immovable object category, and the duration is less than the preset duration, the target is switched from the locked classification label state to the unlocked classification label state.

4. The target state switching method according to claim 3, characterized in that: The determining that, among the classification results, there are a second preset number of consecutive classification results indicating that the classification of the target is the immovable object category, and the duration is less than a preset duration, then switching the target from the locked classification label state to the unlocked classification label state includes: counting a second number of times, in each of the classification results, the target is continuously classified as the immovable object category; Determining whether the second number is greater than or equal to a second preset number; If yes, then obtaining the duration used for continuously classifying the target as the immovable object category; It is determined whether the duration is less than the preset duration; if so, the target is switched from the locked classification tag state to the unlocked classification tag state.

5. A target state switching device, characterized in that: include: A first acquisition module, configured to acquire a current first state of the target; A second acquisition module is configured to acquire a state switching indication parameter, wherein the state switching indication parameter includes: classification results obtained by classifying the target at least twice; a control module, configured to control the target to switch from the first state to the second state, or to control the target to maintain the first state, according to the state switching indication parameter, wherein, when it is determined that among the classification results, there are classification results that exceed a preset number of times continuously, indicating that the classification of the target is all the target category, the first state of the target is switched to the second state; Wherein, when the first state is an unlocked classification tag state, the second state is a locked classification tag state; when the first state is a locked classification tag state, the second state is an unlocked classification tag state, the locked classification tag state means that the classification tag of the target is fixed, and the unlocked classification tag state means that the classification tag of the target is not fixed; Among them, when the first state is an unlocked classification tag state and the second state is a locked classification tag state, according to the state switching indication parameter, the first state of the target is controlled to be switched to the second state, including: determining that among the classification results, there are classification results that exceed the first preset number of times continuously, indicating that the classification of the target is a movable object category, and the moving distance of the target exceeds the preset distance, then the target is switched from the unlocked classification tag state to the locked classification tag state.

6. An electronic device, characterized in that: include: A processor, a memory, and a communication bus, wherein the processor and the memory communicate with each other via the communication bus; The memory is used to store computer programs; The processor is configured to execute the program stored in the memory to implement the target state switching method according to any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the state switching method of any one of claims 1 to 4 is implemented.

Images