CN114842410A - Data detection method and device, storage medium and electronic device - Google Patents

Abstract

The embodiment of the invention provides a data detection method, a data detection device, a storage medium and an electronic device, wherein target area data are obtained, the target area data are detected through a detection model, first characteristic information of a target area is obtained, the first characteristic information comprises depth characteristics of all target objects in the target area, multi-target association is carried out on the first characteristic information to establish a target association group, the target association group is detected based on a characteristic database to obtain a detection result, and the characteristic database comprises the characteristic information of the target objects, so that the accuracy of data detection is improved.

Description

A data detection method, device, storage medium and electronic device

technical field

Embodiments of the present invention relate to the field of data processing, and in particular, to a data detection method, device, storage medium, and electronic device.

Background technique

Both motor vehicles and non-motor vehicles are the most commonly used types of transportation. The traffic rules related to motor vehicles have been developed relatively maturely, while the traffic management of non-motor vehicles is still in the early stage of development. Due to the sharp increase in the number of non-motorized vehicles, the management data on non-motorized traffic violations has also grown exponentially, thus placing new requirements on data management.

For example, for the new type of traffic violations such as over-carrying of non-motor vehicles, the current mainly relies on human monitoring to identify the violations, that is, law enforcement officers supervise the intersections where cases often occur, and detect abnormal violations through human eyes. The method is the current mainstream non-motor vehicle violation identification method, which requires a lot of manpower and material resources, and the false alarm rate is also high.

However, if the neural network deep learning is used for supervision, it is limited by the complexity of the actual application scenario, and there are too many false positives and poor practicability.

Therefore, how to effectively improve the validity of data is one of the main problems to be solved in this field.

SUMMARY OF THE INVENTION

According to an embodiment of the present invention, a data detection method is provided, which solves the problem of low data validity in current data management by performing multi-target association on the first feature information to establish a target association group.

According to an embodiment of the present invention, a data detection method is provided. Through classification detection, the data is re-detected to solve the problem of misjudgment of the target object in the target association group.

According to an embodiment of the present invention, a data detection method is provided, wherein the feature database can be continuously updated according to the detection result, and the detection range can be continuously expanded.

According to an embodiment of the present invention, a data detection method is provided, which can identify non-motor vehicle overloading behavior through data detection, replace manual detection, reduce labor and material costs, and improve detection efficiency.

According to an embodiment of the present invention, a data detection method is provided, including:

Obtaining target area data, detecting the target area data through a detection model, and obtaining first feature information of the target area, wherein the first feature information includes the depth features of all target objects in the target area;

performing multi-target association on the first feature information to establish a target association group;

Based on a feature database, the target association group is detected to obtain a detection result, wherein the feature database includes feature information of the target object.

According to an exemplary embodiment of the present invention, performing multi-target association on the first feature information to establish a target association group, including:

Perform multi-target tracking on the first feature information to obtain second feature information, wherein the second feature information includes an identifier and a motion track of the target object;

Perform association processing on the second feature information to obtain the target association group.

According to an exemplary embodiment of the present invention, performing association processing on the second feature information to obtain the target association group, including:

based on the first feature information and the second feature information, setting a target associated area;

Based on the target association area, perform association screening on the target object to obtain the target association group;

The target association group is updated to obtain a new target association group.

According to an exemplary embodiment of the present invention, the target association group is detected based on a feature database, and a detection result is obtained, wherein the feature database includes feature information of the target object, including:

Based on the feature database, the types of the target objects in the target association group are matched to obtain a matching result, where the matching result includes a first matching result and a second matching result;

Based on the matching result, the first matching result is detected with a first reference threshold, and the second matching result is detected with a second reference threshold to obtain a first detection result;

Based on the first detection result, the target association group is classified and detected to obtain a second detection result.

According to an exemplary embodiment of the present invention, it further includes:

Based on the detection results, the feature database is updated.

According to another embodiment of the present invention, a data detection device is provided, comprising:

a first acquisition module, configured to acquire target area data, detect the target area data through a detection model, and obtain first feature information of the target area, wherein the first feature information includes the depth features of all target objects in the target area ;

a target association module, configured to perform multi-target association on the first feature information to establish a target association group;

The detection module detects the target association group based on a feature database, and obtains a detection result, wherein the feature database includes feature information of the target object.

According to an exemplary embodiment of the present invention, the target association module includes:

a multi-target tracking module, configured to perform multi-target tracking on the first feature information to obtain second feature information, wherein the second feature information includes an identifier and a motion trajectory of the target object;

An association processing module, configured to perform association processing on the second feature information to obtain the target association group.

According to an exemplary embodiment of the present invention, the detection module includes:

a matching unit, configured to match the types of the target objects in the target association group based on the feature database to obtain a matching result, where the matching result includes a first matching result and a second matching result;

a first detection unit, configured to detect the first matching result with a first reference threshold based on the matching result, and detect the second matching result with a second reference threshold to obtain a first detection result;

A classification detection unit, configured to perform classification detection on the target association group based on the first detection result to obtain a second detection result.

According to yet another embodiment of the present invention, a computer-readable storage medium is also provided, and a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute any one of the above methods when running steps in the examples.

According to yet another embodiment of the present invention, there is also provided an electronic device comprising a memory and a processor, wherein the memory stores a computer program, the processor is configured to run the computer program to execute any of the above Steps in Method Examples.

Description of drawings

1 is a block diagram of a hardware structure of a mobile terminal according to a data detection method according to an embodiment of the present invention;

2 is a flowchart of a data detection method according to an embodiment of the present invention;

3 is a flowchart of a data detection method according to an embodiment of the present invention;

4 is a flowchart of a data detection method according to an embodiment of the present invention;

5 is a flowchart of a data detection method according to an embodiment of the present invention;

FIG. 6 is a structural block diagram of a data detection apparatus according to an embodiment of the present invention.

Detailed ways

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and in conjunction with the embodiments.

It should be noted that the terms "first", "second" and the like in the description and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence.

The method embodiments provided in the embodiments of this application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Taking running on a mobile terminal as an example, FIG. 1 is a block diagram of a hardware structure of a mobile terminal according to a detection method according to an embodiment of the present invention. As shown in FIG. 1 , the mobile terminal may include one or more (only one is shown in FIG. 1 ) processors 102 (the processors 102 may include but are not limited to processing devices such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned mobile terminal may also include a transmission device 106 and an input and output device 108 for communication functions. Those of ordinary skill in the art can understand that the structure shown in FIG. 1 is only a schematic diagram, which does not limit the structure of the above-mentioned mobile terminal. For example, the mobile terminal may also include more or fewer components than those shown in FIG. 1 , or have a different configuration than that shown in FIG. 1 .

The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as a computer program corresponding to a detection method in the embodiment of the present invention, and the processor 102 executes the computer program stored in the memory 104 by running the computer program. Various functional applications and data processing implement the above method. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory located remotely from the processor 102, and these remote memories may be connected to the mobile terminal through a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

Transmission means 106 are used to receive or transmit data via a network. The specific example of the above-mentioned network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC for short), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (Radio Frequency, RF for short) module, which is used to communicate with the Internet in a wireless manner.

As can be seen from the above background technology, how to effectively detect non-motor vehicle passenger-carrying behavior from traffic data, and reduce misjudgment behavior and human and material cost in the detection process are the main problems that need to be solved at present.

In order to better solve the above technical problems, the present invention discloses a data detection method, device, storage medium and electronic device, which will be described in detail one by one in the following embodiments.

According to an embodiment of the present invention, the main application scenario of the data detection method is to detect whether a non-motor vehicle has a violation according to the detection result, and the violation mainly refers to the non-motor vehicle exceeding the prescribed number of passengers, that is, illegal overloading.

Referring to FIG. 2, FIG. 2 shows a flowchart of a data detection method provided according to an embodiment of the present specification, which specifically includes the following steps:

S202, obtaining target area data, and detecting the target area data through a detection model to obtain first feature information of the target area, wherein the first feature information includes depth features of all target objects in the target area;

According to an embodiment of the present invention, the target area data refers to image data in the target area, wherein the image data includes but is not limited to picture data and video data, and the target area data is transmitted by a data acquisition device by receiving data information and decoded. The data collection device may refer to data collection equipment such as cameras and video cameras. In this embodiment, the image data is mainly obtained by decoding the data information collected by the cameras and the video cameras. The target area can be set by setting the installation position of the data acquisition device according to specific needs.

According to an embodiment of the present invention, wherein the detection model is a deep learning detection algorithm model, such as the YOLOX detection algorithm, the detection model is pre-trained by the ImageNet data set, and then fine-tuned with the non-motor vehicle and human body data in the traffic scene, The final detector outputs the coordinate frame, position, confidence and other information of the two categories of non-motor vehicle and human body. By detecting the target area data through the detection model, all object points of non-motor vehicles and people in the target area data can be detected correspondingly, and the characteristic information of the non-motor vehicles and people corresponding to the object points can be detected and output.

According to an embodiment of the present invention, the non-motor vehicle and human are the target objects to be detected in the target area, wherein the first feature information refers to all the target objects in the target area data, that is, non-motor vehicles. Depth features of motor vehicles and people, the depth features include but are not limited to features such as type, size, location, timestamp, and confidence of non-motor vehicles, as well as features such as size, location, timestamp, and confidence of people.

S204, performing multi-target association on the first feature information to establish a target association group;

Referring to Figure 3, step S204 includes:

S2042, performing multi-target tracking on the first feature information to obtain second feature information, wherein the second feature information includes an identifier and a motion trajectory of the target object;

According to an embodiment of the present invention, the second feature information of the target object is obtained by performing multi-target tracking on the first feature information. That is, according to the depth features of the target objects in the target area, namely the non-motor vehicle and the person, that is, the type, size, location, timestamp and confidence of the non-motor vehicle, as well as the size, location, timestamp and The second feature information of the target object is obtained through multi-target tracking processing, wherein the second feature information includes the IDs and motion trajectories of all the target objects in the target area.

According to an embodiment of the present invention, an IouTrack tracker is used to perform multi-target tracking on the first feature information, so as to assign a unique identifier to the target object, that is, the ID of the target object, and form the target object represented by each ID movement trajectory. In other embodiments provided by the present invention, other multi-target tracking algorithms can also be used to perform multi-target tracking on the first feature information, and the present invention can be applied as long as the same function can be achieved, which is not limited herein.

S2044. Perform association processing on the second feature information to obtain the target association group.

According to an embodiment of the present invention, due to the large occlusion of the human body on the non-motor vehicle and the influence of the installation positions of different data acquisition devices, continuous detection is difficult, and the tracking ID continuity is difficult to ensure, that is, an effective and continuous movement cannot be formed. trajectory. Therefore, in this embodiment, it is necessary to perform correlation processing on the second feature information after the multi-target tracking processing is performed on the first feature information, so as to improve the detection accuracy by obtaining a target correlation group.

According to an embodiment of the present invention, referring to FIG. 3 , step S2044 specifically includes:

a) based on the first feature information and the second feature information, setting a target associated area;

b) performing association screening on the target object based on the target association area to obtain the target association group;

c) Update the target association group to obtain the new target association group.

According to an embodiment of the present invention, the step a) may be specifically implemented as follows: according to the second characteristic information, the ID of the non-motor vehicle is set as the main ID, and the association threshold is set, and the association threshold may refer to a distance threshold, to The main ID is the center, and the target association area is formed according to the association threshold.

According to an embodiment of the present invention, the step b) can be specifically implemented as, with the main ID as the center,

An association condition is set, the target objects in the target association area are filtered according to the association condition, and the target object that meets the association condition is set as a sub-ID corresponding to the main ID.

According to an embodiment of the present invention, the association condition includes a time condition, a trajectory similarity condition, a direction condition, and a speed condition. When the target object in the target association area satisfies all the above-mentioned association conditions, the The target object is set as the sub-ID corresponding to the main ID.

According to an embodiment of the present invention, the association condition may include a trajectory similarity condition, the similarity between the human trajectory and the non-motor vehicle trajectory is calculated through the trajectory similarity matching algorithm model, and a trajectory similarity threshold is set, for example, the trajectory similarity is set The threshold is 90%. Based on the first feature information and the second feature information, that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, the trajectory similarity is calculated. When the When the trajectory similarity is greater than (or greater than or equal to) 90%, the target object satisfies the trajectory similarity condition. When the target object does not meet the trajectory similarity condition, the association screening is not continued. Wherein, the trajectory similarity matching algorithm model may be the Freycher distance algorithm, or may be other algorithms with similar functions, which are not limited herein. It is worth mentioning that the trajectory similarity threshold can be specifically set according to specific situations and needs.

According to an embodiment of the present invention, the association condition may include a time condition, and the time condition may be specifically set as a time threshold, for example, set the time threshold to 10s, based on the first feature information and the second feature information , that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, calculate the duration of the similar trajectory of the human body and non-motor vehicles, if the duration is greater than (or greater than or equal to) 10s, then the target object The time condition is satisfied. If the duration is less than 10s, it is determined that the time condition is not satisfied, and the association screening is not continued. It is worth mentioning that the time threshold can be specifically set according to specific conditions and needs.

According to an embodiment of the present invention, the association condition may include a direction condition, and the direction condition may be specifically set as a movement direction deviation threshold, for example, the movement direction deviation threshold is set to 5%, based on the first feature information and all The second feature information, that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, calculate the movement direction deviation of the similar trajectory of the human body and the non-motor vehicle, if the movement direction deviation is less than (or less than or equal to) 5%, the target object satisfies the direction condition. If the movement direction deviation is greater than 5%, it is determined that the direction condition is not satisfied, and the association screening is not continued. It is worth mentioning that the movement direction deviation threshold can be specifically set according to specific conditions and needs.

According to an embodiment of the present invention, the association condition may include a speed condition, and the speed condition may be specifically set as a movement speed deviation threshold, for example, setting the movement speed deviation threshold to 5%, based on the first feature information and the The second feature information, that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, calculate the movement speed deviation of the similar trajectory of the human body and the non-motor vehicle, if the movement speed deviation is less than (or less than or equal to) 5%, the target object satisfies the speed condition. If the movement speed deviation is greater than 5%, it is determined that the speed condition is not satisfied, and the association screening is not continued. It is worth mentioning that the motion speed deviation threshold can be specifically set according to specific conditions and needs.

In some other embodiments of the present invention, the association conditions may also include other conditions than time conditions, trajectory similarity conditions, direction conditions, and speed conditions, that is, conditions that can be increased or decreased, and the association conditions can be set according to specific needs . The association conditions listed above are not limiting.

In some other embodiments of the present invention, the association condition may only include several conditions among the time condition, the trajectory similarity condition, the direction condition and the speed condition, that is, the condition that can be filtered out. The association conditions can be set according to specific requirements. The association conditions listed above are not limiting.

According to an embodiment of the present invention, with the main ID as the center, when the target object in the target association area satisfies the association condition set when performing association screening, the target object of the association condition is set to correspond to The sub-ID of the main ID, the main ID and the sub-ID form the target association group.

According to an embodiment of the present invention, the step c) may be specifically implemented as performing continuous association screening on the target area. For different sub-IDs that meet the association conditions after association screening, if it is a new sub-ID, extract the depth features of the human body represented by the sub-ID, and match with the human body features of the target association group. If the matching is successful, use The track of the new sub-ID is updated to the corresponding sub-ID in the target association group. If the matching fails, the sub-ID is included in the target association group as a new sub-ID. That is, if it is an old sub-ID, update the motion trajectory of the corresponding sub-ID in the target association group. The above implementation process is to update the target association group to obtain the new target association group.

It is worth mentioning that the sub-ID of the target association group may temporarily disappear in the target area due to occlusion and other reasons. The ID is added to the target association group, otherwise two identical sub-IDs will appear in the target association group, which will have a significant impact on the subsequent overload detection results of non-motor vehicles, and greatly reduce the accuracy of the detection results. Through the step c), this situation can be effectively detected. When the sub-ID appears again, the track of the new sub-ID is updated to the corresponding old sub-ID in the target association group, avoiding the same sub-ID. The ID is added to the target association group for many times, which improves the detection accuracy.

S206: Detect the target association group based on a feature database to obtain a detection result, where the feature database includes feature information of the target object.

According to an embodiment of the present invention, the method of collecting the characteristic information in the characteristic database includes, but is not limited to, the method of collecting characteristic information through road condition data. The way in which characteristic information is collected.

According to an embodiment of the present invention, the feature information mainly refers to information such as basic features of the target object, and the feature information may be image data and information data, frame picture data, or video in a specific time period. The data may also be information data extracted for the target object. The representation forms of the feature information include, but are not limited to, pictures, videos, texts, and audios. The content of the feature information of the target object includes but is not limited to the type, detection information and status information of the non-motor vehicle.

According to an embodiment of the present invention, when the target object refers to a non-motor vehicle, the target object may include various types of non-motor vehicles. The occupancy capacity of different types of non-motor vehicles may be the same or different, so the types of non-motor vehicles are also a prerequisite for overload detection for the target association group. The detection information of the non-motor vehicle mainly includes the standard of the number of people corresponding to different types of non-motor vehicles, that is, the relevant threshold for detecting overloading. The type and detection information of the non-motor vehicle can be set according to specific conditions, which are not limited here.

According to an embodiment of the present invention, the state information of the non-motor vehicle is specifically: for example, the state information of the non-motor vehicle when the non-motor vehicle is not overloaded and the state information of the non-motor vehicle when the non-motor vehicle is overloaded. The number exceeds the number specified in the traffic regulations.

It is worth mentioning that the status information of non-motor vehicles includes the status information of different types of non-motor vehicles, and the status information of non-motor vehicles in different scenarios, wherein the status information includes pictures and videos used in the status scene. , text and audio, etc., and the information of people and vehicles as the target association group.

According to an embodiment of the present invention, the feature database can be updated, including but not limited to online update or offline update, that is, the feature information in the feature database can be increased, decreased or changed.

According to an embodiment of the present invention, referring to FIG. 5 , the step S206 further includes:

S2061, based on the feature database, match the types of the target objects in the target association group to obtain a matching result, where the matching result includes a first matching result and a second matching result;

S2062, based on the matching result, use a first reference threshold to detect the first matching result, and use a second reference threshold to detect the second matching result to obtain a first detection result;

S2063, based on the first detection result, perform classification detection on the target association group to obtain a second detection result.

According to an embodiment of the present invention, the step S2061 may be specifically implemented as: performing type matching on the main ID of the target object of the target association group according to the feature database, that is, matching the specific type of the non-motor vehicle, wherein the non-motor vehicle type is matched. The type of motor vehicle determines the number of people in the motor vehicle, so as to determine whether the non-motor vehicle is overloaded. When the primary ID of the target object of the target association group can be matched with the type of non-motor vehicle in the feature database, it is the first matching result; When the primary ID is a non-motor vehicle type matched to the feature database, it is the second matching result.

According to an embodiment of the present invention, according to step S2062, wherein the first matching result corresponds to a first reference threshold, that is, each type of non-motor vehicle corresponds to a first reference threshold for judging whether the non-motor vehicle is overloaded, the A reference threshold is the maximum number of people carried by the non-motor vehicle within the range specified by the traffic rules; when the matching result is the second matching result, that is, the primary ID of the target object in the target association group does not have If the type of non-motor vehicle is matched to the feature database, there is no corresponding maximum number of people in the non-motor vehicle within the range specified by the traffic rules, and in this case, the default corresponds to the second reference threshold. Preferably, the second reference threshold is 2, where 2 is a numerical value assigned through experience, and the nuclear occupancy of most non-motor vehicles is 2. The second reference threshold may be specifically set, and a preferred value is provided here instead of limiting.

According to an embodiment of the present invention, the first matching result is detected by using a first reference threshold to obtain a first detection result, wherein the first detection result specifically refers to: the first matching result is greater than the first reference threshold, Then, it is judged that the target association group has a high probability of overload behavior, and the first matching result is less than or equal to the first reference threshold, and it is judged that the target association group does not have overload behavior.

According to an embodiment of the present invention, the second matching result is detected by using a second reference threshold to obtain a first detection result, where the first detection result specifically refers to: the second matching result is greater than the second reference threshold, Then, it is judged that the target association group has a high probability of overload behavior, and the second matching result is less than or equal to the second reference threshold, and it is judged that the target association group does not have overload behavior.

According to an embodiment of the present invention, according to the step S2062, based on the first detection result, the target association group with a high overload behavior probability is classified and detected, wherein the classification detection is mainly performed by a classification model , mainly excluded from the target association group, the sub-ID is not the main ID, that is, the driver or passenger of the non-motor vehicle, but other people who have been around the non-motor vehicle for a long time, and excluded from the target In the association group, the sub-ID is not the main ID, that is, the driver or passenger of the non-motor vehicle, but some other misdetected objects.

It is worth mentioning that in step S204, multi-target tracking and association are performed on the first feature information in the 2D plane. Due to the limitation of the 2D plane viewing angle and the first feature information does not include 3D depth information, the The main ID and sub-ID in the target association group have a high probability of false detection, so the target association group needs to be classified and detected, so as to exclude some cases of staying around non-motor vehicles for a long time or driving for a long time. And the sub-ID is not the main ID, that is, the driver or passenger of the non-motor vehicle, but some other misdetected objects (such as goods), etc., so as to improve the accuracy of the detection result.

S208, based on the detection result, update the feature database.

According to an embodiment of the present invention, in step S206, when the type of the target object in the target association group does not match the corresponding type in the feature database, the target object in the target association group can be The feature information of the object is updated into the feature database, thereby expanding the feature information included in the feature database. When the primary ID of the target object, that is, a non-motor vehicle, appears again, the target The behavior of the associated group is effectively detected.

The method according to the above-mentioned embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by means of hardware, but the former is a better implementation in many cases. Based on this understanding, the technical solutions of the present invention essentially or the parts that contribute to the prior art can be embodied in the form of software products, and the computer software products are stored in a storage medium (such as ROM/RAM, magnetic disk, CD-ROM), including several instructions to make a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

In this embodiment, a data detection apparatus is also provided, and the apparatus is used to implement the above-mentioned embodiments and preferred implementations, and what has been described will not be repeated. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, implementations in hardware, or a combination of software and hardware, are also possible and contemplated.

According to another embodiment of the present invention, referring to FIG. 6, a data detection apparatus is provided, including:

The first acquisition module 30 is configured to acquire target area data, detect the target area data through a detection model, and obtain first feature information of the target area, wherein the first feature information includes the depths of all target objects in the target area feature;

According to an embodiment of the present invention, the target area data refers to image data in the target area, wherein the image data includes but not limited to picture data and video data, and the target area data is transmitted by a data acquisition device by receiving data information and decoded. The data collection device may refer to data collection equipment such as cameras and video cameras. In this embodiment, the image data is mainly obtained by decoding the data information collected by the cameras and the video cameras. The target area can be set by setting the installation position of the data acquisition device according to specific needs.

According to an embodiment of the present invention, wherein the detection model is a deep learning detection algorithm model, such as the YOLOX detection algorithm, the detection model is pre-trained by the ImageNet data set, and then fine-tuned with the non-motor vehicle and human body data in the traffic scene, The final detector outputs the coordinate frame, position, confidence and other information of the two categories of non-motor vehicle and human body. By detecting the target area data through the detection model, all object points of non-motor vehicles and people in the target area data can be detected correspondingly, and the characteristic information of the non-motor vehicles and people corresponding to the object points can be detected and output.

According to an embodiment of the present invention, the non-motor vehicle and human are the target objects to be detected in the target area, wherein the first feature information refers to all the target objects in the target area data, that is, non-motor vehicles. Depth features of motor vehicles and people, the depth features include but are not limited to features such as type, size, location, timestamp, and confidence of non-motor vehicles, as well as features such as size, location, timestamp, and confidence of people.

A target association module 40, configured to perform multi-target association on the first feature information to establish a target association group;

The target association module 40 further includes:

A multi-target tracking module 41, configured to perform multi-target tracking on the first feature information to obtain second feature information, wherein the second feature information includes an identifier and a motion trajectory of the target object;

According to an embodiment of the present invention, the second feature information of the target object is obtained by performing multi-target tracking on the first feature information. That is, according to the depth features of the target objects in the target area, namely the non-motor vehicle and the person, that is, the type, size, location, timestamp and confidence of the non-motor vehicle, as well as the size, location, timestamp and The second feature information of the target object is obtained through multi-target tracking processing, wherein the second feature information includes the IDs and motion trajectories of all the target objects in the target area.

According to an embodiment of the present invention, an IouTrack tracker is used to perform multi-target tracking on the first feature information, so as to assign a unique identifier to the target object, that is, the ID of the target object, and form the target object represented by each ID movement trajectory. In other embodiments provided by the present invention, other multi-target tracking algorithms can also be used to perform multi-target tracking on the first feature information, and the present invention can be applied as long as the same function can be achieved, which is not limited herein.

The association processing module 42 is configured to perform association processing on the second feature information to obtain the target association group.

According to an embodiment of the present invention, due to the large occlusion of the human body on the non-motor vehicle and the influence of the installation positions of different data acquisition devices, continuous detection is difficult, and the tracking ID continuity is difficult to ensure, that is, an effective and continuous movement cannot be formed. trajectory. Therefore, in this embodiment, it is necessary to perform correlation processing on the second feature information after the multi-target tracking processing is performed on the first feature information, so as to improve the detection accuracy by obtaining a target correlation group.

According to an embodiment of the present invention, the association processing module 42 specifically includes:

a target associated area setting unit 421, configured to set a target associated area based on the first feature information and the second feature information;

an association screening unit 422, configured to perform association screening on the target object based on the target association area to obtain the target association group;

The target association group updating unit 423 is configured to update the target association group to obtain the new target association group.

According to an embodiment of the present invention, according to the second characteristic information, the ID of the non-motor vehicle is set as the main ID, and a correlation threshold is set, and the correlation threshold may refer to a distance threshold, with the main ID as the center, according to the correlation threshold The target association area is formed.

According to an embodiment of the present invention, an association condition is set centered on the main ID, the target objects in the target association area are screened according to the association condition, and the target objects that meet the association condition are set as corresponding The sub-ID of the main ID.

According to an embodiment of the present invention, the association condition includes a time condition, a trajectory similarity condition, a direction condition, and a speed condition. When the target object in the target association area satisfies all the above-mentioned association conditions, the The target object is set as the sub-ID corresponding to the main ID.

According to an embodiment of the present invention, the association condition may include a track similarity condition, and the similarity between the human body track and the non-motor vehicle track is calculated through a track similarity matching algorithm model, and a track similarity threshold is set, for example, setting the track similarity The threshold is 90%. Based on the first feature information and the second feature information, that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, the trajectory similarity is calculated. When the When the trajectory similarity is greater than (or greater than or equal to) 90%, the target object satisfies the trajectory similarity condition. When the target object does not meet the trajectory similarity condition, the association screening is not continued. Wherein, the trajectory similarity matching algorithm model may be the Frecher distance algorithm, or may be other algorithms with similar functions, which are not limited herein. It is worth mentioning that the trajectory similarity threshold can be specifically set according to specific situations and needs.

According to an embodiment of the present invention, the association condition may include a time condition, and the time condition may be specifically set as a time threshold, for example, set the time threshold to 10s, based on the first feature information and the second feature information , that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, calculate the duration of the similar trajectory of the human body and non-motor vehicles, if the duration is greater than (or greater than or equal to) 10s, then the target object The time condition is satisfied. If the duration is less than 10s, it is determined that the time condition is not satisfied, and the association screening is not continued. It is worth mentioning that the time threshold can be specifically set according to specific conditions and needs.

According to an embodiment of the present invention, the association condition may include a direction condition, and the direction condition may be specifically set as a movement direction deviation threshold, for example, the movement direction deviation threshold is set to 5%, based on the first feature information and all The second feature information, that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, calculate the movement direction deviation of the similar trajectory of the human body and the non-motor vehicle, if the movement direction deviation is less than (or less than or equal to) 5%, the target object satisfies the direction condition. If the movement direction deviation is greater than 5%, it is determined that the direction condition is not satisfied, and the association screening is not continued. It is worth mentioning that the movement direction deviation threshold can be specifically set according to specific conditions and needs.

According to an embodiment of the present invention, the association condition may include a speed condition, and the speed condition may be specifically set as a movement speed deviation threshold, for example, setting the movement speed deviation threshold to 5%, based on the first feature information and the The second feature information, that is, the depth feature of the target object and the ID and corresponding trajectory of the target object, calculate the movement speed deviation of the similar trajectory of the human body and the non-motor vehicle, if the movement speed deviation is less than (or less than or equal to) 5%, the target object satisfies the speed condition. If the movement speed deviation is greater than 5%, it is determined that the speed condition is not satisfied, and the association screening is not continued. It is worth mentioning that the motion speed deviation threshold can be specifically set according to specific conditions and needs.

In some other embodiments of the present invention, the association conditions may also include other conditions than time conditions, trajectory similarity conditions, direction conditions, and speed conditions, that is, conditions that can be increased or decreased, and the association conditions can be set according to specific needs . The association conditions listed above are not limiting.

In some other embodiments of the present invention, the association condition may only include several conditions among the time condition, the trajectory similarity condition, the direction condition and the speed condition, that is, the condition that can be filtered out. The association conditions can be set according to specific requirements. The association conditions listed above are not limiting.

According to an embodiment of the present invention, with the main ID as the center, when the target object in the target association area satisfies the association condition set when performing association screening, the target object of the association condition is set to correspond to The sub-ID of the main ID, the main ID and the sub-ID form the target association group.

According to an embodiment of the present invention, continuous association screening is performed on the target area. For different sub-IDs that meet the association conditions after association screening, if it is a new sub-ID, extract the depth features of the human body represented by the sub-ID, and match with the human body features of the target association group. If the matching is successful, use The track of the new sub-ID is updated to the corresponding sub-ID in the target association group. If the matching fails, the sub-ID is included in the target association group as a new sub-ID. That is, if it is an old sub-ID, update the motion trajectory of the corresponding sub-ID in the target association group. The above implementation process is to update the target association group to obtain the new target association group.

It is worth mentioning that the sub-ID of the target association group may temporarily disappear in the target area due to occlusion and other reasons. The ID is added to the target association group, otherwise two identical sub-IDs will appear in the target association group, which will have a significant impact on the subsequent overload detection results of non-motor vehicles, and greatly reduce the accuracy of the detection results. This situation can be effectively detected by the target association group updating unit 423. When the sub-ID appears again, the track of the new sub-ID is updated to the corresponding old sub-ID in the target association group to avoid The same sub-ID is added to the target association group multiple times, which improves the detection accuracy.

The detection module 50 detects the target association group based on a feature database, and obtains a detection result, wherein the feature database includes feature information of the target object.

According to an embodiment of the present invention, the method of collecting the characteristic information in the characteristic database includes, but is not limited to, the method of collecting characteristic information through road condition data. The way in which characteristic information is collected.

According to an embodiment of the present invention, the feature information mainly refers to information such as basic features of the target object, and the feature information may be image data and information data, frame picture data, or video within a specific time period. The data may also be information data extracted for the target object. The representation forms of the feature information include, but are not limited to, pictures, videos, texts, and audios. The content of the feature information of the target object includes but is not limited to the type, detection information and status information of the non-motor vehicle.

According to an embodiment of the present invention, when the target object refers to a non-motor vehicle, the target object may include various types of non-motor vehicles. The occupancy capacity of different types of non-motor vehicles may be the same or different, so the types of non-motor vehicles are also a prerequisite for the overload detection of the target association group. The detection information of the non-motor vehicle mainly includes the standard of the number of passengers corresponding to different types of non-motor vehicles, that is, the relevant threshold for detecting overloading. The type and detection information of the non-motor vehicle can be set according to specific conditions, which are not limited here.

According to an embodiment of the present invention, the state information of the non-motor vehicle is specifically: for example, the state information of the non-motor vehicle when the non-motor vehicle is not overloaded and the state information of the non-motor vehicle when the non-motor vehicle is overloaded. The number exceeds the number specified in the traffic regulations.

It is worth mentioning that the status information of non-motor vehicles includes the status information of different types of non-motor vehicles, and the status information of non-motor vehicles in different scenarios, wherein the status information includes pictures and videos used in the status scene. , text and audio, etc., and the information of people and vehicles as the target association group.

According to an embodiment of the present invention, the feature database can be updated, including but not limited to online update or offline update, that is, the feature information in the feature database can be increased, decreased or changed.

According to an embodiment of the present invention, the detection module 50 further includes:

A matching unit 51, configured to match the types of the target objects in the target association group based on the feature database to obtain a matching result, where the matching result includes a first matching result and a second matching result;

The first detection unit 52 is configured to, based on the matching result, use a first reference threshold to detect the first matching result, and use a second reference threshold to detect the second matching result to obtain a first detection result;

The classification detection unit 53 is configured to perform classification detection on the target association group based on the first detection result to obtain a second detection result.

According to an embodiment of the present invention, type matching is performed on the main ID of the target object of the target association group according to the feature database, that is, matching the specific type of non-motor vehicle, wherein the type of non-motor vehicle determines the motor vehicle The number of people carried, so as to determine whether the non-motor vehicle is overloaded. When the primary ID of the target object of the target association group can be matched with the type of non-motor vehicle in the feature database, it is the first matching result; When the primary ID is a non-motor vehicle type matched to the feature database, it is the second matching result.

According to an embodiment of the present invention, the first matching result corresponds to a first reference threshold, that is, each type of non-motor vehicle corresponds to a first reference threshold for judging whether the non-motor vehicle is overloaded, and the first reference threshold is in traffic Within the scope specified by the rule, the maximum number of people carrying the non-motor vehicle; when the matching result is the second matching result, that is, the primary ID of the target object in the target association group does not match the feature database The non-motor vehicle type in , there is no corresponding maximum number of people in the non-motor vehicle within the range specified by the traffic rules, which corresponds to the second reference threshold by default. Preferably, the second reference threshold is 2, and 2 is a numerical value assigned by experience, and the nuclear occupancy of most non-motor vehicles is 2. The second reference threshold may be specifically set, and a preferred value is provided here instead of limiting.

According to an embodiment of the present invention, the first matching result is detected by using a first reference threshold to obtain a first detection result, wherein the first detection result specifically refers to: the first matching result is greater than the first reference threshold, Then, it is judged that the target association group has a high probability of overload behavior, and the first matching result is less than or equal to the first reference threshold, and it is judged that the target association group does not have overload behavior.

According to an embodiment of the present invention, the second matching result is detected by using a second reference threshold to obtain a first detection result, where the first detection result specifically refers to: the second matching result is greater than the second reference threshold, Then, it is judged that the target association group has a high probability of overload behavior, and the second matching result is less than or equal to the second reference threshold, and it is judged that the target association group does not have overload behavior.

According to an embodiment of the present invention, based on the first detection result, a classification detection is performed on the target association group with a high overload behavior probability, wherein the classification detection is mainly performed by a classification model, which is mainly excluded from the target. In the association group, the sub-ID is not the main ID, that is, the driver or passenger of the non-motor vehicle, but other people who have been around the non-motor vehicle for a long time, and excluded from the target association group, the sub-ID It is not the main ID, that is, the driver or passenger of a non-motor vehicle, but some other false detections.

It is worth mentioning that the target association module 40 specifically performs multi-target tracking and association on the first feature information in the 2D plane. Due to the limitations of the 2D plane viewing angle and the first feature information does not include 3D depth information, Therefore, the main ID and sub-ID in the target association group have a high probability of false detection, so the target association group needs to be classified and detected, so as to exclude some people who stay around non-motor vehicles for a long time or keep driving for a long time. situation, and the sub-ID is not the main ID, that is, the driver or passenger of a non-motor vehicle, but some other misdetected objects (such as goods), etc., so as to improve the accuracy of the detection results.

The updating module 60, based on the detection result, updates the feature database.

According to an embodiment of the present invention, when the type of the target object in the target association group does not match the corresponding type in the feature database, the feature information of the target object in the target association group can be updated into the feature database, thereby expanding the feature information included in the feature database, and when the primary ID of the target object, that is, a non-motor vehicle, reappears, the behavior of the target association group can be performed. Effective detection.

Embodiments of the present invention further provide a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

In an exemplary embodiment, the above-mentioned computer-readable storage medium may include, but is not limited to, a USB flash drive, a read-only memory (Read-Only Memory, referred to as ROM for short), and a random access memory (Random Access Memory, referred to as RAM for short) , mobile hard disk, magnetic disk or CD-ROM and other media that can store computer programs.

An embodiment of the present invention also provides an electronic device, comprising a memory and a processor, where a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any of the above method embodiments.

In an exemplary embodiment, the above-mentioned electronic device may further include a transmission device and an input-output device, wherein the transmission device is connected to the above-mentioned processor, and the input-output device is connected to the above-mentioned processor.

For specific examples in this embodiment, reference may be made to the examples described in the foregoing embodiments and exemplary implementation manners, and details are not described herein again in this embodiment.

Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device, which can be centralized on a single computing device, or distributed in a network composed of multiple computing devices On the other hand, they can be implemented in program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, can be performed in a different order than shown here. Or the described steps, or they are respectively made into individual integrated circuit modules, or a plurality of modules or steps in them are made into a single integrated circuit module to realize. As such, the present invention is not limited to any particular combination of hardware and software.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. a data detection method, is characterized in that, comprises: Obtaining target area data, detecting the target area data through a detection model, and obtaining first feature information of the target area, wherein the first feature information includes the depth features of all target objects in the target area; performing multi-target association on the first feature information to establish a target association group; Based on a feature database, the target association group is detected to obtain a detection result, wherein the feature database includes feature information of the target object. 2. The method according to claim 1, wherein performing multi-target association on the first feature information to establish a target association group, comprising: Perform multi-target tracking on the first feature information to obtain second feature information, wherein the second feature information includes an identifier and a motion track of the target object; Perform association processing on the second feature information to obtain the target association group. 3. The method according to claim 2, wherein, performing association processing on the second feature information to obtain the target association group, comprising: based on the first feature information and the second feature information, setting a target associated area; Based on the target association area, perform association screening on the target object to obtain the target association group; The target association group is updated to obtain a new target association group. 4. The method according to claim 1, wherein the target association group is detected based on a feature database, and a detection result is obtained, wherein the feature database comprises feature information of the target object, including: Based on the feature database, the types of the target objects in the target association group are matched to obtain a matching result, where the matching result includes a first matching result and a second matching result; Based on the matching result, the first matching result is detected with a first reference threshold, and the second matching result is detected with a second reference threshold to obtain a first detection result; Based on the first detection result, the target association group is classified and detected to obtain a second detection result. 5. The method of claim 1, further comprising: Based on the detection results, the feature database is updated. 6. A data detection device, characterized in that, comprising: a first acquisition module, configured to acquire target area data, detect the target area data through a detection model, and obtain first feature information of the target area, wherein the first feature information includes the depth features of all target objects in the target area ; a target association module, configured to perform multi-target association on the first feature information to establish a target association group; The detection module detects the target association group based on a feature database, and obtains a detection result, wherein the feature database includes feature information of the target object. 7. The apparatus according to claim 6, wherein the target association module comprises: a multi-target tracking module, configured to perform multi-target tracking on the first feature information to obtain second feature information, wherein the second feature information includes an identifier and a motion trajectory of the target object; An association processing module, configured to perform association processing on the second feature information to obtain the target association group. 8. The device according to claim 6, wherein the detection module comprises: a matching unit, configured to match the types of the target objects in the target association group based on the feature database to obtain a matching result, where the matching result includes a first matching result and a second matching result; a first detection unit, configured to detect the first matching result with a first reference threshold based on the matching result, and detect the second matching result with a second reference threshold to obtain a first detection result; A classification detection unit, configured to perform classification detection on the target association group based on the first detection result to obtain a second detection result. 9. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute any one of the claims 1 to 5 when running the method described. 10. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any one of claims 1 to 5 method described in.

Images