Patent Application
20230394848
This application claims the benefit of China Patent Application No. 202210639071.0 filed Jun. 7, 2022, the entire contents of which are incorporated herein by reference in its entirety.
The disclosure relates to the technical field of deep learning, and specifically provides a parking space detection method and device, a vehicle, and a storage medium.
In the design of automatic parking products, a user needs to be informed of available (parkable) parking spaces on an interaction interface, so that the user can make interactive choices. The accuracy of determining the availability of a parking space is required to be high, to minimize a possibility that the user cannot choose a parking space.
In related technologies, an image classification method is usually used to obtain the availability of a parking space. However, during parking space detection, there are some complex scenarios during entrance. For example, the complex scenarios may include the following: A front/rear parking space is blocked. There are many non-standard elements in a parking lot, such as cabbages piled in a parking space, user graffiti, and various charging signs and patterns. There are dynamic or static obstacles of different sizes near the parking space, such as vehicles, pedestrians, cone barrels, and sign boards. In a complex scenario, the availability of a parking space obtained only by using the image classification method is unreliable.
To overcome the above defects, the disclosure is proposed to provide a parking space detection method and device, a vehicle, and a storage medium, to solve or at least partially solve the technical problem that the availability of a parking space obtained only by using an image classification method is unreliable in a complex scenario.
According to a first aspect, the disclosure provides a parking space detection method, including:
According to a second aspect, the disclosure provides a parking space detection device, including at least one processor and a storage apparatus configured to store a plurality of program codes, where the program codes are adapted to be loaded and executed by the at least one processor to perform the parking space detection method according to any one of the above implementations.
According to a third aspect, a vehicle is provided, including the above parking space detection device.
According to a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores a plurality of program codes, where the program codes are adapted to be loaded and executed by at least one processor to perform the parking space detection method according to any one of the above technical solutions.
The above one or more technical solutions of the disclosure have at least one or more of the following beneficial effects:
In implementing the technical solutions of the disclosure, the current frame image that is obtained from the vehicle-mounted camera and that is of the scenario of the vehicle is separately input into the pre-trained parking space detection model, the pre-trained obstacle detection model, and the pre-trained scenario detection model for detection, to separately obtain the parking space prediction result, the obstacle prediction result, and the scenario prediction result. Then, it is determined, based on the detected positional relationship between any target parking space and the vehicle-mounted camera, whether the target parking space is the parking space where the vehicle-mounted camera is located. If it is determined that the target parking space is the parking space where the vehicle-mounted camera is located, verification is performed on the parking space prediction result of the target parking space by using the obstacle prediction result and the scenario prediction result, to obtain the single-frame prediction result of the target parking space. If it is determined that the target parking space is a parking space other than the parking space where the vehicle-mounted camera is located, verification is performed on the parking space prediction result of the target parking space by using the scenario prediction result, to obtain the single-frame prediction result of the target parking space. In this way, after the verification based on a plurality of verification mechanisms, a highly precise parking space detection result can be given in a complex scenario. In addition, a precise prediction result can be given while the vehicle does not need to pass the target parking space completely, which improves a parking space release rate.
The disclosed content of the disclosure will become more readily understood with reference to the accompanying drawings. Those skilled in the art readily understand that these accompanying drawings are merely for illustrative purposes and are not intended to limit the scope of protection of the disclosure. In addition, similar components are represented by similar numbers in the figures, in which:
Some implementations of the disclosure are described below with reference to the accompanying drawings. Those skilled in the art should understand that these implementations are only used to explain the technical principles of the disclosure, and are not intended to limit the scope of protection of the disclosure.
In the description of the disclosure, a “module” or “processor” may include hardware, software, or a combination thereof. A module may include a hardware circuit, various suitable sensors, a communication port, and a memory, or may include a software part, for example, program codes, or may be a combination of software and hardware. The at least one processor may be a central processing unit, a microprocessor, a graphics processing unit, a digital signal processor, or any other suitable processor. The at least one processor has a data and/or signal processing function. The at least one processor may be implemented in software, hardware, or a combination thereof. A non-transitory computer-readable storage medium includes any suitable medium that may store program codes, for example, a magnetic disk, a hard disk, an optical disc, a flash memory, a read-only memory, or a random access memory. The term “A and/or B” indicates all possible combinations of A and B, for example, only A, only B, or A and B. The term “at least one of A or B” or “at least one of A and B” has a meaning similar to “A and/or B” and may include only A, only B, or A and B. The terms “a/an” and “this” in the singular form may also include the plural form.
In automatic parking technologies, an image classification method is usually used to obtain the availability of a parking space. However, during parking space detection, there are some complex scenarios during entrance. For example, the complex scenarios may include the following: A front/rear parking space is blocked. There are many non-standard elements in a parking lot, such as cabbages piled in a parking space, user graffiti, and various charging signs and patterns. There are dynamic or static obstacles of different sizes near the parking space, such as vehicles, pedestrians, cone barrels, and sign boards. In a complex scenario, the availability of a parking space obtained only by using the image classification method is unreliable.
Therefore, to solve the above technical problems, the disclosure provides the following technical solutions.
In step 101, a current frame image of a scenario of a vehicle is obtained from a vehicle-mounted camera.
In a specific implementation process, the vehicle-mounted camera may be installed in the vehicle, and the vehicle-mounted camera may capture the current frame image of the scenario of the vehicle. Exemplarily, the vehicle-mounted camera may be a fisheye camera or any other form of camera, and the vehicle-mounted camera may be installed in corresponding areas of the front, the rear, and both sides of the vehicle body of the vehicle, so that the vehicle-mounted camera can obtain the current frame image of the scenario of the vehicle.
In a specific implementation process, for example, the vehicle-mounted camera is installed on a vehicle ear 11 (a vehicle rearview mirror). A traveling vehicle 1 travels along a road, there are parking spaces on both sides of the road, and scenario images are continuously captured along the road and then are spliced to obtain an aerial view of the scenario of the vehicle, to serve as the current frame image.
In a specific implementation process, different detection models for predicting the availability of parking spaces may be pre-trained, so as to perform a plurality of verifications on the availability of parking spaces by using prediction results of a plurality of detection models.
Specifically, the parking space detection model, the obstacle detection model, and the scenario detection model may be pre-trained, and after the current frame image of the scenario of the vehicle is obtained, the image may be separately input into the pre-trained parking space detection model, the pre-trained obstacle detection model, and the pre-trained scenario detection model for detection, to obtain the parking space prediction result, the obstacle prediction result, and the scenario prediction result.
The parking space prediction result is an initial result obtained based on image classification, the obstacle prediction result is whether there are obstacles such as a started parking lock and a cone barrel in the parking space, and the scenario prediction result is the drivable area in the scenario of the vehicle, display information of the parking space in the current frame image in the current scenario, and the like.
In step 103, it is determined, based on a detected positional relationship between any target parking space and the vehicle-mounted camera, whether the target parking space is a parking space where the vehicle-mounted camera is located. If yes, step 104 is performed, and if no, step 105 is performed.
In a specific implementation process, there may be one or more parking spaces in the current frame image, and for any target parking space, the following operation may be performed:
Specifically, center coordinates of the target parking space and coordinates of the vehicle-mounted camera in the current frame image may be obtained, and then it may be determined, based on a geometric relationship between the center coordinates of the target parking space and the coordinates of the vehicle-mounted camera, whether the target parking space is the parking space where the vehicle-mounted camera is located. Exemplarily, if a deviation angle between the target parking space and the vehicle-mounted camera falls within a preset range, it may be determined that the target parking space is a parking space where the vehicle-mounted camera is located; otherwise, it may be determined that the target parking space is a parking space other than the parking space where the vehicle-mounted camera is located.
It should be noted that the above method of determining, based on the geometric relationship between the center coordinates of the target parking space and the coordinates of the vehicle-mounted camera, whether the target parking space is the parking space where the vehicle-mounted camera is located is only an exemplary method, and other methods are not limited in this embodiment. For example, if the coordinates of the vehicle-mounted camera are located between two corner coordinates of a parking space parallel to the driving road, it may also be determined that the target parking space is the parking space where the vehicle-mounted camera is located; otherwise, it may be determined that the target parking space is a parking space other than the parking space where the vehicle-mounted camera is located.
In a specific implementation process, as shown in
In a specific implementation process, if it is determined that the target parking space is the parking space where the vehicle-mounted camera is located, it means that the vehicle has completely passed the target parking space. In this case, the target parking space is no longer blocked, and obstacle information can be clearly obtained. In this case, verification may be performed on the parking space prediction result of the target parking space by using the obstacle prediction result and the scenario prediction result, to obtain a single-frame prediction result of the target parking space.
Exemplarily, verification may be performed on the parking space prediction result of the target parking space for the first time by using the obstacle prediction result, to obtain the intermediate prediction result of the target parking space, and then verification may be performed on the intermediate prediction result of the target parking space for the second time by using the scenario prediction result, to obtain the single-frame prediction result of the target parking space. Alternatively, verification may be performed on the parking space prediction result of the target parking space for the first time by using the scenario prediction result, to obtain the intermediate prediction result of the target parking space, and then verification may be performed on the intermediate prediction result of the target parking space for the second time by using the obstacle prediction result, to obtain the single-frame prediction result of the target parking space.
In step 105, verification is performed on the parking space prediction result of the target parking space by using the scenario prediction result, to obtain a single-frame prediction result of the target parking space.
In a specific implementation process, if it is determined that the target parking space is a parking space other than the parking space where the vehicle-mounted camera is located, it means that the target parking space may be blocked. In this way, the obstacle information of the target parking space cannot be clearly obtained. Therefore, verification may be performed on the parking space prediction result of the target parking space only by using the scenario prediction result, to obtain the single-frame prediction result of the target parking space.
In a specific implementation process, in this embodiment, the availability of the parking space other than the parking space where the vehicle-mounted camera is located may be predicted. In this way, a precise prediction result can be obtained while the vehicle does not need to completely pass the target parking space, which improves a parking space release rate.
As shown in
It should be noted that, to distinguish parking spaces in different states, other methods may also be used, such as using different colors for distinction, which is not specifically limited in this embodiment.
In the parking space detection method of this embodiment, the current frame image that is obtained from the vehicle-mounted camera and that is of the scenario of the vehicle is separately input into the pre-trained parking space detection model, the pre-trained obstacle detection model, and the pre-trained scenario detection model for detection, to separately obtain the parking space prediction result, the obstacle prediction result, and the scenario prediction result. Then, it is determined, based on the detected positional relationship between any target parking space and the vehicle-mounted camera, whether the target parking space is the parking space where the vehicle-mounted camera is located. If it is determined that the target parking space is the parking space where the vehicle-mounted camera is located, verification is performed on the parking space prediction result of the target parking space by using the obstacle prediction result and the scenario prediction result, to obtain the single-frame prediction result of the target parking space. If it is determined that the target parking space is a parking space other than the parking space where the vehicle-mounted camera is located, verification is performed on the parking space prediction result of the target parking space by using the scenario prediction result, to obtain the single-frame prediction result of the target parking space. In this way, after the verification based on a plurality of verification mechanisms, a highly precise parking space detection result can be given in a complex scenario. In addition, a precise prediction result can be given while the vehicle does not need to pass the target parking space completely, which improves a parking space release rate.
In a specific implementation process, the scenario prediction result may include a drivable area in the scenario of the vehicle.
Step 104 above may specifically include the following steps:
In a specific implementation process, the parking space prediction result of the target parking space may include an available state or an unavailable state. If it is determined that the target parking space is the parking space where the vehicle-mounted camera is located, verification may be first performed on the parking space prediction result of the target parking space by using the obstacle prediction result, to obtain the intermediate prediction result of the target parking space. The intermediate prediction result of the target parking space also includes an available state or an unavailable state.
Specifically, if the obstacle prediction result indicates that there is an obstacle in the target parking space, regardless of the state of the parking space prediction result of the target parking space, the state may be verified as an unavailable state to serve as the intermediate prediction result of the target parking space. If the obstacle prediction result indicates that there is no obstacle in the target parking space, the parking space prediction result of the target parking space is maintained as the intermediate prediction result of the target parking space.
In a specific implementation process, points on the boundary corresponding to the drivable area may be used as road edge points, and points inside the boundary may be used as non-road-edge points. Then, it is detected whether there are non-road-edge points (hereinafter referred to as fs points) of the drivable area in the target parking space, to obtain a detection result, and verification is performed on the parking space prediction result of the target parking space based on the obtained detection result and the intermediate prediction result of the target parking space, so as to obtain the single-frame prediction result of the target parking space.
Specifically, if there is an fs point in the target parking space, it means that there may be an object in the target parking space. In this case, if the intermediate prediction result of the target parking space is an available state, the available state may be verified as an unknown state to serve as the single-frame prediction result of the target parking space.
In a specific implementation process, if there is no fs point in the target parking space, it means that there is no object in the target parking space. In this case, if the intermediate prediction result of the target parking space is an available state, the available state is maintained as the single-frame prediction result of the target parking space.
In a specific implementation process, if there is an fs point in the target parking space, it means that there may be an object in the target parking space. In this case, if the intermediate prediction result of the target parking space is an unavailable state, the unavailable state is maintained as the single-frame prediction result of the target parking space.
In a specific implementation process, if there is no fs point in the target parking space, it means that there may be an object in the target parking space. In this case, if the intermediate prediction result of the target parking space is an unavailable state, the unavailable state is verified as an available state to serve as the single-frame prediction result of the target parking space.
Step 105 above may specifically include the following steps:
In a specific implementation process, the scenario prediction result may include display information of the parking space in the current frame image in the scenario of the vehicle. The display information includes displaying the entire target parking space or displaying a part of the target parking space.
Step 104 above may specifically include the following steps:
This implementation process is similar to the verification process using the drivable area. For details, reference may be made to the above relevant records. Details are not repeated herein.
Step 105 above may specifically include the following steps:
In a specific implementation process, the single-frame prediction result of the target parking space may be shown in Table 1. Table 1 is a verification table of the target parking space and the fs point or the display information of the parking space in the current frame image. The display information of the parking space in the current frame image is represented by the parking space being inside the image or the parking space being outside the image. Exemplarily, the vehicle-mounted camera is installed in a rearview mirror of the vehicle. A parking space where the vehicle-mounted camera is located may be referred to as a vehicle-ear parking space, and a parking space other than a parking space where the vehicle-mounted camera is located may be referred to as a non-vehicle-ear parking space.
In a specific implementation process, when the vehicle is driven, a plurality of frame images are captured at different times, and single-frame prediction results of the target parking space in frames of images may be different. Therefore, to more precisely determine the availability of the target parking space in the current frame image, the disclosure further provides the following technical solutions.
In step 201, a historical frame image having a same parking space identifier as that of the target parking space and a historical single-frame prediction result of the target parking space in the historical frame image are obtained.
In a specific implementation process, after the single-frame prediction result of the target parking space is obtained, the historical frame image having a same parking space identifier as that of the target parking space and the historical single-frame prediction result of the target parking space in the historical frame image may be obtained based on the identifier of the target parking space.
In step 202, it is detected whether the historical frame image includes a frame of target image adjacent to the current frame image in time sequence. If yes, step 203 is performed, and if no, step 205 is performed.
In a specific implementation process, each frame of image has a corresponding time sequence, and it may be detected whether the historical frame image includes a frame of target image adjacent to the current frame image in time sequence. Specifically, if the time sequence of the current frame image is t, a time sequence of a frame of target image adjacent to the time sequence of the current frame image is t−1. If the historical frame image includes an image with a time sequence t−1, it may be determined that there is a frame of target image adjacent to the current frame image in time sequence, and step 203 is performed. If the historical frame image includes no image with a time sequence t−1, it may be determined that there is no frame of target image adjacent to the current frame image in time sequence, and step 205 is performed.
In step 203, it is detected whether a historical single-frame prediction result of the target parking space in the target image is an unknown state. If yes, step 205 is performed, and if no, step 204 is performed.
In a specific implementation process, if the historical frame image includes a frame of target image adjacent to the current frame image in time sequence, it may be detected whether a historical single-frame prediction result of the target parking space in the target image is an unknown state. If yes, step 204 is performed, and if no, step 205 is performed.
In step 204, verification is performed on the single-frame prediction result of the target parking space in the current frame image based on the historical single-frame prediction result of the target parking space in the historical frame image and the positional relationship between the target parking space and the vehicle-mounted camera, to obtain a final prediction result of the target parking space in the current frame image.
In a specific implementation process, if the historical single-frame prediction result of the target parking space in the target image is not an unknown state, verification may be performed on the single-frame prediction result of the target parking space in the current frame image based on the historical single-frame prediction result of the target parking space in the historical frame image and the positional relationship between the target parking space and the vehicle-mounted camera, to obtain a final prediction result of the target parking space in the current frame image.
Specifically, if it is determined, based on the positional relationship between the target parking space and the vehicle-mounted camera, that the target parking space is the parking space where the vehicle-mounted camera is located, and the historical single-frame prediction result of the parking space in the historical frame image includes at least N unavailable states with specified reasons, the unavailable states with specified reasons are used as the final prediction result of the target parking space in the current frame image. If it is determined, based on the positional relationship between the target parking space and the vehicle-mounted camera, that the target parking space is the parking space where the vehicle-mounted camera is located, and/or the single-frame prediction result of the parking space in the historical frame image does not include at least N unavailable states with specified reasons, the historical single-frame prediction result of the target parking space in the target image is used as the final prediction result of the target parking space in the current frame image. N may be 3. The unavailable state with a specified reason may be but is not limited to: being unavailable because a parking lock is started.
In step 205, a state voting result of the target parking space in the current frame image is calculated based on the historical single-frame prediction result of the target parking space in the historical frame image.
In a specific implementation process, if the historical frame image includes a frame of target image adjacent to the current frame image in time sequence, but the historical single-frame prediction result of the target parking space in the target image is not an unknown state, the state voting result of the target parking space in the current frame image may be calculated based on the historical single-frame prediction result of the target parking space in the historical frame image. Alternatively, if the historical frame image includes no frame of target image adjacent to the current frame image in time sequence, the state voting result of the target parking space in the current frame image may be calculated based on the historical single-frame prediction result of the target parking space in the historical frame image.
In a specific implementation process, in calculating the state voting result of the target parking space in the current frame image, a state that has the largest number among states in the historical single-frame prediction result of the target parking space in the historical frame image may be selected as the state voting result of the target parking space in the current frame image. For example, the unavailable state with a specified reason has the largest number among states in the historical single-frame prediction result of the target parking space in the historical frame image, and the state voting result of the target parking space in the current frame image is the unavailable state with a specified reason.
In step 206, verification is performed on the state voting result based on the historical single-frame prediction result of the target parking space in the historical frame image and the positional relationship between the target parking space and the vehicle-mounted camera, to obtain the final prediction result of the target parking space in the current frame image.
In a specific implementation process, after the state voting result of the target parking space in the current frame image is calculated, verification may be further performed on the state voting result based on the historical single-frame prediction result of the target parking space in the historical frame image and the positional relationship between the target parking space and the vehicle-mounted camera, to obtain the final prediction result of the target parking space in the current frame image.
Specifically, if it is determined, based on the positional relationship between the target parking space and the vehicle-mounted camera, that the target parking space is the parking space where the vehicle-mounted camera is located, when a second preset condition is met, an available state is used as the final prediction result of the target parking space in the current frame image. When the second preset condition is not met and a third preset condition is not met, an unknown state is used as the final prediction result of the target parking space in the current frame image. When the second preset condition is not met, but the third preset condition and a fourth preset condition are met, an unavailable state with a specified reason is used as the final prediction result of the target parking space in the current frame image. When the second preset condition and the fourth preset condition are not met, but the third preset condition is met, an unavailable state with a non-specified reason is used as the final prediction result of the target parking space in the current frame image. The unavailable state with a non-specified reason is calculated by voting.
The second preset condition includes that the historical single-frame prediction result of the target parking space in the historical frame image includes at least M available states, and does not include at least P unavailable states with specified reasons; and M may be 3, and P may be 1.
The third preset condition includes that the historical single-frame prediction result of the target parking space in the historical frame image includes at least P unavailable states with specified reasons or includes at least Q unavailable states with non-specified reasons; and Q may be 3.
The fourth preset condition includes that the historical single-frame prediction result of the target parking space in the historical frame image includes at least P unavailable states with specified reasons.
In a specific implementation process, if it is determined, based on the positional relationship between the target parking space and the vehicle-mounted camera, that the target parking space is a parking space other than the parking space where the vehicle-mounted camera is located, it is detected whether the historical single-frame prediction result of the target parking space in the historical frame image includes at least M available states.
If the historical single-frame prediction result of the target parking space in the historical frame image includes at least M available states, an available state is used as the final prediction result of the target parking space in the current frame image.
If the historical single-frame prediction result of the parking space in the historical frame image does not include the at least M available states, an unknown state is used as the final prediction result of the target parking space in the current frame image.
In the parking space detection method of this embodiment, after the single-frame prediction result of the target parking space in the current frame image is obtained, verification may be further performed on the single-frame prediction result of the target parking space based on the single-frame prediction result of the target parking space in the historical frame image, to obtain the final prediction result of the target parking space in the current frame image. In this way, the obtained final prediction result of the target parking space in the current frame image is more precise and reliable.
In a specific implementation process, for a process of performing verification on the single-frame prediction result of the target parking space in the current frame image by using the historical single-frame prediction result of the target parking space in the historical frame image, reference may be made to an example in
As shown in
In step 301, a single-frame prediction result of a current frame is input into a time sequence queue.
The single-frame prediction result of the current frame may be understood as the single-frame prediction result of the target parking space in the current frame image.
In step 302, a historical single-frame prediction result with a same id in the time sequence queue is calculated.
This step is equivalent to: the process of obtaining a historical frame image having a same parking space identifier as that of the target parking space and a historical single-frame prediction result of the target parking space in the historical frame image.
In step 303, it is determined whether there is an adjacent target image in time sequence. If yes, step 308 is performed, and if no, step 304 is performed.
This step is equivalent to: detecting whether the historical frame image includes a frame of target image adjacent to the current frame image in time sequence.
In step 304, it is determined whether the parking space in the target image is in an unknown state. If yes, step 308 is performed, and if no, step 305 is performed.
This step is equivalent to: detecting whether a historical single-frame prediction result of the target parking space in the target image is an unknown state.
In step 305, it is determined whether the target parking space is a vehicle-ear parking space, and a number of frames of images in an unavailable state in the case of the parking lock is ≥3. If yes, step 306 is performed, and if no, step 307 is performed.
This step is equivalent to: if it is determined, based on the positional relationship between the target parking space and the vehicle-mounted camera, that the target parking space is the parking space where the vehicle-mounted camera is located, and the historical single-frame prediction result of the parking space in the historical frame image includes at least N unavailable states with specified reasons.
In step 306, it is the unavailable state in the case of the parking lock.
This step is equivalent to: if it is determined, based on the positional relationship between the target parking space and the vehicle-mounted camera, that the target parking space is the parking space where the vehicle-mounted camera is located, and the historical single-frame prediction result of the parking space in the historical frame image includes at least N unavailable states with specified reasons, using the unavailable states with specified reasons as the final prediction result of the target parking space in the current frame image.
In step 307, the result is consistent with a state of the parking space in the target image.
This step is equivalent to: if it is determined, based on the positional relationship between the target parking space and the vehicle-mounted camera, that the target parking space is the parking space where the vehicle-mounted camera is located, and/or the single-frame prediction result of the parking space in the historical frame image does not include at least N unavailable states with specified reasons, using the historical single-frame prediction result of the target parking space in the target image as the final prediction result of the target parking space in the current frame image.
In step 308, a voting result of the target parking space is calculated.
This step is equivalent to: calculating a state voting result of the target parking space in the current frame image based on the historical single-frame prediction result of the target parking space in the historical frame image.
In step 309, it is determined whether the parking space is a vehicle-ear parking space. If yes, step 310 is performed, and if no, step 317 is performed.
In step 310, a number of time sequence frames of the available state is ≥3, and a number of time sequence frames of the unavailable state in the case of the parking lock is ≤1. If yes, step 311 is performed, and if no, step 312 is performed.
This step is equivalent to the second preset condition.
In step 311, it is the available state.
This step is equivalent to: using an available state as the final prediction result of the target parking space in the current frame image.
In step 312, a number of time sequence frames of another unavailable state is ≥3, or a number of time sequence frames of the unavailable state in the case of the parking lock is ≥1. If yes, step 313 is performed, and if no, step 316 is performed.
This step is equivalent to the third preset condition.
In step 313, a number of time sequence frames of the unavailable state in the case of the parking lock is ≥1. If yes, step 306 is performed, and if no, step 314 is performed.
This step is equivalent to the fourth preset condition.
In step 314, a voting result of another unavailable state is used.
This step is equivalent to: using an unavailable state with a non-specified reason as the final prediction result of the target parking space in the current frame image.
In step 315, it is the unknown state.
This step is equivalent to: using the unknown state as the final prediction result of the target parking space in the current frame image.
In step 316, a number of time sequence frames of the available state is ≥3. If yes, step 311 is performed, and if no, step 315 is performed.
It should be noted that, although the steps are described in a specific order in the above embodiments, those skilled in the art may understand that in order to implement the effects of the disclosure, different steps are not necessarily performed in such an order, but may be performed simultaneously (in parallel) or in other orders, and these changes shall all fall within the scope of protection of the disclosure.
Those skilled in the art can understand that all parking space boxes or some flows in the above method in an embodiment of the disclosure may also be implemented by a computer program instructing relevant hardware. The computer program may be stored in a computer-readable storage medium, and when the computer program is executed by at least one processor, the steps of the above method embodiments may be implemented. The computer program includes computer program codes, which may be in a source code form, an object code form, an executable file form, some intermediate forms, or the like. The computer-readable storage medium may include: any entity or apparatus that can carry the computer program codes, a medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory, a random access memory, an electric carrier signal, a telecommunications signal, and a software distribution medium. It should be noted that the content included in the computer-readable storage medium may be appropriately added or deleted depending on requirements of the legislation and patent practice in a jurisdiction. For example, in some jurisdictions, according to the legislation and patent practice, the computer-readable storage medium does not include an electric carrier signal and a telecommunications signal.
Furthermore, the disclosure further provides a parking space detection device.
Further, the disclosure further provides a vehicle, which may include the parking space detection device according to the above embodiment. The vehicle may be an autonomous vehicle.
Further, the disclosure further provides a computer-readable storage medium. In an embodiment of the computer-readable storage medium according to the disclosure, the computer-readable storage medium may be configured to store a program for performing the parking space detection method of the above method embodiments, and the program may be loaded and executed by at least one processor to implement the above parking space detection method. For ease of description, only parts related to the embodiments of the disclosure are shown. For specific technical details that are not disclosed, reference may be made to the method part of the embodiments of the disclosure. The computer-readable storage medium may be a storage apparatus formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the disclosure is a non-transitory computer-readable storage medium.
Further, it should be understood that since the configuration of the modules is only intended to illustrate the functional units of the apparatus of the disclosure, physical devices corresponding to these modules may be at least one processor itself, or a part of the parking space box in software, or a part of the parking space box in hardware, or a part of the parking space box in a combination of software and hardware in the at least one processor. Therefore, the number of modules in the figure is merely illustrative.
Those skilled in the art can understand that the modules in the apparatus may be adaptively split or merged. Such a split or combination of specific modules does not cause the technical solutions to depart from the principle of the disclosure. Therefore, technical solutions after any such split or combination shall all fall within the scope of protection of the disclosure.
Heretofore, the technical solutions of the disclosure have been described with reference to the preferred implementations shown in the accompanying drawings. However, those skilled in the art can readily understand that the scope of protection of the disclosure is apparently not limited to these specific implementations. Those skilled in the art may make equivalent changes or substitutions to the related technical features without departing from the principle of the disclosure, and all the technical solutions with such changes or substitutions shall fall within the scope of protection of the disclosure.
| Number | Date | Country | Kind |
|---|---|---|---|
| 202210639071.0 | Jun 2022 | CN | national |
