Method, device and storage medium for detecting drivable area of vehicle

Abstract

The application discloses a kind of detection method, device and storage medium of vehicle drivable area, the method comprises: obtaining the image to be processed of the preset area corresponding to target vehicle;The distance feature and boundary point category feature of boundary point are obtained by extracting features to the image to be processed;The distance feature represents the distance between the center point of the image to be processed and the boundary point of the image to be processed;According to the distance feature of the boundary point, the predicted position of the boundary point of the image to be processed is determined;According to the predicted position of the boundary point of the image to be processed and the boundary point category feature, the predicted category of the boundary point is determined;According to the predicted position of the boundary point of the image to be processed and the predicted category of the boundary point, the drivable area of the target vehicle is determined.The application improves the accuracy of vehicle detection drivable area, improves the safety in the process of vehicle driving.

Description

Technical Field

[0001] This application relates to the field of intelligent automatic parking technology, and in particular to a method, device and storage medium for detecting the drivable area of ​​a vehicle. Background Technology

[0002] With the development of autonomous driving technology, automated parking has become an indispensable implementation scenario. Drivable area detection algorithms are a key component of autonomous driving technology, crucial for ensuring the safe and efficient operation of autonomous vehicles on roads. In existing technologies, advancements in computer vision have provided the foundation for drivable area detection. Through image processing and analysis techniques, algorithms can extract road features and obstacle information from visual data captured by cameras. The development of deep learning, especially convolutional neural networks (CNNs), has greatly promoted the advancement of image segmentation technology. Deep learning models can automatically learn feature representations in images, thereby more accurately identifying and classifying drivable areas. Sensors such as LiDAR, radar, GPS, and inertial measurement units (IMUs) provide autonomous vehicles with rich environmental perception capabilities. This sensor data can be combined with visual data to provide more accurate and robust drivable area detection. Data from a single sensor may have limitations, while multi-sensor fusion technology can effectively combine information from different sensors, improving the accuracy and reliability of detection.

[0003] However, in daily life, vehicles may face a variety of complex environments when parking automatically, including city streets, highways, and rural roads. In order to ensure safe driving, the drivable area detection algorithm must be able to process sensor data and make decisions within milliseconds, which places high demands on the algorithm's computational efficiency. At the same time, the drivable area detection algorithm needs to maintain high accuracy under various weather conditions and changes in lighting. In existing technologies, the real-time performance and computational efficiency of the algorithms are not good, and the accuracy of the detection is not precise enough. Summary of the Invention

[0004] This application provides a method, apparatus, and storage medium for detecting the drivable area of ​​a vehicle, which can improve the accuracy of vehicle detection of the drivable area and enhance safety during driving.

[0005] On the one hand, this application provides a method for detecting a vehicle-drivable area, the method comprising:

[0006] Obtain the image to be processed from a preset area corresponding to the target vehicle; the preset area is the area whose distance from the target vehicle meets a preset distance condition;

[0007] Feature extraction is performed on the image to be processed to obtain the distance features and boundary point category features corresponding to the boundary points; the distance features represent the distance between the boundary points of the image to be processed and the center point of the image to be processed.

[0008] Based on the distance features corresponding to the boundary points, the predicted positions of the boundary points in the image to be processed are determined.

[0009] The predicted category of the boundary points is determined based on the predicted location of the boundary points in the image to be processed and the category features of the boundary points.

[0010] The drivable area of ​​the target vehicle is determined based on the predicted location and predicted category of the boundary points in the image to be processed.

[0011] In one exemplary embodiment, the step of extracting features from the image to be processed to obtain distance features and boundary point category features corresponding to the boundary points includes:

[0012] The image to be processed is input into the distance prediction network in the feature extraction model for distance prediction processing. The distance between the center point and the boundary point of the image to be processed is predicted at preset increments along the target direction of the image to be processed, so as to obtain the target number of distances between the center point and the boundary point of the image to be processed.

[0013] The distance feature of the target quantity dimension is determined based on the distance between the center point and the boundary point of the image to be processed of the target quantity.

[0014] The image to be processed is input into the boundary point category recognition network in the feature extraction model for boundary point category recognition processing. The category of the boundary points of the image to be processed is recognized at each preset increment along the target direction of the image to be processed, so as to obtain the target number of boundary point categories.

[0015] Based on the boundary point categories of the target quantity, determine the boundary point category features of the target quantity dimension.

[0016] In one exemplary embodiment, determining the predicted location of the boundary points of the image to be processed based on the distance features corresponding to the boundary points includes:

[0017] The coordinate information of the target boundary point is determined based on the distance between any target boundary point and the center point of the image to be processed; the target boundary point is any boundary point among the target number of boundary points;

[0018] Based on the coordinate information of the target boundary point, determine the predicted position of the target boundary point;

[0019] Based on the predicted positions of the target boundary points, the predicted positions of each boundary point in the image to be processed are determined.

[0020] In one exemplary embodiment, determining the predicted category of the boundary points based on the predicted locations of the boundary points in the image to be processed and the category features of the boundary points includes:

[0021] Based on the boundary point category features, determine the location category corresponding to the boundary point of the image to be processed;

[0022] The target prediction category of the target boundary point is determined based on the location category corresponding to any target boundary point and the predicted location of the target boundary point;

[0023] Based on the target prediction category of the target boundary point, the prediction category of each boundary point in the image to be processed is determined.

[0024] In one exemplary embodiment, determining the drivable area of ​​the target vehicle based on the predicted locations and predicted categories of the boundary points in the image to be processed includes:

[0025] Based on the predicted positions of the boundary points of the image to be processed, the boundary points of the image to be processed are connected to obtain a closed region;

[0026] Based on the predicted category of the boundary points, determine the drivable area boundary points and the prohibited drivable area boundary points of the image to be processed.

[0027] The drivable area of ​​the target vehicle is determined based on the boundary points of the drivable area, the boundary points of the prohibited area, and the closed area.

[0028] In one exemplary embodiment, the training method of the feature extraction model includes:

[0029] Obtain sample top-view image data; the sample top-view image data is labeled with sample distance labels and sample boundary point category labels;

[0030] Based on a feature extraction network with a preset model, feature extraction processing is performed on the sample top view image data to obtain sample distance features and sample boundary point category features corresponding to the sample top view image data.

[0031] Based on the distance prediction network, the distance prediction process is performed on the sample distance features to obtain the sample distance results of the sample top view image data;

[0032] Based on the boundary point category recognition network, the sample boundary point category features are processed to identify the boundary point category, and the sample boundary point category results of the top view image data are obtained.

[0033] Based on the first difference between the sample distance result and the sample distance label, and the second difference between the sample boundary point category result and the sample boundary point category label, the preset model is trained, and the feature extraction network in the preset model after training is used as the feature extraction model.

[0034] In one exemplary embodiment, training the preset model based on the first difference between the sample distance result and the sample distance label, and the second difference between the sample boundary point category result and the sample boundary point category label, and using the feature extraction network in the preset model after training as the feature extraction model, includes:

[0035] Based on the first difference between the sample distance result and the sample distance label, the first loss information is determined;

[0036] The second loss information is determined based on the second difference between the sample boundary point category results and the sample boundary point category labels;

[0037] Based on the first loss information and the second loss information, determine the target loss information;

[0038] Based on the target loss information, the parameters of the preset model are adjusted, and the feature extraction network in the preset model after training is used as the feature extraction model.

[0039] On the other hand, a device for detecting a vehicle-drivable area is provided, the device comprising:

[0040] The image to be processed module is used to acquire an image of a preset area corresponding to the target vehicle; the preset area is an area whose distance from the target vehicle meets a preset distance condition.

[0041] The feature extraction module is used to extract features from the image to be processed, and obtain the distance features and boundary point category features corresponding to the boundary points; the distance features represent the distance between the boundary points of the image to be processed and the center point of the image to be processed.

[0042] The prediction location module is used to determine the predicted location of the boundary points of the image to be processed based on the distance features corresponding to the boundary points.

[0043] The prediction category module is used to determine the predicted category of the boundary points based on the predicted positions of the boundary points in the image to be processed and the category features of the boundary points.

[0044] The drivable area module is used to determine the drivable area of ​​the target vehicle based on the predicted positions of the boundary points of the image to be processed and the predicted categories of the boundary points.

[0045] On the other hand, an electronic device is provided, the device including a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded by the processor and executed as described above for the method of detecting a vehicle drivable area.

[0046] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction or at least one program is stored therein, the at least one instruction or at least one program being loaded and executed by a processor to implement the vehicle drivable area detection method as described above.

[0047] On the other hand, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method for detecting a vehicle drivable area as described above.

[0048] The method, apparatus, and storage medium for detecting the drivable area of ​​a vehicle provided in this application have the following technical advantages:

[0049] This application acquires a pre-defined image of a target vehicle; the pre-defined area is a region whose distance from the target vehicle meets a pre-defined distance condition; features are extracted from the image to be processed to obtain distance features and boundary point category features corresponding to boundary points; the distance features characterize the distance between the boundary points of the image to be processed and the center point of the image to be processed; the predicted position of the boundary points of the image to be processed is determined based on the distance features corresponding to the boundary points; the predicted category of the boundary points is determined based on the predicted position and the boundary point category features; and the drivable area of ​​the target vehicle is determined based on the predicted position and the predicted category of the boundary points of the image to be processed. This application acquires a pre-defined image of a target vehicle, extracts features from the image to be processed to obtain distance features and boundary point category features, and uses these two features to determine the predicted position and category of the boundary points of the image to be processed, thereby determining the drivable area. This improves the accuracy of drivable area prediction, enhances the precision and reliability of vehicle drivable area detection, and improves the safety of vehicle operation. Attached Figure Description

[0050] To more clearly illustrate the technical solutions and advantages in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating a method for detecting a vehicle's drivable area provided in an embodiment of this specification.

[0052] Figure 2 This is a flowchart illustrating a method for detecting a vehicle's drivable area provided in an embodiment of this specification.

[0053] Figure 3 This is a flowchart illustrating a method for detecting a vehicle's drivable area provided in an embodiment of this specification.

[0054] Figure 4 This is a schematic diagram of the structure of a feature extraction network provided in the embodiments of this specification;

[0055] Figure 5 This is a flowchart illustrating a method for detecting a vehicle's drivable area provided in an embodiment of this specification.

[0056] Figure 6 This is a flowchart illustrating a method for detecting a vehicle's drivable area provided in an embodiment of this specification.

[0057] Figure 7 This is a schematic diagram of a boundary point prediction method provided in the embodiments of this specification;

[0058] Figure 8 This is a flowchart illustrating a method for detecting a vehicle's drivable area provided in an embodiment of this specification.

[0059] Figure 9 This is a flowchart illustrating a method for detecting a vehicle's drivable area provided in an embodiment of this specification.

[0060] Figure 10 This is a schematic diagram of a vehicle-driving area provided in the embodiments of this specification;

[0061] Figure 11 This is a schematic diagram of a vehicle-driving area provided in the embodiments of this specification;

[0062] Figure 12 This is a schematic diagram of the structure of a vehicle driving area detection device provided in the embodiments of this specification.

[0063] Figure 13This is a schematic diagram of the server structure for a vehicle driving area detection method provided in the embodiments of this specification. Detailed Implementation

[0064] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0065] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0066] The following describes a method for detecting the drivable area of ​​a vehicle, as described in this application. Figure 1 This is a flowchart illustrating a method for detecting a vehicle's drivable area according to an embodiment of this specification. This specification provides the operational steps of the method described in the embodiment or flowchart, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiment is merely one possible execution order among many and does not represent the only possible execution order. In actual system or server product execution, the method can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment) as shown in the embodiment or drawings. Specifically, as... Figure 1 As shown, the method can be applied to a control unit in a vehicle, and the method includes:

[0067] S101: Obtain the image to be processed of the preset area corresponding to the target vehicle; the preset area is the area whose distance from the target vehicle meets the preset distance condition.

[0068] In the embodiments of this specification, the image to be processed can be a top-view image obtained by performing IPM (Inverse Perspective Mapping) processing on an image captured by a camera, webcam, or other image acquisition device on the target vehicle. Due to perspective effects, objects that are originally parallel appear not to be parallel in the image captured by the image acquisition device. The idea behind IPM is to transform the original perspective result into a bird's-eye view, eliminating perspective distortion and making the image appear as if it is being viewed vertically downwards. The preset distance condition can be defined as the area centered on the target vehicle where the distance to the target vehicle is less than or equal to a distance value 'a'. Converting the perspective image into a bird's-eye view through IPM transformation makes lane lines, road edges, and other ground features more linear, simplifying the subsequent feature extraction process. Furthermore, IPM helps the system more accurately understand the relative position and distance of objects, which is beneficial for vehicle obstacle avoidance.

[0069] S103: Perform feature extraction on the image to be processed to obtain the distance features and boundary point category features corresponding to the boundary points; the distance features represent the distance between the boundary points of the image to be processed and the center point of the image to be processed.

[0070] In the embodiments of this specification, feature extraction is performed on the image to be processed, and distance features and boundary point category features are extracted, which facilitates the subsequent determination of the drivable area and improves the accuracy of vehicle detection of the drivable area.

[0071] In the embodiments of this specification, feature extraction is performed on the image to be processed to obtain the distance features and boundary point category features corresponding to the boundary points, such as... Figure 2 As shown, Figure 2 A flowchart illustrating a method for detecting a vehicle's drivable area is provided, including:

[0072] S201: Input the image to be processed into the distance prediction network in the feature extraction model for distance prediction processing. Predict the distance between the center point of the image to be processed and the boundary point of the image to be processed at preset increments along the target direction of the image to be processed, and obtain the target number of distances between the center point of the image to be processed and the boundary point of the image to be processed.

[0073] In the embodiments of this specification, for example, the target direction can be a clockwise direction starting from the 12 o'clock position of the top view of the image to be processed; the preset increment can be 1 degree; along the target direction of the image to be processed, the distance between the center point and the boundary point of the image to be processed is predicted at preset increments, that is, along the 12 o'clock position of the top view of the image to be processed, one boundary point is predicted at 1 degree intervals, and the predicted result is the distance between the center point of the image to be processed and the current boundary point of the image to be processed. Since the image has 360 degrees, a total of 360 boundary points can be predicted, and the target number can be 360.

[0074] In the embodiments of this specification, the training method of the feature extraction model is as follows: Figure 3 As shown, Figure 3 A flowchart illustrating a method for detecting a vehicle's drivable area is provided, including:

[0075] S301: Obtain sample top-view image data; the sample top-view image data is labeled with sample distance labels and sample boundary point category labels.

[0076] In the embodiments described in this specification, such as Figure 4 As shown, Figure 4 This is a schematic diagram of the structure of a feature extraction network. Figure 4 In the diagram, 401 represents the input image with dimensions 1024*1024*3, 402 represents the feature extraction network using RegNet as its backbone, and 403 represents the output of the last layer of the feature extraction network, consisting of two sets of 360-dimensional feature vectors. One set represents the distance to points in the drivable area, and the other represents the category of these points. Different categories can be represented by different color channels; for example, blue points represent drivable areas, red points represent obstacles (vehicles), and green points represent obstacles (walls). Specifically, the sample top-down image data is the input data, with dimensions of 1024*1024*3 (1024x1024 pixels) and three color channels (RGB). This means each pixel consists of red, green, and blue colors.

[0077] S302: Based on a preset model, a feature extraction network is used to perform feature extraction processing on the sample top-view image data to obtain sample distance features and sample boundary point category features corresponding to the sample top-view image data.

[0078] In the embodiments of this specification, specifically, the feature extraction model uses Regnet as the backbone network, i.e., the feature extraction network. The sample top-view image data is input into the preset model, and the feature extraction network Regnet performs feature extraction processing. First, it learns from shallow texture, color and boundary information to obtain shallow features. Then, it learns semantic feature information through deep networks, and fuses shallow features and high-level semantic information. Finally, the last layer of the network obtains sample distance features and sample boundary point category features. The sample distance features and sample boundary point category features have the same dimension, which can be 360 ​​dimensions, for example. The sample distance feature represents the predicted distance between the boundary points and the center point of the sample top-view image data; the sample boundary point category feature represents the predicted category of the boundary points in the sample top-view image data. By choosing RegNet as the backbone network, the RegNet network optimizes the ratio of network width and depth, allowing each layer to utilize its computational resources more efficiently, reducing the amount of computation, thereby accelerating the training process and improving the model's convergence speed, reducing the total time required for training iterations. Furthermore, the RegNet network design considers the effective use of computational resources, thus saving hardware resources. At the same time, because the RegNet architecture can flexibly adjust its width and depth according to needs, lightweight versions of the model can be built. These models maintain high accuracy while significantly reducing the requirements for computational resources and optimizing the model inference process. The model's design considers computational efficiency, ensuring timely processing of input data and output of results in practical applications, and ensuring the model's real-time processing capability on various hardware platforms. Furthermore, when an image is input into the model, the model directly outputs a highly structured feature vector, which can be directly used in the decision-making logic of an autonomous driving system. It can be seen that while achieving an end-to-end network, there is almost no post-processing computation, which reduces the resource consumption in actual use.

[0079] S303: Based on the distance prediction network, perform distance prediction processing on the sample distance features to obtain the sample distance results of the sample top view image data.

[0080] In the embodiments of this specification, a network architecture specifically designed for distance prediction is used to perform distance prediction processing on sample distance features, thereby enabling the vehicle to achieve fast response and accurate distance from the center point of the image to the boundary point of the drivable area. At the same time, it can work under different environmental conditions and maintain high prediction accuracy even in the presence of light changes, occlusion or other interference factors.

[0081] S304: Based on the boundary point category recognition network, perform boundary point category recognition processing on the sample boundary point category features to obtain the sample boundary point category results of the top view image data.

[0082] In the embodiments of this specification, by designing a network architecture specifically for category recognition, the boundary point category network performs distance prediction processing on the category features of sample boundary points, thereby enabling the vehicle to respond quickly and accurately identify the object category, and the vehicle can maintain high recognition accuracy under different lighting conditions, viewing angles and occlusion conditions.

[0083] S305: Based on the first difference between the sample distance result and the sample distance label, and the second difference between the sample boundary point category result and the sample boundary point category label, train the preset model, and use the feature extraction network in the preset model after training as the feature extraction model.

[0084] In the embodiments of this specification, the model is trained by combining distance differences and boundary point category differences, which enables the model to achieve accurate recognition in complex environments, thereby improving the model's accuracy, robustness, and real-time performance.

[0085] In this embodiment of the specification, the preset model is trained based on the first difference between the sample distance result and the sample distance label, and the second difference between the sample boundary point category result and the sample boundary point category label. The feature extraction network in the preset model after training is used as the feature extraction model. Figure 5 As shown, Figure 5 A flowchart illustrating a method for detecting a vehicle's drivable area is provided, including:

[0086] S501: Determine the first loss information based on the first difference between the sample distance result and the sample distance label.

[0087] In the embodiments of this specification, the sample distance result is the predicted distance between the boundary point and the center point of the sample top-view image data, and the sample distance label is the true distance between the boundary point and the center point of the sample top-view image data. First loss information is determined based on the difference between the true distance and the predicted distance. The first loss information can be determined using a drivable area detection algorithm loss function. For example, the Smooth L1 (Smoothed Average Absolute Error) loss function can be used to measure the distance between the predicted distance and the true distance. The formula for the Smooth L1 loss function is as follows:

[0088]

[0089] Among them, l loss For the loss value of a single sample, x n The distance to the truth value, y nFor distance prediction, beta is a hyperparameter that determines the inflection point where the loss function transitions from quadratic loss to linear loss.

[0090] Using the Smooth L1 loss function, when the absolute value of the difference between the predicted distance and the true distance is less than beta, the loss function takes the form of mean squared error, i.e., 0.5 × (x n -y n ) 2 This form encourages the model to minimize residuals when the error is small, because the loss function penalizes the error quadratically at this point; however, when the absolute value of the difference between the predicted distance and the true distance is greater than or equal to beta, the loss function takes the form of the absolute error, but with linear scaling, i.e., |x n -y n The form -0.5×beta provides a gentler penalty for larger errors, preventing outliers from having an excessive impact on model training and improving the model's robustness. It is evident that the Smooth L1 loss function combines the advantages of mean squared error and absolute error, balancing the model's sensitivity to small errors with its robustness to large errors.

[0091] S502: Determine the second loss information based on the second difference between the sample boundary point category result and the sample boundary point category label.

[0092] In the embodiments of this specification, the sample boundary point category result is the predicted category of the boundary points in the sample top-view image data, and the sample boundary point category label is the actual category of the boundary points in the sample top-view image data. Second loss information is determined based on the difference between the actual category and the predicted category. The second loss information can be determined using a category loss function. For example, the cross-entropy loss function can be used to measure the difference between the predicted category and the actual category. The formula for the cross-entropy loss function is:

[0093]

[0094] Among them, L CE Let be the loss value for a single sample, N be the total number of classes, i represent the class number, and y be the vector of actual classes. y is the probability vector for predicting the class. i Let be the value of the i-th element in the vector y representing the actual category. The probability vector for predicting the class The value of the i-th element in the array.

[0095] Using the cross-entropy loss function, when calculating categorical loss information, if the probability of the predicted category is close to 1, then... It will be a small negative number, so that A probability close to 0 results in a smaller loss; conversely, if the probability of the predicted class is far from 1, then... This will be a large negative number, leading to increased loss; by minimizing the cross-entropy loss to predict a probability distribution that is closer to the actual class, the accuracy and reliability of the predicted class are improved.

[0096] S503: Determine the target loss information based on the first loss information and the second loss information.

[0097] In the embodiments of this specification, target loss information representing distance difference and category difference is determined based on first loss information representing the difference between true distance and predicted distance, and second loss information representing the difference between actual category and predicted category. For example, the target loss information is determined based on the drivable area detection algorithm loss function and the category loss function; the formula for the target loss information can be...

[0098] L total =λ·L SmoothL1 +(1-λ)·L CE

[0099] Among them, L total L represents the target loss information, λ is a hyperparameter that adjusts the weights between the two loss information values. SmoothL1 As the first piece of information regarding loss, L CE This is the second piece of information that is lost.

[0100] Since different loss functions have different numerical ranges, the category loss function fluctuates within a smaller range, while the drivable area detection algorithm loss function has a larger value. For example, the weight of the category loss function can be 0.3, while the weight of the drivable area detection algorithm loss function can be 1. The impact of different losses on model training can be balanced by adjusting the hyperparameters.

[0101] S504: Based on the target loss information, adjust the parameters of the preset model, and use the feature extraction network in the preset model after training as the feature extraction model.

[0102] In the embodiments of this specification, after determining the target loss information, the gradient can be calculated using the target loss information, and then the parameters of the preset model can be updated through backpropagation. The feature extraction network in the preset model after training is used as the feature extraction model. By combining the first loss information and the second loss information, the model parameters are adjusted, which balances the model's sensitivity to small errors and its robustness to large errors, and improves the accuracy and reliability of the model's prediction results.

[0103] S202: Determine the distance feature of the target quantity dimension based on the distance between the center point of the target quantity image and the boundary point of the target quantity image.

[0104] In the embodiments of this specification, since 360 ​​boundary points of the images to be processed are predicted, a 360-dimensional distance feature can be obtained. Each dimension represents the distance between the center point of an image to be processed and the boundary point of the image to be processed in the current dimension. By predicting the distance from the center point of the image to be processed to the boundary point of the drivable area in each dimension, the accuracy and reliability of the detection are improved.

[0105] S203: Input the image to be processed into the boundary point category recognition network in the feature extraction model for boundary point category recognition processing. Identify the category of the boundary points of the image to be processed at preset increments along the target direction of the image to be processed to obtain the target number of boundary point categories.

[0106] In the embodiments of this specification, for example, the target direction can be a clockwise direction starting from the 12 o'clock position of the top view of the image to be processed; the preset increment can be 1 degree; along the target direction of the image to be processed, the category of the boundary points of the image to be processed is identified at each preset increment, that is, along the 12 o'clock position of the top view of the image to be processed, the category of one boundary point is predicted at each 1 degree interval, and the predicted result is the category of the boundary points of the current image to be processed. Since the image has 360 degrees, a total of 360 boundary point categories can be predicted, and the target number is 360.

[0107] S204: Determine the boundary point category features of the target quantity dimension based on the boundary point category of the target quantity.

[0108] In the embodiments of this specification, since the categories of the boundary points of 360 images to be processed are predicted, 360-dimensional boundary point category features can be obtained. Each dimension represents the category of the boundary points of the current dimension of the image to be processed. By predicting the category of the boundary points of each dimension of the image to be processed, it is possible to predict the drivable area and detect the categories of complex boundaries and obstacles, thereby improving the accuracy and diversity of detection.

[0109] S105: Determine the predicted position of the boundary point of the image to be processed based on the distance features corresponding to the boundary point.

[0110] In the embodiments of this specification, the predicted position of the boundary point in each dimension of the image to be processed is determined by the distance between the boundary point in each dimension of the image to be processed and the center point of the image to be processed. This achieves a comprehensive judgment of the drivable area in the image to be processed, and improves the accuracy and reliability of vehicle detection of drivable areas.

[0111] In the embodiments of this specification, the step of determining the predicted position of the boundary points of the image to be processed based on the distance features corresponding to the boundary points is as follows: Figure 6 As shown, Figure 6 A flowchart illustrating a method for detecting a vehicle's drivable area is provided, including:

[0112] S601: Determine the coordinate information of the target boundary point based on the distance between any target boundary point and the center point of the image to be processed; the target boundary point is any boundary point among the target number of boundary points.

[0113] In the embodiments of this specification, the distance value between any target boundary point and the center point of the image to be processed is mapped to the center point of the image to be processed, and the distance value is converted into a coordinate value, thereby determining the coordinate information of the target boundary point. For example, assuming that the horizontal direction of the image to be processed is the x-axis, the vertical direction of the image to be processed is the y-axis, the center point of the image to be processed is the origin, the predicted distance of the target boundary point is y, and the coordinate information of the target boundary point is determined as (a, b) based on the predicted distance y.

[0114] S602: Determine the predicted position of the target boundary point based on the coordinate information of the target boundary point.

[0115] In the embodiments of this specification, for example, it is assumed that the horizontal direction of the image to be processed is the x-axis, the vertical direction of the image to be processed is the y-axis, the center point of the image to be processed is the origin, and the coordinates of the target boundary point are (a, b). Based on the coordinate information of the target boundary point, the point with a value of a in the horizontal direction and a value of b in the vertical direction is determined, thereby determining the predicted position of the target boundary point.

[0116] S603: Determine the predicted positions of each boundary point in the image to be processed based on the predicted positions of the target boundary points.

[0117] In the embodiments described in this specification, such as Figure 7 As shown, Figure 7 This is a schematic diagram of boundary point prediction. Figure 7 The points on the boundary represent the boundary points of the image to be processed. In fact, there should be a total of 360 boundary points for each dimension. The figure is only for illustration. By converting the predicted distance of each dimension into coordinate values ​​and then determining the predicted position of the boundary point of each dimension, the accuracy and reliability of vehicle detection of drivable areas are improved.

[0118] S107: Determine the predicted category of the boundary points based on the predicted positions of the boundary points in the image to be processed and the category features of the boundary points.

[0119] In the embodiments of this specification, based on the boundary points mapped on the boundary of the image to be processed and combined with the boundary point category features, the predicted category of the boundary points in each dimension is determined, thereby realizing comprehensive category prediction of the boundary of the image to be processed and achieving comprehensiveness and reliability of detection.

[0120] In the embodiments of this specification, the step of determining the predicted category of the boundary points based on the predicted positions of the boundary points in the image to be processed and the category features of the boundary points is as follows: Figure 8 As shown, Figure 8 A flowchart illustrating a method for detecting a vehicle's drivable area is provided, including:

[0121] S801: Determine the position category corresponding to the boundary point of the image to be processed based on the boundary point category features.

[0122] In the embodiments of this specification, the boundary point category feature represents the category of the boundary points of the image to be processed in each dimension. Here, the boundary point is a vague concept and is not precise to a point, but rather represents the category of an approximate boundary point position corresponding to this dimension. Therefore, the position category corresponding to the boundary point of the image to be processed in each dimension can be determined based on the boundary point category feature.

[0123] S802: Determine the target prediction category of the target boundary point based on the location category corresponding to any target boundary point and the predicted location of the target boundary point.

[0124] In the embodiments of this specification, the position category of the dimension corresponding to any target boundary point is determined according to the position category corresponding to the target boundary point, and the specific boundary point position of the current dimension corresponding to the target boundary point is determined according to the predicted position of the target boundary point, thereby determining the target prediction category of the target boundary point.

[0125] S803: Determine the prediction category of each boundary point in the image to be processed based on the target prediction category of the target boundary point.

[0126] In the embodiments described in this specification, such as Figure 7 As shown, Figure 7 This is a schematic diagram of boundary point prediction. Figure 7In the diagram, 701 represents a vehicle, 702 a wall, and 703 indicates no obstacles. Different color channels can represent different categories of points in the output. Blue areas represent drivable areas, red areas represent obstacles (vehicles), and green areas represent obstacles (walls). Therefore, boundary points next to obstacle 701 are red, those next to obstacle 702 are green, and those next to obstacle 703 are blue. Combining the predicted categories of all target boundary points, the predicted category of each boundary point in the image to be processed is determined. By determining the category of the boundary points corresponding to each dimension, the category of all boundary points in the image to be processed can be accurately identified, facilitating the identification of drivable areas or obstacle categories and improving detection accuracy and efficiency.

[0127] S109: Determine the drivable area of ​​the target vehicle based on the predicted positions of the boundary points of the image to be processed and the predicted categories of the boundary points.

[0128] In the embodiments of this specification, the drivable area detected by the target vehicle is determined by the predicted boundary points on the boundary of the image to be processed and the prediction category corresponding to each boundary point. This not only detects the drivable area but also ensures the integrity of the prediction of complex boundaries, achieving comprehensive area detection and improving the accuracy and efficiency of detection.

[0129] In the embodiments of this specification, the drivable area of ​​the target vehicle is determined based on the predicted positions and predicted categories of the boundary points of the image to be processed. Figure 9 As shown, Figure 9 A flowchart illustrating a method for detecting a vehicle's drivable area is provided, including:

[0130] S901: Based on the predicted positions of the boundary points of the image to be processed, connect the boundary points of the image to be processed to obtain a closed region.

[0131] In the embodiments of this specification, the specific location of each boundary point is located based on the predicted positions of the boundary points of the image to be processed, and the boundary points of the image to be processed are connected to obtain a closed region. This simplifies the detection process while improving the accuracy and reliability of the detection.

[0132] S902: Based on the predicted category of the boundary points, determine the drivable area boundary points and the prohibited drivable area boundary points of the image to be processed.

[0133] In the embodiments of this specification, drivable area boundary points represent all drivable areas, and prohibited drivable area boundary points represent the presence of obstacles and their categories. The categories of obstacles can include walls, curbs, people, vehicles, pillars, and other obstacles. This achieves accurate detection of the type of drivable area boundaries, ensuring the completeness of complex boundary predictions and enabling vehicles to avoid obstacles in a timely manner.

[0134] S903: Determine the drivable area of ​​the target vehicle based on the boundary points of the drivable area, the boundary points of the prohibited area, and the closed area.

[0135] In the embodiments described in this specification, such as Figure 10-11 As shown, Figure 10-11 This is a schematic diagram of a drivable area. Figure 10 In the diagram, 1001 represents the target vehicle, 1003 represents the obstacle vehicle, 1004 represents the obstacle vehicle, and 1002 represents the closed area. Figure 11 In the diagram, 1101 represents the target vehicle, 1102 represents the closed area, and 1103 represents the obstacle wall. Based on the boundary points of the drivable area, the boundary points of the prohibited area, and the connected closed area, the drivable area is determined. In practical applications, the drivable area can be marked in green for a more prominent effect, making it easier for users to distinguish the drivable area and avoid obstacles in time, thus improving safety during driving.

[0136] This manual also provides a device for detecting the drivable area of ​​a vehicle, such as... Figure 12 As shown, the device includes:

[0137] The image to be processed module 1201 is used to acquire an image of a preset area corresponding to the target vehicle; the preset area is an area whose distance from the target vehicle meets a preset distance condition.

[0138] The feature extraction module 1202 is used to extract features from the image to be processed, and obtain the distance features and boundary point category features corresponding to the boundary points; the distance features represent the distance between the boundary points of the image to be processed and the center point of the image to be processed.

[0139] The prediction position module 1203 is used to determine the predicted position of the boundary points of the image to be processed based on the distance features corresponding to the boundary points.

[0140] The prediction category module 1204 is used to determine the predicted category of the boundary points based on the predicted positions of the boundary points in the image to be processed and the category features of the boundary points.

[0141] The drivable area module 1205 is used to determine the drivable area of ​​the target vehicle based on the predicted position of the boundary points of the image to be processed and the predicted category of the boundary points.

[0142] In some embodiments, the feature extraction module further includes:

[0143] The target quantity submodule is used to input the image to be processed into the distance prediction network in the feature extraction model for distance prediction processing. It predicts the distance between the center point and the boundary point of the image to be processed at preset increments along the target direction of the image to be processed, and obtains the target quantity of the distance between the center point and the boundary point of the image to be processed.

[0144] The target quantity dimension submodule is used to determine the distance feature of the target quantity dimension based on the distance between the center point of the image to be processed and the boundary point of the image to be processed, according to the target quantity.

[0145] The boundary point category submodule is used to input the image to be processed into the boundary point category recognition network in the feature extraction model for boundary point category recognition processing. It identifies the category of the boundary points of the image to be processed at preset increments along the target direction of the image to be processed, thereby obtaining the target number of boundary point categories.

[0146] The category feature submodule is used to determine the boundary point category features of the target quantity dimension based on the boundary point categories of the target quantity.

[0147] In some embodiments, the predicted location module further includes:

[0148] The coordinate information submodule is used to determine the coordinate information of the target boundary point based on the distance between any target boundary point and the center point of the image to be processed; the target boundary point is any boundary point among the target number of boundary points.

[0149] The predicted location determination submodule is used to determine the predicted location of the target boundary point based on the coordinate information of the target boundary point.

[0150] The location determination submodule is used to determine the predicted location of each boundary point in the image to be processed based on the predicted location of the target boundary point.

[0151] In some embodiments, the prediction category module further includes:

[0152] The location category submodule is used to determine the location category corresponding to the boundary point of the image to be processed based on the boundary point category features.

[0153] The target prediction category submodule is used to determine the target prediction category of the target boundary point based on the location category corresponding to any target boundary point and the predicted location of the target boundary point.

[0154] The prediction category determination submodule is used to determine the prediction category of each boundary point in the image to be processed based on the target prediction category of the target boundary point.

[0155] In some embodiments, the drivable area module further includes:

[0156] The closed region submodule is used to connect the boundary points of the image to be processed based on the predicted positions of the boundary points of the image to be processed, so as to obtain a closed region.

[0157] The boundary point submodule is used to determine the drivable area boundary points and the prohibited drivable area boundary points of the image to be processed based on the predicted category of the boundary points.

[0158] The drivable area determination submodule is used to determine the drivable area of ​​the target vehicle based on the drivable area boundary points, the prohibited drivable area boundary points, and the closed area.

[0159] In some embodiments, the target quantity submodule further includes:

[0160] The sample top-view image data unit is used to acquire sample top-view image data; the sample top-view image data is labeled with sample distance labels and sample boundary point category labels.

[0161] The sample feature unit is used to perform feature extraction processing on the sample top view image data based on the feature extraction network of the preset model, so as to obtain the sample distance features and sample boundary point category features corresponding to the sample top view image data.

[0162] The sample distance result unit is used to perform distance prediction processing on the sample distance features based on the distance prediction network to obtain the sample distance result of the sample top view image data.

[0163] The sample boundary point category result unit is used to perform boundary point category recognition processing on the sample boundary point category features based on the boundary point category recognition network to obtain the sample boundary point category result of the top view image data.

[0164] The training unit is used to train the preset model based on the first difference between the sample distance result and the sample distance label, and the second difference between the sample boundary point category result and the sample boundary point category label, and to use the feature extraction network in the preset model after training as the feature extraction model.

[0165] In some embodiments, the training unit further includes:

[0166] The first loss information subunit is used to determine the first loss information based on the first difference between the sample distance result and the sample distance label.

[0167] The second loss information subunit is used to determine the second loss information based on the second difference between the sample boundary point category result and the sample boundary point category label.

[0168] The target loss information subunit is used to determine the target loss information based on the first loss information and the second loss information.

[0169] The parameter update subunit is used to adjust the parameters of the preset model according to the target loss information, and to use the feature extraction network in the preset model after training as the feature extraction model.

[0170] The apparatus and method embodiments described herein are based on the same inventive concept.

[0171] This specification provides a vehicle electronic device, which includes a processor and a memory. The memory stores at least one instruction or at least one program, which is loaded and executed by the processor to implement the vehicle drivable area detection method provided in the above method embodiments.

[0172] Embodiments of this application also provide a computer storage medium, which can be disposed in a terminal to store at least one instruction or at least one program related to implementing a method for detecting a drivable area of ​​a vehicle in the method embodiment. The at least one instruction or at least one program is loaded and executed by the processor to implement the method for detecting a drivable area of ​​a vehicle provided in the above method embodiment.

[0173] Embodiments of this application also provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method for detecting a drivable area of ​​a vehicle provided in the above-described method embodiments.

[0174] The memory described in the embodiments of this specification can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for the functions, etc.; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0175] The vehicle drivable area detection method embodiments provided in this specification can be executed on a mobile terminal, computer terminal, server, or similar computing device. Taking running on a server as an example, Figure 13 This is a hardware structure block diagram of a server for a vehicle drivable area detection method provided in the embodiments of this specification. Figure 13 As shown, the server 1300 can vary significantly due to different configurations or performance. It may include one or more Central Processing Units (CPUs) 1310 (CPUs 1310 may include, but are not limited to, microprocessors (MCUs) or programmable logic devices (FPGAs), a memory 1330 for storing data, and one or more storage media 1320 (e.g., one or more mass storage devices) for storing application programs 1323 or data 1322. The memory 1330 and storage media 1320 may be temporary or persistent storage. The program stored in the storage media 1320 may include one or more modules, each module may include a series of instruction operations on the server. Furthermore, the CPU 1310 may be configured to communicate with the storage media 1320 and execute the series of instruction operations stored in the storage media 1320 on the server 1300. Server 1300 may also include one or more power supplies 1360, one or more wired or wireless network interfaces 1350, one or more input / output interfaces 1340, and / or one or more operating systems 1321, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0176] The input / output interface 1340 can be used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of server 1300. In one example, the input / output interface 1340 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the input / output interface 1340 may be a radio frequency (RF) module for wireless communication with the Internet.

[0177] Those skilled in the art will understand that Figure 13 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, server 1300 may also include... Figure 13 The more or fewer components shown, or having the same Figure 13 The different configurations shown.

[0178] As can be seen from the embodiments of the vehicle drivable area detection method, apparatus, and storage medium provided in this application, this application acquires a pre-defined image of a target vehicle; the pre-defined area is a region whose distance from the target vehicle meets a pre-defined distance condition; features are extracted from the image to be processed to obtain distance features and boundary point category features corresponding to boundary points; the distance features characterize the distance between the boundary points of the image to be processed and the center point of the image to be processed; the predicted position of the boundary points of the image to be processed is determined based on the distance features corresponding to the boundary points; the predicted category of the boundary points is determined based on the predicted position of the boundary points of the image to be processed and the boundary point category features; and the drivable area of ​​the target vehicle is determined based on the predicted position of the boundary points of the image to be processed and the predicted category of the boundary points. This application uses Regnet as the backbone network to extract 360-dimensional distance features and 360-dimensional boundary point category features from the image to be processed by IPM, achieving end-to-end network-based feature vector generation, thus improving the efficiency, accuracy, and flexibility of detection. A 360-degree prediction method is used to predict the boundary points and their categories in each dimension of the image to be processed from the IPM perspective, thereby obtaining the drivable area. This ensures accurate prediction of the drivable area around the vehicle in a top-down view with limited computational resources, improving prediction accuracy in blind spots. Furthermore, the predicted boundary point categories ensure the completeness of complex boundary prediction and the identification of the drivable area boundary type. This achieves comprehensive and accurate detection of the drivable area, improving vehicle safety during driving.

[0179] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0180] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0181] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware, or by a program instructing related hardware. The program can be stored in a computer storage medium, such as a read-only memory, a disk, or an optical disk. The above descriptions are merely preferred embodiments of this application and are not intended to limit the application. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for detecting a drivable area, characterized in that, The method includes: Obtain the image to be processed from a preset area corresponding to the target vehicle; the preset area is the area whose distance from the target vehicle meets a preset distance condition; The image to be processed is input into the distance prediction network in the feature extraction model for distance prediction processing. The distance between the center point and the boundary point of the image to be processed is predicted at preset increments along the target direction of the image to be processed, so as to obtain the target number of distances between the center point and the boundary point of the image to be processed. The distance feature of the target quantity dimension is determined based on the distance between the center point and the boundary point of the image to be processed of the target quantity; the target quantity dimension represents the same dimension as the target quantity. The image to be processed is input into the boundary point category recognition network in the feature extraction model for boundary point category recognition processing. The category of the boundary points of the image to be processed is recognized at each preset increment along the target direction of the image to be processed, so as to obtain the target number of boundary point categories. Based on the boundary point categories of the target quantity, determine the boundary point category features of the target quantity dimension; Based on the distance features corresponding to the boundary points, the predicted positions of the boundary points in the image to be processed are determined. The predicted category of the boundary points is determined based on the predicted location of the boundary points in the image to be processed and the category features of the boundary points. The drivable area of ​​the target vehicle is determined based on the predicted location and predicted category of the boundary points in the image to be processed.

2. The detection method according to claim 1, characterized in that, Determining the predicted location of the boundary points in the image to be processed based on the distance features corresponding to the boundary points includes: The coordinate information of the target boundary point is determined based on the distance between any target boundary point and the center point of the image to be processed; the target boundary point is any boundary point among the target number of boundary points; Based on the coordinate information of the target boundary point, determine the predicted position of the target boundary point; Based on the predicted positions of the target boundary points, the predicted positions of each boundary point in the image to be processed are determined.

3. The detection method according to claim 2, characterized in that, The step of determining the predicted category of the boundary points based on the predicted positions of the boundary points in the image to be processed and the category features of the boundary points includes: Based on the boundary point category features, determine the location category corresponding to the boundary point of the image to be processed; The target prediction category of the target boundary point is determined based on the location category corresponding to any target boundary point and the predicted location of the target boundary point; Based on the target prediction category of the target boundary point, the prediction category of each boundary point in the image to be processed is determined.

4. The detection method according to claim 3, characterized in that, Determining the drivable area of ​​the target vehicle based on the predicted positions and predicted categories of the boundary points in the image to be processed includes: Based on the predicted positions of the boundary points of the image to be processed, the boundary points of the image to be processed are connected to obtain a closed region; Based on the predicted category of the boundary points, determine the drivable area boundary points and the prohibited drivable area boundary points of the image to be processed. The drivable area of ​​the target vehicle is determined based on the boundary points of the drivable area, the boundary points of the prohibited area, and the closed area.

5. The detection method according to claim 1, characterized in that, The training method for the feature extraction model includes: Obtain sample top-view image data; the sample top-view image data is labeled with sample distance labels and sample boundary point category labels; Based on a feature extraction network with a preset model, feature extraction processing is performed on the sample top view image data to obtain sample distance features and sample boundary point category features corresponding to the sample top view image data. Based on the distance prediction network, the distance prediction process is performed on the sample distance features to obtain the sample distance results of the sample top view image data; Based on the boundary point category recognition network, the boundary point category features of the sample are processed to identify the boundary point category, and the sample boundary point category results of the sample top view image data are obtained. Based on the first difference between the sample distance result and the sample distance label, and the second difference between the sample boundary point category result and the sample boundary point category label, the preset model is trained, and the feature extraction network in the preset model after training is used as the feature extraction model.

6. The detection method according to claim 5, characterized in that, The step of training the preset model based on the first difference between the sample distance result and the sample distance label, and the second difference between the sample boundary point category result and the sample boundary point category label, and using the feature extraction network in the preset model after training as the feature extraction model, includes: Based on the first difference between the sample distance result and the sample distance label, the first loss information is determined; The second loss information is determined based on the second difference between the sample boundary point category results and the sample boundary point category labels; Based on the first loss information and the second loss information, determine the target loss information; Based on the target loss information, the parameters of the preset model are adjusted, and the feature extraction network in the preset model after training is used as the feature extraction model.

7. A device for detecting a drivable area, characterized in that, The device includes: The image to be processed module is used to acquire an image of a preset area corresponding to the target vehicle; the preset area is an area whose distance from the target vehicle meets a preset distance condition. The feature extraction module is used to input the image to be processed into the distance prediction network in the feature extraction model for distance prediction processing. It predicts the distance between the center point and the boundary points of the image to be processed at preset increments along the target direction of the image to be processed, obtaining a target number of distances between the center point and the boundary points of the image to be processed. Based on the target number of distances between the center point and the boundary points of the image to be processed, it determines the distance features of the target quantity dimension. The module then inputs the image to be processed into the boundary point category recognition network in the feature extraction model for boundary point category recognition processing. It identifies the category of the boundary points of the image to be processed at preset increments along the target direction of the image to be processed, obtaining the boundary point category of the target number. Based on the boundary point category of the target number, it determines the boundary point category features of the target quantity dimension. The target quantity dimension represents the same dimension as the target quantity. The prediction location module is used to determine the predicted location of the boundary points of the image to be processed based on the distance features corresponding to the boundary points. The prediction category module is used to determine the predicted category of the boundary points based on the predicted positions of the boundary points in the image to be processed and the category features of the boundary points. The drivable area module is used to determine the drivable area of ​​the target vehicle based on the predicted positions of the boundary points of the image to be processed and the predicted categories of the boundary points.

8. An electronic device, characterized in that, The device includes a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded by the processor and executed as described in any one of claims 1-6 for detecting drivable areas.

9. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction or at least one program, which is loaded and executed by a processor to implement the method for detecting drivable areas as described in any one of claims 1-6.

Images