Method, system, device and medium for identifying the number of queues at battery swap stations

Abstract

The present invention discloses a method, system, device and medium for identifying the number of queues at a battery swap station. The identification method comprises: obtaining an environmental image of a vehicle queue area at a battery swap station; identifying queued vehicles included in the environmental image; and determining the number of queues at the battery swap station according to the identified queued vehicles. The present invention automatically obtains an environmental image of a vehicle queue area at a battery swap station, identifies queued vehicles from the environmental image to determine the number of queues at the battery swap station, and can effectively improve the efficiency of identifying the number of queues, and can also effectively save human resource consumption at the battery swap station and reduce costs.

Description

Technical Field

[0001] The present invention relates to the field of battery swapping technology, and in particular to a method, system, device and medium for identifying the number of queues at a battery swap station. Background Art

[0002] New energy vehicles such as electric vehicles can effectively alleviate air pollution and are also the main direction of future automobile development. Battery swap stations are a place to quickly and efficiently replenish electricity for new energy vehicles. Battery swap stations can not only save new energy vehicles from the requirements of charging stations / piles and other equipment, but also improve equipment utilization and meet the needs of new energy drivers to the greatest extent.

[0003] Battery swap stations can provide electric vehicles with fully automatic battery swap and fast charging services. With the growing demand for electric vehicles, the number of battery swap users has gradually increased, especially taxis. In order to save time and cost, they prefer efficient battery swap services. Due to the limited number of battery swap stations and batteries, battery swap users may need to queue for a certain period of time in the station before they can swap batteries for their electric vehicles.

[0004] At present, the number of vehicles queuing for battery swapping is generally manually counted by employees at battery swap stations, resulting in a waste of human resources, increased costs of the battery swap stations, and low work efficiency. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the defect that the number of queues for battery swapping vehicles in the prior art is manually counted, which leads to a waste of human resources and reduced efficiency, and to provide a method, system, equipment and medium for identifying the number of queues at battery swapping stations.

[0006] The present invention solves the above technical problems through the following technical solutions:

[0007] According to one embodiment of the present invention, a method for identifying the number of queues at a battery swap station is provided, comprising:

[0008] Acquire an environmental image of a vehicle queuing area of ​​a battery swap station;

[0009] identifying queued vehicles included in the environment image;

[0010] The number of queues at the battery swap station is determined based on the identified queued vehicles.

[0011] In this solution, by automatically acquiring the environmental image of the vehicle queuing area of ​​the battery swap station and identifying the queuing vehicles from the environmental image to determine the queue number of the battery swap station, the efficiency of identifying the queue number can be effectively improved, and the human resource consumption of the battery swap station can also be effectively saved, thereby reducing costs.

[0012] Optionally, the identifying the queued vehicles included in the environment image comprises:

[0013] The environment image is input into a queuing vehicle recognition model, and the queuing vehicles in the environment image are recognized by the queuing vehicle recognition model; the queuing vehicle recognition model is trained based on the environment image of the vehicle queuing area of ​​the battery swap station and the corresponding training labels;

[0014] The step of determining the queue number of the battery swap station according to the identified queued vehicles comprises:

[0015] The number of queues at the battery swap station is determined by the recognition result output by the queuing vehicle recognition model.

[0016] In this scheme, historical environmental images are used in advance to train the queuing vehicle recognition model, so that the queuing vehicle recognition model can learn the ability to recognize queuing vehicles based on environmental images. The trained model is then used to recognize queuing vehicles in environmental images, further improving recognition accuracy and efficiency.

[0017] Optionally, the training step of the queuing vehicle recognition model includes:

[0018] Obtain the initialized queuing vehicle recognition model;

[0019] Obtain video streams of the vehicle queuing area of ​​the battery swap station from multiple perspectives;

[0020] Extracting environmental images including the vehicle queuing area from multiple perspectives from the video stream;

[0021] The environment images from multiple perspectives are de-overlapped and merged and used as model training samples;

[0022] Obtaining training labels corresponding to each of the model training samples;

[0023] The queuing vehicle recognition model is trained according to the model training samples and corresponding training labels.

[0024] In this solution, multiple cameras are used to capture video streams from different perspectives to extract environmental images, which are then de-duplicated and merged to be used as model training samples, so that the environmental images can cover the vehicle queuing area as comprehensively as possible, thereby improving the accuracy of model recognition and the stability of the training model.

[0025] Optionally, removing overlaps and merging the environment images from multiple perspectives to use them as model training samples includes:

[0026] De-overlapping and merging the environment images from multiple perspectives to obtain an intermediate image;

[0027] dividing a vehicle queuing area in the intermediate image;

[0028] Blurring the area in the intermediate image except the vehicle queuing area;

[0029] The blurred intermediate image is used as a model training sample.

[0030] In this scheme, a vehicle queuing area is divided in the environmental image, and the non-queuing area in the environmental image is blurred. On the one hand, it can effectively eliminate the interference of other non-battery-swapping vehicles in the invalid area on the identification of battery-swapping vehicles and improve the recognition accuracy of queuing vehicles. On the other hand, it can also avoid the workload of identifying invalid areas and improve the recognition efficiency of queuing vehicles.

[0031] Optionally, training the queuing vehicle recognition model according to the model training samples and corresponding training labels includes:

[0032] Compressing and / or encoding the model training samples and inputting them into the queuing vehicle recognition model;

[0033] Obtaining the predicted recognition result output by the queuing vehicle recognition model;

[0034] The model parameters of the queuing vehicle recognition model are optimized according to the difference between the predicted recognition result and the training labels corresponding to the model training samples.

[0035] In this solution, corresponding image processing is performed on the model training samples to facilitate model training and reduce the amount of transmitted data, thereby improving the stability of model training, effectively saving transmission network bandwidth, reducing costs, and effectively ensuring data security.

[0036] Optionally, the step of updating the queuing vehicle identification model includes:

[0037] When the accuracy of the queuing vehicle identification model is lower than a preset threshold, updating the queuing vehicle identification model;

[0038] or,

[0039] When the usage time of the queuing vehicle identification model is longer than a preset time, the queuing vehicle identification model is updated.

[0040] In this step, when the update conditions are met, the model is updated in a timely manner to follow up on actual needs in real time, thereby further improving the accuracy and efficiency of model recognition.

[0041] Optionally, the training step is performed by a dedicated server; after the training of the queuing vehicle recognition model is completed, the dedicated server sends the queuing vehicle recognition model to each of the battery swap stations and / or the battery swap cloud, so that the battery swap station and / or the battery swap cloud adopts the queuing vehicle recognition model to identify the environmental image and output the queue number.

[0042] In this solution, the trained model is sent to each battery swap station and / or battery swap cloud, so that each battery swap station and / or battery swap cloud can perform model recognition, thereby arranging the model training process and usage process in an orderly manner, effectively saving costs and improving the stability of model recognition.

[0043] According to another embodiment of the present invention, a system for identifying the number of queues at a battery swap station is provided, comprising:

[0044] An environmental image acquisition module is configured to acquire an environmental image of a vehicle queuing area of ​​a battery swap station;

[0045] A queue vehicle recognition module, configured to recognize queue vehicles included in the environment image;

[0046] The queue quantity determination module is configured to determine the queue quantity of the battery swap station according to the identified queued vehicles.

[0047] In this solution, by automatically acquiring the environmental image of the vehicle queuing area of ​​the battery swap station and identifying the queuing vehicles from the environmental image to determine the queue number of the battery swap station, the efficiency of identifying the queue number can be effectively improved, and the human resource consumption of the battery swap station can also be effectively saved, thereby reducing costs.

[0048] Optionally, the queue vehicle identification module is configured to: input the environment image into a queue vehicle identification model, and identify the queue vehicles in the environment image through the queue vehicle identification model; the queue vehicle identification model is trained based on the environment image of the vehicle queue area of ​​the battery swap station and the corresponding training labels;

[0049] The queue quantity determination module is configured to determine the queue quantity of the battery swap station according to the recognition result output by the queue vehicle recognition model.

[0050] In this scheme, historical environmental images are used in advance to train the queuing vehicle recognition model, so that the queuing vehicle recognition model can learn the ability to recognize queuing vehicles based on environmental images. The trained model is then used to recognize queuing vehicles in environmental images, further improving recognition accuracy and efficiency.

[0051] Optionally, the system also includes a model training module;

[0052] The model training module is configured as follows:

[0053] Obtain the initialized queuing vehicle recognition model;

[0054] Obtain video streams of the vehicle queuing area of ​​the battery swap station from multiple perspectives;

[0055] Extracting environmental images including the vehicle queuing area from multiple perspectives from the video stream;

[0056] The environment images from multiple perspectives are de-overlapped and merged and used as model training samples;

[0057] Obtaining training labels corresponding to each of the model training samples;

[0058] The queuing vehicle recognition model is trained according to the model training samples and corresponding training labels.

[0059] In this solution, multiple cameras are used to capture video streams from different perspectives to extract environmental images, which are then de-duplicated and merged to be used as model training samples, so that the environmental images can cover the vehicle queuing area as comprehensively as possible, thereby improving the accuracy of model recognition and the stability of the training model.

[0060] Optionally, the model training module is further configured to:

[0061] De-overlapping and merging the environment images from multiple perspectives to obtain an intermediate image;

[0062] dividing a vehicle queuing area in the intermediate image;

[0063] Blurring the area in the intermediate image except the vehicle queuing area;

[0064] The blurred intermediate image is used as a model training sample.

[0065] In this scheme, a vehicle queuing area is divided in the environmental image, and the non-queuing area in the environmental image is blurred. On the one hand, it can effectively eliminate the interference of other non-battery-swapping vehicles in the invalid area on the identification of battery-swapping vehicles and improve the recognition accuracy of queuing vehicles. On the other hand, it can also avoid the workload of identifying invalid areas and improve the recognition efficiency of queuing vehicles.

[0066] Optionally, the model training module is further configured to:

[0067] Compressing and / or encoding the model training samples and inputting them into the queuing vehicle recognition model;

[0068] Obtaining the predicted recognition result output by the queuing vehicle recognition model;

[0069] The model parameters of the queuing vehicle recognition model are optimized according to the difference between the predicted recognition result and the training labels corresponding to the model training samples.

[0070] In this solution, corresponding image processing is performed on the model training samples to facilitate model training and reduce the amount of transmitted data, thereby improving the stability of model training, effectively saving transmission network bandwidth, reducing costs, and effectively ensuring data security.

[0071] Optionally, the model training module is further configured to:

[0072] When the accuracy of the queuing vehicle identification model is lower than a preset threshold, updating the queuing vehicle identification model;

[0073] or,

[0074] When the usage time of the queuing vehicle identification model is longer than a preset time, the queuing vehicle identification model is updated.

[0075] In this solution, when the update conditions are met, the model is updated in a timely manner to follow up on actual needs in real time, thereby further improving the accuracy and efficiency of model recognition.

[0076] Optionally, the model training module is configured on a dedicated server. After the training of the queuing vehicle recognition model is completed, the dedicated server is configured to send the queuing vehicle recognition model to each battery swap station and / or battery swap cloud, so that the battery swap station and / or battery swap cloud uses the queuing vehicle recognition model to identify the environmental image and output the queue number.

[0077] In this solution, the trained model is sent to each battery swap station and / or battery swap cloud, so that each battery swap station and / or battery swap cloud performs model recognition, thereby arranging the model training process and use process in an orderly manner, effectively saving costs and improving the stability of model recognition. According to another embodiment of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for identifying the number of queues at battery swap stations as described above are implemented.

[0078] According to another embodiment of the present invention, a computer-readable medium is provided, on which computer instructions are stored, and when the computer instructions are executed by a processor, the steps of the method for identifying the number of queues at battery swap stations as described above are implemented.

[0079] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present invention.

[0080] The positive and progressive effects of the present invention are:

[0081] The present invention automatically obtains the environmental image of the vehicle queuing area of ​​the battery swap station to identify the queuing vehicles from the environmental image to determine the queue number of the battery swap station, which can effectively improve the efficiency of identifying the queue number and effectively save the human resource consumption of the battery swap station and reduce costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] The features and advantages of the present invention can be better understood after reading the detailed description of the embodiments of the present disclosure in conjunction with the following drawings. In the drawings, the components are not necessarily drawn to scale, and components with similar related properties or features may have the same or similar reference numerals.

[0083] Figure 1 The figure is a flow chart of a method for identifying the number of queues at a battery swap station according to an embodiment of the present invention.

[0084] Figure 2 The figure is a schematic diagram of the module structure of a system for identifying the number of queues at battery swap stations according to another embodiment of the present invention.

[0085] Figure 3 The present invention is a schematic diagram of the structure of an electronic device for implementing a method for identifying the number of queues at a battery swap station according to another embodiment of the present invention. DETAILED DESCRIPTION

[0086] The present invention is further described below by way of examples, but the present invention is not limited to the scope of the examples.

[0087] In order to overcome the above-mentioned defects that currently exist, this embodiment provides a method for identifying the number of queues at a battery swap station, including: acquiring an environmental image of a vehicle queuing area of ​​the battery swap station; identifying queued vehicles included in the environmental image; and determining the number of queues at the battery swap station based on the identified queued vehicles.

[0088] In this embodiment, the method can be applied to a battery swap station or the cloud, but the application scenario is not specifically limited and can be selected and adjusted accordingly according to actual needs.

[0089] In this embodiment, by automatically acquiring an environmental image of the vehicle queuing area of ​​the battery swap station and identifying the queuing vehicles from the environmental image to determine the number of queues at the battery swap station, the efficiency of identifying the number of queues can be effectively improved, and the human resource consumption of the battery swap station can also be effectively saved, thereby reducing costs.

[0090] Specifically, as an embodiment, Figure 1 As shown, the method for identifying the number of queues at a battery swap station provided in this embodiment mainly includes the following steps:

[0091] Step 101: Acquire an environmental image of a vehicle queuing area of ​​a battery swap station.

[0092] Among them, the vehicle queuing area can be a designated area for vehicle queuing in the battery swap station, or it can be an area formed by natural queuing of vehicles.

[0093] In this step, when the method is applied to a battery swap station, the local server of the battery swap station calls the battery swap station camera to collect the video stream of the vehicle queuing area of ​​the battery swap station in real time, and converts the video stream into an image to capture the environmental image of the vehicle queuing area of ​​the battery swap station.

[0094] When this method is applied to a battery swap station, image recognition can be performed through the local server of the battery swap station, which can reduce bandwidth pressure and cloud computing pressure.

[0095] As a preferred embodiment, in this step, multiple battery swap station cameras at different locations are called to capture real-time video streams of the vehicle queuing area of ​​the battery swap station from multiple perspectives, and the video streams are converted into images to capture environmental images including the vehicle queuing area from multiple perspectives.

[0096] Video streams from different perspectives are collected by multiple cameras to ensure that the environmental image can cover the vehicle queuing area as comprehensively as possible, thereby further improving the accuracy of identifying the number of queues.

[0097] In one feasible method, a camera can be used to change the viewing angle in real time to capture video streams from different viewing angles, so as to ensure that the environmental image can cover the vehicle queuing area as comprehensively as possible, thereby further improving the accuracy of identifying the number of queues.

[0098] In this step, when the method is applied to the cloud, the battery swap station reports the collected video stream to the cloud server, and may also report the captured environmental image of the vehicle queuing area of ​​the battery swap station to the cloud server. The cloud server receives the video stream to obtain the environmental image or directly receives the environmental image.

[0099] When this method is applied to the cloud, image recognition can be performed through the cloud server to facilitate unified management and scheduling of data, which can effectively avoid data tampering and the like.

[0100] Step 102: Input the environment image into the trained queuing vehicle recognition model to output the queuing vehicles in the environment image.

[0101] In this step, the environment image is input into the queuing vehicle recognition model, and the queuing vehicles in the environment image are recognized by the queuing vehicle recognition model. The queuing vehicle recognition model is trained based on the environment image of the vehicle queuing area of ​​the battery swap station and the corresponding training labels.

[0102] The following describes in detail the queuing vehicle recognition model training process.

[0103] First, an initialized queuing vehicle recognition model is obtained. The queuing vehicle recognition model may be a neural network model for image recognition. As long as the vehicle recognition function can be realized, the present embodiment does not specifically limit the model type, and corresponding selection and adjustment may be made according to actual needs.

[0104] Secondly, environmental images including the vehicle queuing area from multiple perspectives captured from video streams shot from multiple perspectives are collected, the environmental images from multiple perspectives are de-duplicated and merged to be used as model training samples, and the corresponding training labels of each model training sample are obtained.

[0105] As a preferred embodiment, the acquired environment image, ie, the model training sample, is further subjected to image processing.

[0106] Specifically, the image processing includes at least one of compression processing, encoding processing and encryption processing. For example, the environment image is compressed, and the compressed image is converted into Base64 (one of the most common encoding methods for transmitting 8-bit bytecodes on the Internet) encoding, and then the Base64 encoding is compressed and encrypted. However, this embodiment does not specifically limit the image processing method, and a corresponding selection can be made according to actual needs.

[0107] Perform corresponding image processing on the model training samples to facilitate model training and reduce the amount of data transmitted, thereby improving the stability of model training, effectively saving transmission network bandwidth, reducing costs, and effectively ensuring data security.

[0108] As another preferred embodiment, the environment images from multiple perspectives are de-overlapped and merged to obtain an intermediate image, a vehicle queuing area is divided in the intermediate image, the area except the vehicle queuing area in the intermediate image is blurred, and the blurred intermediate image is used as a model training sample.

[0109] Specifically, when there are repeated areas between the environment images under multiple perspectives, the environment images are cropped and then merged. The environment images with repeated areas are cropped and then merged to improve the comprehensiveness of model training.

[0110] In an practicable manner, the vehicle queuing area is an area for vehicle queuing that is predetermined by the battery swap station. In this embodiment, the vehicle queuing area can be divided from the environment image based on the template image, or the queuing path in the environment image can be identified to divide the vehicle queuing area.

[0111] In an practicable manner, the vehicle queuing area is an area formed by natural queuing of vehicles. In this embodiment, the vehicle queuing area can be identified based on the positional relationship of the first connected positions of a plurality of vehicles.

[0112] In the above-mentioned feasible method, the non-queue area in the model training sample is blurred, thereby effectively excluding other non-battery-swap vehicles, further improving the accuracy of model recognition.

[0113] Finally, the queuing vehicle recognition model is trained based on the model training samples and corresponding training labels.

[0114] Specifically, the model training samples are input into the queuing vehicle recognition model to obtain the predicted recognition results output by the queuing vehicle recognition model, and the model parameters of the queuing vehicle recognition model are optimized according to the difference between the predicted recognition results and the corresponding training labels of the model training samples.

[0115] As a preferred embodiment, when the accuracy of the queuing vehicle identification model is lower than a preset threshold, the queuing vehicle identification model is updated. The number of queues obtained according to the location information of the vehicles connected to the network or the number of queues manually identified can also be obtained, and the number of queues determined according to the identified queuing vehicles is verified by the number of queues obtained according to the location information of the vehicles connected to the network or the number of queues manually identified to obtain the accuracy of the queuing vehicle identification model. This embodiment does not specifically limit the preset threshold, and can be set accordingly according to actual needs.

[0116] As another preferred embodiment, when the usage time of the queuing vehicle identification model is longer than the preset time, the queuing vehicle identification model is updated. This embodiment does not specifically limit the preset time, and can be set accordingly according to actual needs.

[0117] When the update conditions are met, the model is updated in a timely manner and actual needs are followed up in real time, thereby further improving the accuracy and efficiency of model recognition.

[0118] As a preferred embodiment, the step of training the model is executed by a dedicated server. After the training of the queuing vehicle recognition model is completed, the dedicated server sends the queuing vehicle recognition model to each battery swap station and / or the battery swap cloud, so that the battery swap station and / or the battery swap cloud uses the queuing vehicle recognition model to identify the environmental image and output the queue number.

[0119] The trained model is sent to each battery swap station and / or battery swap cloud, so that each battery swap station and / or battery swap cloud can perform model recognition, thereby arranging the model training process and usage process in an orderly manner, effectively saving costs and improving the stability of model recognition.

[0120] When the vehicle identification model is deployed at a battery swap station, the model deployed on the local server of the battery swap station can be used to quickly and efficiently identify queued vehicles, saving operating costs. In addition, in the event of anomalies on the Internet, vehicle identification can also be achieved through the local LAN, thus ensuring identification stability and reliability.

[0121] Step 103: Determine the queue number of the battery swap station through the recognition result output by the queue vehicle recognition model.

[0122] In this step, the number of vehicles in the queue at the battery swap station is determined through the recognition results after model recognition, and whether the identified vehicles are queued vehicles can be further determined through a preset method. For example, vehicles that have left after battery swapping or vehicles that are currently accurately swapping batteries can be excluded.

[0123] Step 104: Verify the determined queue quantity.

[0124] In this step, the queue number obtained according to the positioning information of the vehicles entering the network is obtained, and the queue number determined according to the identified queue vehicles is verified using the queue number obtained according to the positioning information of the vehicles entering the network.

[0125] Specifically, based on the positioning information of the vehicles entering the network, the number of vehicles in the vehicle queuing area at the current position is determined and used as the queue number, and it is judged whether the queue number is consistent with the queue number output in step 103. If so, the verification is successful. If not, it is prompted that the verification failed, and which data to use as the queue number is selected.

[0126] The number of queues obtained through the positioning information of the vehicles connected to the network is used to verify the number of queues determined by the environmental images, thereby further improving the accuracy of the data.

[0127] As another embodiment, the number of queues obtained by manual statistics can be used to verify the number of queues determined by the environmental image, thereby further improving the accuracy of the data and further selecting which data to use.

[0128] Step 105: recommend a battery swap station based on the verified queue number.

[0129] In this step, the determined queue number is uploaded to the cloud, so that when the cloud receives a request to search for a battery swap station, it can recommend a battery swap station based on the queue number reported by each battery swap station.

[0130] The number of queues determined through environmental images can be used to recommend battery swap stations to users, so that users in queue or waiting to swap batteries can easily choose battery swap stations with fewer queues, thereby saving user time and improving battery swap efficiency.

[0131] In this embodiment, the battery swap waiting time can also be determined based on the identified queue number, and the queue number and / or battery swap waiting time can be output to the display terminal or user terminal of the battery swap station.

[0132] Users in queue can choose a suitable battery swap station for battery swap based on the waiting time and queue length, thereby improving user experience.

[0133] In one practicable manner, when the method is applied to a battery swap station, the battery swap station can upload the queue number of the identified battery swap station to the cloud according to the time period, so that the cloud can recommend the battery swap station to the driver user according to the queue number of the battery swap station when necessary. When the number of queues at the identified battery swap station changes, the battery swap station can upload the new queue number to the cloud, so that the cloud can recommend the battery swap station to the driver user according to the queue number of the battery swap station when necessary.

[0134] The method for identifying the number of queues at a battery swap station provided in this embodiment automatically obtains an environmental image of the vehicle queuing area of ​​the battery swap station to identify the queued vehicles from the environmental image to determine the number of queues at the battery swap station, and further verifies the accuracy of the queue number. At the same time, it can also recommend suitable battery swap stations to users based on the queue number, thereby effectively improving the efficiency of identifying the queue number, and can also effectively save human resource consumption at the battery swap station and reduce costs.

[0135] In order to overcome the above-mentioned defects that currently exist, a system for identifying the number of queues at a battery swap station is also provided, including: an environmental image acquisition module, configured to acquire an environmental image of a vehicle queuing area of ​​a battery swap station; a queuing vehicle identification module, configured to identify queuing vehicles included in the environmental image; and a queue number determination module, configured to determine the number of queues at the battery swap station based on the identified queuing vehicles.

[0136] Specifically, Figure 2 As shown, as another embodiment, the system for identifying the number of queues at battery swap stations provided in this embodiment mainly includes an environmental image acquisition module 21, a queue vehicle identification module 22, a queue number determination module 23, a queue number verification module 24, a battery swap station recommendation module 25 and a model training module 26. The system utilizes the method for identifying the number of queues at battery swap stations as described above.

[0137] When the system is applied to a battery swap station, the environmental image acquisition module 21 is configured to call the battery swap station camera to acquire a video stream of the vehicle queuing area of ​​the battery swap station in real time, and convert the video stream into an image to extract an environmental image of the vehicle queuing area of ​​the battery swap station.

[0138] When the system is applied to battery swap stations, image recognition can be performed through the local server of the battery swap station, which can reduce bandwidth pressure and cloud computing pressure.

[0139] As a preferred embodiment, the environmental image acquisition module 21 is also configured to call multiple battery swap station cameras at different locations to acquire in real time video streams of the vehicle queuing area of ​​the battery swap station from multiple perspectives, and convert the video streams into images to extract environmental images including the vehicle queuing area from multiple perspectives.

[0140] Video streams from different perspectives are collected by multiple cameras to ensure that the environmental image can cover the vehicle queuing area as comprehensively as possible, thereby further improving the accuracy of identifying the number of queues.

[0141] In one feasible method, a camera can be used to change the viewing angle in real time to capture video streams from different viewing angles, so as to ensure that the environmental image can cover the vehicle queuing area as comprehensively as possible, thereby further improving the accuracy of identifying the number of queues.

[0142] When the system is applied to the cloud, the environmental image acquisition module 21 is configured to acquire the video stream reported by the battery swap station, and may also acquire the environmental image of the vehicle queuing area reported by the battery swap station.

[0143] When the system is applied to the cloud, image recognition can be performed through the cloud server to facilitate unified management and scheduling of data, which can effectively avoid data tampering and other situations.

[0144] The queuing vehicle recognition module 22 is configured to input the environmental image into the queuing vehicle recognition model, and recognize the queuing vehicles in the environmental image through the queuing vehicle recognition model. The queuing vehicle recognition model is trained by calling the model training module 26 based on the environmental image of the vehicle queuing area of ​​the battery swap station and the corresponding training labels.

[0145] The following specifically describes the process of calling the model training module 26 to train the queuing vehicle recognition model.

[0146] The model training module 26 is configured to obtain an initialized queuing vehicle recognition model, which may be a neural network model for image recognition. As long as the vehicle recognition function can be realized, this embodiment does not specifically limit the model type, and corresponding selection and adjustment may be made according to actual needs.

[0147] The model training module 26 is also configured to collect environmental images including vehicle queuing areas from multiple perspectives captured from video streams shot from multiple perspectives, de-duplicate and merge the environmental images from multiple perspectives to use them as model training samples, and obtain corresponding training labels for each model training sample.

[0148] As a preferred embodiment, the model training module 26 is further configured to perform further image processing on the acquired environment image, ie, the model training sample.

[0149] Specifically, the image processing includes at least one of compression processing, encoding processing and encryption processing. For example, the environment image is compressed, and the compressed image is converted into Base64 (one of the most common encoding methods for transmitting 8-bit bytecodes on the Internet) encoding, and then the Base64 encoding is compressed and encrypted. However, this embodiment does not specifically limit the image processing method, and a corresponding selection can be made according to actual needs.

[0150] Perform corresponding image processing on the model training samples to facilitate model training and reduce the amount of data transmitted, thereby improving the stability of model training, effectively saving transmission network bandwidth, reducing costs, and effectively ensuring data security.

[0151] As another preferred embodiment, the model training module 26 is further configured to de-overlap and merge the environment images under multiple perspectives to obtain an intermediate image, divide the vehicle queuing area in the intermediate image, blur the area except the vehicle queuing area in the intermediate image, and use the blurred intermediate image as a model training sample. The vehicle queuing area can be a predetermined area for vehicle queuing in the battery swap station, or an area formed by natural queuing of vehicles.

[0152] Specifically, when there are repeated areas between the environment images under multiple perspectives, the environment images are cropped and then merged. The environment images with repeated areas are cropped and then merged to improve the comprehensiveness of model training.

[0153] In an practicable manner, the vehicle queuing area is an area for vehicle queuing that is predetermined by the battery swap station. In this embodiment, the vehicle queuing area can be divided from the environment image based on the template image, or the queuing path in the environment image can be identified to divide the vehicle queuing area.

[0154] In an practicable manner, the vehicle queuing area is an area formed by natural queuing of vehicles. In this embodiment, the vehicle queuing area can be identified based on the positional relationship of the first connected positions of a plurality of vehicles.

[0155] In the above-mentioned feasible method, the non-queue area in the model training sample is blurred, thereby effectively excluding other non-battery-swap vehicles, further improving the accuracy of model recognition.

[0156] The model training module 26 is also configured to train a queuing vehicle recognition model based on model training samples and corresponding training labels.

[0157] Specifically, the model training samples are input into the queuing vehicle recognition model to obtain the predicted recognition results output by the queuing vehicle recognition model, and the model parameters of the queuing vehicle recognition model are optimized according to the difference between the predicted recognition results and the corresponding training labels of the model training samples.

[0158] As a preferred embodiment, the model training module 26 is further configured to update the queue vehicle identification model when the accuracy of the queue vehicle identification model is lower than a preset threshold. The number of queues obtained according to the location information of the vehicles connected to the network or the number of queues manually identified can also be obtained, and the number of queues determined according to the identified queue vehicles is verified by the number of queues obtained according to the location information of the vehicles connected to the network or the number of queues manually identified to obtain the accuracy of the queue vehicle identification model. This embodiment does not specifically limit the preset threshold, and can be set accordingly according to actual needs.

[0159] As another preferred embodiment, the model training module 26 is also configured to update the queuing vehicle identification model when the usage time of the queuing vehicle identification model is longer than a preset time. This embodiment does not specifically limit the preset time and can be set accordingly according to actual needs.

[0160] When the update conditions are met, the model is updated in a timely manner and actual needs are followed up in real time, thereby further improving the accuracy and efficiency of model recognition.

[0161] As a preferred embodiment, the model training module 26 can be configured on a dedicated server. After the training of the queuing vehicle recognition model is completed, the dedicated server will send the queuing vehicle recognition model to each battery swap station and / or battery swap cloud, so that the battery swap station and / or battery swap cloud can use the queuing vehicle recognition model to identify the environmental image and output the queue number.

[0162] The trained model is sent to each battery swap station and / or battery swap cloud, so that each battery swap station and / or battery swap cloud can perform model recognition, thereby arranging the model training process and usage process in an orderly manner, effectively saving costs and improving the stability of model recognition.

[0163] When the vehicle identification model is deployed at a battery swap station, the model deployed on the local server of the battery swap station can be used to quickly and efficiently identify queued vehicles, saving operating costs. In addition, in the event of anomalies on the Internet, vehicle identification can also be achieved through the local LAN, thus ensuring identification stability and reliability.

[0164] The queue quantity determination module 23 is configured to determine the queue quantity of the battery swap station through the recognition results after the vehicle recognition model is identified, and can further determine whether the identified vehicles are queue vehicles in a preset manner. For example, vehicles that have left after battery swapping or vehicles that are currently being accurately swapped can be excluded.

[0165] The queue number verification module 24 is configured to obtain the queue number obtained according to the positioning information of the vehicles entering the network, and use the queue number obtained according to the positioning information of the vehicles entering the network to verify the queue number determined according to the identified queue vehicles.

[0166] Specifically, the queue number verification module 24 is configured to determine the number of vehicles in the vehicle queuing area at the current position according to the positioning information of the vehicles entering the network and use it as the queue number, and judge whether the queue number is consistent with the queue number output by the queue number determination module 23. If so, the verification is successful; if not, it prompts that the verification failed, and selects which data to use as the queue number.

[0167] The number of queues obtained through the positioning information of the vehicles connected to the network is used to verify the number of queues determined by the environmental images, thereby further improving the accuracy of the data.

[0168] As another embodiment, the number of queues obtained by manual statistics can be used to verify the number of queues determined by the environmental image, thereby further improving the accuracy of the data and further selecting which data to use.

[0169] The battery swap station recommendation module 25 is configured to receive the determined queue number, and upon receiving a battery swap station search request, recommend a battery swap station according to the queue number reported by each battery swap station.

[0170] The number of queues determined through environmental images can be used to recommend battery swap stations to users, so that users in queue or waiting to swap batteries can easily choose battery swap stations with fewer queues, thereby saving user time and improving battery swap efficiency.

[0171] In this embodiment, the system can also determine the battery swap waiting time based on the identified queue number, and output the queue number and / or battery swap waiting time to the display terminal or user terminal of the battery swap station.

[0172] Users in queue can choose a suitable battery swap station for battery swap based on the waiting time and queue length, thereby improving user experience.

[0173] In one practicable manner, when the system is applied to a battery swap station, the battery swap station can upload the queue number of the identified battery swap station to the cloud according to the time period, so that the cloud can recommend the battery swap station to the driver user according to the queue number of the battery swap station when necessary. When the number of queues at the identified battery swap station changes, the battery swap station can upload the new queue number to the cloud, so that the cloud can recommend the battery swap station to the driver user according to the queue number of the battery swap station when necessary.

[0174] The system for identifying the number of queues at battery swap stations provided in this embodiment automatically obtains environmental images of the vehicle queuing area of ​​the battery swap station to identify the queued vehicles from the environmental images to determine the number of queues at the battery swap station, and further verifies the accuracy of the queue number. It can also recommend suitable battery swap stations to users based on the queue number, thereby effectively improving the efficiency of identifying the queue number, and can also effectively save human resource consumption at the battery swap station and reduce costs.

[0175] Figure 3 The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for identifying the number of queues at the battery swap station in the above embodiment is implemented. Figure 3 The electronic device 30 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0176] like Figure 3 As shown, the electronic device 30 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including the memory 32 and the processor 31).

[0177] The bus 33 includes a data bus, an address bus, and a control bus.

[0178] The memory 32 may include a volatile memory, such as a random access memory (RAM) 321 and / or a cache memory 322 , and may further include a read-only memory (ROM) 323 .

[0179] The memory 32 may also include a program / utility 325 having a set (at least one) of program modules 324, such program modules 324 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0180] The processor 31 executes various functional applications and data processing by running the computer program stored in the memory 32, such as the method for identifying the number of queues at the battery swap station in the above embodiment of the present invention.

[0181] The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 35. Furthermore, the model generating device 30 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 36. Figure 3 As shown, the network adapter 36 communicates with other modules of the model-generated device 30 via the bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the model-generated device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0182] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0183] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the method for identifying the number of queues at battery swap stations in the above embodiment are implemented.

[0184] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0185] In a possible implementation, the present invention can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to enable the terminal device to execute the steps in the method for identifying the number of queues at battery swap stations in the above embodiment.

[0186] Among them, the program code for executing the present invention can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on the remote device.

[0187] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that this is only for illustration and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims

1. A method for identifying the number of queues at a battery swap station, characterized in that: include: Acquire an environmental image of a vehicle queuing area of ​​a battery swap station; identifying queued vehicles included in the environment image; Determine the queue number of the battery swap station according to the identified queued vehicles; The identifying the queued vehicles included in the environment image comprises: The environment image is input into a queuing vehicle recognition model, and the queuing vehicles in the environment image are recognized by the queuing vehicle recognition model; the queuing vehicle recognition model is trained based on the environment image of the vehicle queuing area of ​​the battery swap station and the corresponding training labels; The step of determining the queue number of the battery swap station according to the identified queued vehicles comprises: Determine the queue number of the battery swap station according to the recognition result output by the queue vehicle recognition model; The training steps of the queuing vehicle recognition model include: Obtain the initialized queuing vehicle recognition model; Obtain video streams of the vehicle queuing area of ​​the battery swap station from multiple perspectives; Extracting environmental images including the vehicle queuing area from multiple perspectives from the video stream; The environment images from multiple perspectives are de-overlapped and merged and used as model training samples; Obtaining training labels corresponding to each of the model training samples; The queuing vehicle recognition model is trained according to the model training samples and corresponding training labels.

2. The method according to claim 1, characterized in that The de-overlapping and merging of the environment images from multiple perspectives to use as model training samples includes: De-overlapping and merging the environment images from multiple perspectives to obtain an intermediate image; dividing a vehicle queuing area in the intermediate image; Blurring the area in the intermediate image except the vehicle queuing area; The blurred intermediate image is used as a model training sample.

3. The method according to claim 1, characterized in that The step of training the queuing vehicle recognition model according to the model training samples and the corresponding training labels includes: Compressing and / or encoding the model training samples and inputting them into the queuing vehicle recognition model; Obtaining the predicted recognition result output by the queuing vehicle recognition model; The model parameters of the queuing vehicle recognition model are optimized according to the difference between the predicted recognition result and the training labels corresponding to the model training samples.

4. The method according to claim 1, characterized in that: The updating step of the queuing vehicle identification model comprises: When the accuracy of the queuing vehicle identification model is lower than a preset threshold, updating the queuing vehicle identification model; or, When the usage time of the queuing vehicle identification model is longer than a preset time, the queuing vehicle identification model is updated.

5. The method according to claim 1, characterized in that The training step is performed by a dedicated server; after the training of the queuing vehicle recognition model is completed, the dedicated server sends the queuing vehicle recognition model to each of the battery swap stations and / or the battery swap cloud, so that the battery swap station and / or the battery swap cloud adopts the queuing vehicle recognition model to identify the environmental image and output the queue number.

6. A system for identifying the number of queues at a battery swap station, characterized in that: include: An environmental image acquisition module is configured to acquire an environmental image of a vehicle queuing area of ​​a battery swap station; A queue vehicle recognition module, configured to recognize queue vehicles included in the environment image; A queue number determination module, configured to determine the queue number of the battery swap station according to the identified queued vehicles; The queuing vehicle identification module is configured as follows: The environment image is input into a queuing vehicle recognition model, and the queuing vehicles in the environment image are recognized by the queuing vehicle recognition model; the queuing vehicle recognition model is trained based on the environment image of the vehicle queuing area of ​​the battery swap station and the corresponding training labels; The recognition system also includes a model training module; The model training module is configured as follows: Obtain the initialized queuing vehicle recognition model; Obtain video streams of the vehicle queuing area of ​​the battery swap station from multiple perspectives; Extracting environmental images including the vehicle queuing area from multiple perspectives from the video stream; The environment images from multiple perspectives are de-overlapped and merged and used as model training samples; Obtaining training labels corresponding to each of the model training samples; The queuing vehicle recognition model is trained according to the model training samples and corresponding training labels.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for identifying the number of queues at a battery swap station as described in any one of claims 1 to 5 are implemented.

8. A computer readable medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, the steps of the method for identifying the number of queues at a battery swap station as described in any one of claims 1 to 5 are implemented.

Images