Parking space information determination method and device, equipment and medium

Abstract

The invention discloses a parking space information determination method and device, equipment and a medium. The method comprises the following steps: acquiring a real-time scene image, an initial parking scene type and initial sensor data of a to-be-detected area of a to-be-detected vehicle, wherein the initial sensor data comprises at least one vehicle sensor and a sensor weight corresponding to each vehicle sensor; recognizing the real-time scene image to obtain at least one scene label; according to each scene label, updating the initial sensor data, and determining target sensor data; determining a first sensor and a second sensor of the to-be-detected vehicle according to the target sensing data; and detecting the to-be-detected area according to the first sensor and the second sensor, and determining parking space information of the to-be-detected vehicle. According to the embodiment of the invention, the accuracy of determining the parking space information can be improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, device, and medium for determining parking space information. Background Technology

[0002] With the rapid development of technology, vehicles are becoming increasingly functional, among which automatic parking is gradually being applied to various types of vehicles. Vehicles equipped with automatic parking can detect parking spaces and automatically park themselves in those spaces.

[0003] Currently, parking spaces can be detected in the environment surrounding a vehicle using sensors to determine parking space information.

[0004] However, the environment around parking spaces is complex, making it difficult to obtain parking space information in complex scenarios. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for determining parking space information, so as to improve the accuracy of parking space information determination.

[0006] In a first aspect, embodiments of the present invention provide a method for determining parking space information, the method comprising:

[0007] Acquire real-time scene images of the area to be detected for the vehicle to be detected, the initial parking scene type, and the initial sensor data. The initial sensor data includes: at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor.

[0008] At least one scene label is obtained from real-time scene image recognition;

[0009] Based on the scene labels, update the initial sensor data and determine the target sensor data;

[0010] Based on the target sensing data, determine the first and second sensors of the vehicle to be detected;

[0011] Based on the detection of the area to be detected by the first and second sensors, the parking space information of the vehicle to be detected is determined.

[0012] Secondly, embodiments of the present invention also provide a parking space information determination device, the device comprising:

[0013] The initial data acquisition module is used to acquire real-time scene images of the area to be detected of the vehicle to be detected, the initial parking scene type, and the initial sensor data. The initial sensor data includes: at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor.

[0014] The label recognition module is used to identify at least one scene label from real-time scene images;

[0015] The target data acquisition module is used to update the initial sensor data based on the scene labels and determine the target sensor data.

[0016] The sensor determination module is used to determine the first and second sensors of the vehicle to be detected based on the target sensing data.

[0017] The parking space information determination module is used to determine the parking space information of the vehicle to be detected based on the detection of the area by the first sensor and the second sensor.

[0018] Thirdly, embodiments of the present invention also provide a parking space information determination device, the parking space information determination device comprising:

[0019] At least one processor; and

[0020] A memory that is communicatively connected to at least one processor; wherein,

[0021] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the parking space information determination method according to any embodiment of the present invention.

[0022] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the parking space information determination method of any embodiment of the present invention.

[0023] The technical solution of this invention acquires real-time scene images of the area to be detected for a vehicle, initial parking scene type, and initial sensor data. The initial sensor data includes at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor. At least one scene label is obtained by recognizing the real-time scene image. The initial sensor data is updated according to each scene label to determine target sensor data. The first and second sensors of the vehicle to be detected are determined according to the target sensor data. The parking space information of the vehicle to be detected is determined based on the detection of the area to be detected by the first and second sensors. Different sensor allocation schemes can be executed for different scenarios, refining the steps for acquiring parking space information and improving the accuracy of determining parking space information.

[0024] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of a parking space information determination method provided in Embodiment 1 of the present invention;

[0027] Figure 2 This is a flowchart of a parking space information determination method according to Embodiment 2 of the present invention;

[0028] Figure 3 This is a structural diagram of a parking space information determination device according to an embodiment of the present invention;

[0029] Figure 4 This is a schematic diagram of the structure of a parking space information determination device provided in an embodiment of the present invention. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention 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 the invention 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 a non-exclusive inclusion; for example, a process, method, system, product, or apparatus 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 apparatus.

[0032] The acquisition, storage, and application of real-time scene images and other data involved in the technical solutions of this invention comply with relevant laws and regulations and do not violate public order and good morals.

[0033] Example 1

[0034] Figure 1 This is a flowchart illustrating a parking space information determination method according to Embodiment 1 of the present invention. This embodiment of the invention is applicable to situations requiring parking space information determination. The method can be executed by a parking space information determination device, which can be implemented in hardware and / or software.

[0035] See Figure 1 The methods for determining parking space information shown include:

[0036] S101. Obtain the real-time scene image of the area to be detected for the vehicle to be detected, the initial parking scene type, and the initial sensor data. The initial sensor data includes: at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor.

[0037] The vehicle to be detected can be a vehicle undergoing parking space detection. The detection area can be the maximum detectable area of ​​each sensor on the vehicle. The real-time scene image can be an image of the detection area captured in real-time by the camera of the vehicle. The initial parking scene type can be a preset identifier of the parking environment conditions for the vehicle. The initial sensor data can be a set of preset initial sensor configuration information. The vehicle sensors can be sensors installed on the vehicle. The sensor weights can be descriptive information indicating the importance of the data collected by the vehicle sensors.

[0038] Specifically, at least one sensor can be installed on the vehicle to be inspected. For example, the sensor can be a camera, millimeter-wave radar, or ultrasonic sensor. The camera installed on the vehicle captures images of the area to be inspected, obtaining a real-time scene image of that area. The initial parking scenario type and initial sensor data can be determined based on the most frequently occurring operational scenario type and sensor data corresponding to the vehicle's historical time. The initial sensor data includes at least one vehicle sensor and its corresponding sensor weight. For example, parking scenario types can be sunny daytime, foggy / rainy daytime, nighttime, old underground parking garage, tunnel, or obstructed parking garage. In a scenario where the initial parking scenario type is sunny daytime, the vehicle sensors can be a camera, millimeter-wave radar, and ultrasonic sensor, with a sensor weight of 0.6 for the camera, 0.3 for the millimeter-wave radar, and 0.1 for the ultrasonic sensor.

[0039] S102. Obtain at least one scene label from real-time scene image recognition.

[0040] Among them, scene tags can be identifiers for parking scenes.

[0041] Specifically, obtaining at least one scene label for real-time scene image recognition involves using computer vision algorithms to extract and analyze features from the acquired images of the area to be detected, ultimately outputting a label that describes the key attributes of the current environment (such as lighting conditions, weather conditions, or road conditions), providing a basis for subsequent sensor weight adjustments. For example, the visual algorithm could be a convolutional neural network (CNN), utilizing its efficient extraction capabilities of image spatial features to achieve automatic classification and recognition of parking-related environments. Dataset construction: The data mainly comes from vehicle-mounted cameras and needs to cover multiple scenes to avoid data skew, such as different lighting conditions (daytime, nighttime, or underground parking garages) and different weather conditions (sunny, rainy, and cloudy). The acquired data is categorized, such as "rainy day." The input data is cropped (to unify the input resolution for easier learning by the CNN) and normalized (using the Min-Max normalization method to map pixel values ​​from [0,255] to [0,1], accelerating model convergence). By introducing residual connections on top of convolutional neural networks, ResNet can be used to extract more complex semantic features. The classification head maps the learned feature representations to class predictions. Spatial feature compression uses global pooling to calculate the mean or maximum value across the entire spatial dimension for each channel of the feature map, resulting in a 1×1 feature map, which is then flattened into a vector. Feature mapping uses a fully connected layer (FC) with ReLU activation and Dropout to map the compressed vectors. Class probability output uses the Softmax function to convert the feature-mapped vectors into classes. Real-time scene images are input into the model, and based on the trained convolutional neural network model, at least one scene label is output.

[0042] S103. Update the initial sensor data according to the scene labels and determine the target sensor data.

[0043] Specifically, based on a comparison between each scene label and the initial parking scene type, the similarity between the set of scene labels and the initial parking scene type is calculated to obtain the similarity comparison result. If the similarity comparison result is determined to be similar, it indicates that the current real-time scene type is consistent with the initial parking scene type, and therefore, there is no need to update the initial sensor data; the initial sensor data is designated as the target sensor data. If the similarity comparison result is determined to be dissimilar, it indicates that the current real-time scene type is inconsistent with the initial parking scene type, and therefore, the initial sensor data needs to be updated. The initial sensor data is updated according to the weight adjustment rules corresponding to each scene label, and the updated initial sensor data is designated as the target sensor data. Different scene labels correspond to different weight adjustment rules.

[0044] S104. Based on the target sensing data, determine the first sensor and the second sensor of the vehicle to be detected.

[0045] Specifically, the target sensors include at least one vehicle sensor and sensor weights for each vehicle sensor. The sensor weights are sorted, and the vehicle sensor with the highest weight is designated as the first sensor. The remaining vehicle sensors (excluding the first sensor) are designated as second sensors. When determining parking space information, the parking space information detected by the first sensor is primary, while the parking space information detected by the second sensors is secondary. There is at least one second sensor.

[0046] S105. Based on the detection of the area to be detected by the first sensor and the second sensor, determine the parking space information of the vehicle to be detected.

[0047] Specifically, timestamp alignment technology ensures that the first sensor and all second sensors collect data from the area to be detected within the same time window, avoiding collaboration errors caused by timing discrepancies. Preprocessing of the first sensor can involve specific optimizations tailored to its type (e.g., camera image denoising and distortion correction; millimeter-wave radar point cloud filtering and static target screening) to output high-quality core detection data. The second sensors also undergo preprocessing (e.g., ultrasonic filtering to remove abnormal ranging values) to ensure the stability of auxiliary data and provide a reliable foundation for verification. The first sensor is determined based on dynamically updated sensor weights, typically the sensor with the best performance and highest weight in the current scenario (e.g., camera in sunny weather, millimeter-wave radar in rainy or foggy weather). The second sensor is one or more sensors with weights second only to the first sensor, compensating for the shortcomings of the first sensor (e.g., millimeter-wave radar assists in ranging when camera is the primary sensor; camera assists in identifying parking lines when millimeter-wave radar is the primary sensor).

[0048] The technical solution of this invention acquires real-time scene images of the area to be detected for a vehicle, initial parking scene type, and initial sensor data. The initial sensor data includes at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor. At least one scene label is obtained by recognizing the real-time scene image. The initial sensor data is updated according to each scene label to determine target sensor data. The first and second sensors of the vehicle to be detected are determined according to the target sensor data. The parking space information of the vehicle to be detected is determined based on the detection of the area to be detected by the first and second sensors. Different sensor allocation schemes can be executed for different scenarios, refining the steps for acquiring parking space information and improving the accuracy of determining parking space information.

[0049] Example 2

[0050] Figure 2 This is a flowchart illustrating a parking space information determination method according to Embodiment 2 of the present invention. Based on the above embodiments, this embodiment optimizes and improves the parking space information determination operation.

[0051] Furthermore, the process of "updating the initial sensor data based on each scene label and determining the target sensor data" is refined into "identifying the scene label for each scene label and determining the label type corresponding to the scene label; finding the associated information corresponding to the label type based on the label type, including: associated sensors and the update level of the associated sensors; updating the initial sensor data based on the associated information and determining the target sensor data", in order to improve the operation of determining parking space information.

[0052] It should be noted that for parts not described in detail in the embodiments of the present invention, please refer to the descriptions in other embodiments.

[0053] See Figure 2 The methods for determining parking space information shown include:

[0054] S201. Obtain the real-time scene image of the area to be detected for the vehicle to be detected, the initial parking scene type, and the initial sensor data. The initial sensor data includes: at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor.

[0055] S202, Obtain at least one scene label from real-time scene image recognition.

[0056] S203. For each scene label, identify the scene label and determine the label type corresponding to the scene label.

[0057] The tag type can be the type of environmental condition corresponding to the scene tag.

[0058] Specifically, scene labels can be identifiers that directly describe environmental conditions, or identifiers that describe the characteristics of environmental conditions. For example, scene labels can be sunny, rainy, or snowy, or they can be well-lit, contain water droplets, have obstacles, or have sand and dust. For each scene label, scene label identification is performed to determine the corresponding label type. The label type can be lighting type, weather type, or obstacle type, etc.

[0059] S204. Based on the tag type, find the associated information corresponding to the tag type. The associated information includes: associated sensors and the update level of the associated sensors.

[0060] Among them, the associated sensors can be vehicle sensors that are affected by parking space detection under the environmental conditions corresponding to the tag type. The update level can be a description of the urgency of updating the sensor weights of the associated sensors.

[0061] Specifically, one tag type can correspond to at least one associated sensor. Each associated sensor corresponds one-to-one with its update level. Update levels can be level 1, level 2, and level 3, etc. Level 1 indicates a low urgency for updating the sensor weights of the associated sensor; level 2 indicates a medium urgency; and level 3 indicates a high urgency. Based on the tag type, the associated information corresponding to that tag type is retrieved. This associated information includes the associated sensor and its corresponding update level. The retrieval methods include, but are not limited to, sequential search algorithms, binary search algorithms, tree search algorithms, or hash search algorithms. This embodiment of the invention does not impose any limitations on these methods.

[0062] S205. Update the initial sensor data based on the associated information to determine the target sensor data.

[0063] Specifically, based on the associated information, at least one associated sensor affecting the collection of parking space information under the environmental conditions corresponding to the scene label is identified, along with the sensor weights corresponding to each associated sensor. For each associated sensor, the vehicle sensors in the initial sensor data are filtered to determine the vehicle sensors corresponding to the associated sensors. The sensor weights of the vehicle sensors are updated according to the update level of the associated sensors to determine the target sensor data.

[0064] S206. Based on the target sensing data, determine the first sensor and the second sensor of the vehicle to be detected.

[0065] S207. Based on the detection of the area to be detected by the first sensor and the second sensor, determine the parking space information of the vehicle to be detected.

[0066] This invention identifies scene tags for each scene and determines the tag type corresponding to each scene tag; based on the tag type, it finds the associated information corresponding to the tag type, including: associated sensors and the update level of the associated sensors; based on the associated information, it updates the initial sensor data and determines the target sensor data. This allows for updating the sensor weights of different vehicle sensors for different scene tags, improving the accuracy of sensor allocation.

[0067] Optionally, the initial sensor data is updated based on the association information to determine the target sensor data, including: determining the update step size based on the associated sensors in the association information; determining the weight offset based on the update step size and the update level corresponding to the associated sensors; determining the target sensor and the sensor weight corresponding to the target sensor by searching for the associated sensors in the initial sensor data; and updating the sensor weight corresponding to the target sensor based on the weight offset to determine the target sensor data.

[0068] The update step size can be a descriptive information about the amount of data that the sensor weights are adjusted each time.

[0069] Specifically, different associated sensors correspond to different update step sizes. The update step size is determined based on the associated sensors in the association information. The weight offset is determined by multiplying the update step size by the update level corresponding to the associated sensor. The target sensor and its corresponding sensor weight are then determined by searching among the vehicle sensors in the initial sensor data for the associated sensor. The new sensor weight is obtained by adding the target sensor's corresponding sensor weight to the weight offset, thus determining the target sensor data. When the weight offset is positive, the sensor weight increases when the target sensor's weight is updated, indicating good detectability conditions and a significant contribution of the detected information to parking space information decisions. Conversely, when the weight offset is negative, the sensor weight decreases when the target sensor's weight is updated, indicating poor detectability conditions and a small contribution of the detected information to parking space information decisions.

[0070] The update step size is determined based on the associated sensors in the association information; the weight offset is determined based on the update step size and the update level corresponding to the associated sensors; the target sensor and its corresponding sensor weight are determined by searching for the associated sensors in the initial sensor data; the sensor weight corresponding to the target sensor is updated based on the weight offset to determine the target sensor data. The weight of each sensor is determined for different update levels and update step sizes, refining the weight calculation steps and improving the accuracy of the target sensor data determination.

[0071] Optionally, acquiring a real-time scene image of the area to be detected for the vehicle to be detected includes: acquiring the current illumination intensity and illumination intensity threshold corresponding to the current time; comparing the current illumination intensity with the illumination intensity threshold to determine the intensity comparison result; when the intensity comparison result is determined to be that the current illumination intensity is less than the illumination intensity threshold, then the infrared scene image of the area to be detected for the vehicle to be detected is determined as the real-time scene image of the area to be detected for the vehicle to be detected; when the intensity comparison result is determined to be that the current illumination intensity is greater than or equal to the illumination intensity threshold, then the visible light scene image of the area to be detected for the vehicle to be detected is determined as the real-time scene image of the area to be detected for the vehicle to be detected.

[0072] Specifically, the current light intensity and light intensity threshold are obtained at the current time; the current light intensity is compared with the light intensity threshold to determine the intensity comparison result; if the intensity comparison result is that the current light intensity is less than the light intensity threshold, it indicates that the current ambient light conditions are weak and the environmental features of the area to be detected are not clear enough when the visible light image acquisition device is used, then the infrared camera on the vehicle to be detected is used to acquire images, and the infrared scene image of the area to be detected is determined as the real-time scene image of the area to be detected; if the intensity comparison result is that the current light intensity is greater than or equal to the light intensity threshold, it indicates that the current ambient light conditions are good, and the visible light scene image of the area to be detected is acquired by the visible light acquisition device, then the visible light scene image of the area to be detected is determined as the real-time scene image of the area to be detected.

[0073] By acquiring the current light intensity and light intensity threshold corresponding to the current time, comparing the current light intensity with the light intensity threshold, and determining the intensity comparison result, if the intensity comparison result is that the current light intensity is less than the light intensity threshold, then the infrared scene image of the area to be detected of the vehicle to be detected is determined as the real-time scene image of the area to be detected of the vehicle to be detected; if the intensity comparison result is that the current light intensity is greater than or equal to the light intensity threshold, then the visible light scene image of the area to be detected of the vehicle to be detected is determined as the real-time scene image of the area to be detected of the vehicle to be detected. Different acquisition devices are used to acquire images for different lighting environments, thereby improving the accuracy of real-time scene image acquisition.

[0074] Optionally, obtaining at least one scene label from real-time scene image recognition includes: recognizing the real-time scene image and determining image feature data, including pixel features, weather features, and road condition features; obtaining at least one sample feature data and a label identifier corresponding to each sample feature data; determining at least one label identifier based on matching the image feature data with each sample feature data; and determining each label identifier as a scene label.

[0075] Specifically, the acquired real-time scene images are analyzed to extract pixel features, weather features, or road condition features that reflect the essence of the environment, forming structured image feature data. Pixel features (basic visual attributes) focus on the underlying visual characteristics of the image, quantifying basic information such as illumination, color, or sharpness. These can include: Brightness indicators: average grayscale value of the image (0-255, e.g., average value >150 on sunny days, <80 at night), highlight area proportion (number of overexposed pixels / total number of pixels, >30% in backlit scenes). Color distribution: average value of a specific channel in the HSV color space (e.g., blue channel value >200 in sunny sky areas, grayscale channel value >180 for the ground in rainy weather). Sharpness indicators: image gradient value (edge ​​sharpness, gradient value <50 in foggy weather due to blurring, >100 in sunny weather), noise density (percentage of abnormal pixels, a large number of white noise points exist in snowy weather, density >20%). Weather features (environmental interference attributes) identify image texture and morphological features related to weather, determining the presence of interference such as rain, fog, or snow. These include: Rain / fog features: contrast attenuation rate in distant areas (contrast of distant objects <0.3 in foggy weather) or diffuse particle distribution (randomly distributed raindrop reflections in rainy images). Snow features: size and density of white particles (snowflakes 3-5 pixels in diameter, density >15 particles / square inch), ground reflectivity (grayscale value >220 in snowy areas). Clear weather features: uniformity of the sky (standard deviation of the blue channel <30 in cloudless conditions), and recognizability of distant objects (vehicle outlines clearly discernible at 10 meters). Road condition features (parking space related attributes) focus on environmental details directly related to parking space detection, including: parking line features: edge continuity (continuous line segments of complete parking lines >80%), color consistency (pixel value variance of white / yellow parking lines <20%), and occlusion ratio (pixel percentage obscured by fallen leaves / water accumulation; >30% indicates blurriness). Obstacle features: Contour integrity (edge ​​closure > 90%) of static obstacles (parked vehicles or pillars), motion vectors (displacement > 50 pixels over 3 consecutive frames) of dynamic obstacles (pedestrians, moving vehicles). Scene structure: Ground material or surrounding fixed facilities. A feature-label mapping library is established using massive labeled data as a benchmark for real-time image matching. The most matching label is selected by calculating the similarity between real-time image features and sample features. Specifically, feature similarity calculation uses Euclidean distance or cosine similarity algorithms to quantify the difference between the real-time image feature vector and the sample feature vector: the higher the similarity, the closer the real-time scene is to the sample scene. For example, if the average gray value of the real-time image is 170, it has a higher similarity to the features of a sunny daytime sample (gray value 180), and the labels corresponding to the top 3 samples with the highest similarity are selected. If both rainy and sunny days occur simultaneously, the label of the sample with the highest similarity is selected.

[0076] By recognizing real-time scene images, image feature data is determined, including pixel features, weather features, and road condition features. At least one sample feature data and a corresponding label for each sample feature data are obtained. Based on the matching of image feature data and each sample feature data, at least one label is determined. Each label is then used as a scene label. By determining scene labels through multi-dimensional data, the accuracy of scene label acquisition is improved.

[0077] Optionally, the parking space information of the vehicle to be detected is determined based on the detection of the area to be detected by the first sensor and the second sensor, including: obtaining first detection information and a first confidence level based on the detection of the area to be detected by the first sensor; obtaining second detection information and a second confidence level based on the detection of the area to be detected by the second sensor; calculating a target confidence level based on the target sensing data, the first confidence level, and the second confidence level; determining a confidence level comparison result by comparing the target confidence level with a confidence level threshold; and determining the parking space information of the vehicle to be detected based on the confidence level comparison result, the first detection information, and the second detection information.

[0078] Specifically, the first sensor (main sensor) performs core detection on the area to be detected based on its high weight in the current scenario (such as millimeter-wave radar in rainy weather or a camera in sunny weather), obtaining the first detection information: complete core parameters of the parking space, including but not limited to: parking space location, parking space size, parking space type, and occupancy status. The first confidence level (C1): a value between 0 and 1, reflecting the reliability of the main sensor's detection results (e.g., C1=0.9 when the camera clearly identifies the parking space line, C1=0.75 when the millimeter-wave radar detects in rain or fog). The second detection information is supplementary data associated with the first detection information. For example: if the second sensor is millimeter-wave radar: supplementing the distance measurement value of the parking space boundary (e.g., the actual measured width of 2.48 meters) or whether there are static obstacles not identified by the main sensor; if the second sensor is ultrasonic: supplementing the detection results of nearby obstacles within the parking space (e.g., a stone block 0.3 meters from the left boundary). Second confidence level (C2): A value between 0 and 1, reflecting the reliability of the auxiliary detection (e.g., C2=0.85 for ultrasonic close-range detection, C2=0.6 for camera-assisted recognition in low light). Target confidence level is a quantitative indicator combining the reliability of the primary and secondary sensors. It is equal to the sum of the product of the first sensor's first confidence level and its corresponding sensor weight, and the product of the second sensor's second confidence level and its corresponding sensor weight. A preset confidence threshold can be dynamically adjusted according to the scenario. The final parking space information is determined by comparing the target confidence level with the threshold.

[0079] By detecting the area to be detected using the first sensor, first detection information and a first confidence level are obtained; by detecting the area to be detected using the second sensor, second detection information and a second confidence level are obtained; based on the target sensor data, the first confidence level, and the second confidence level, a target confidence level is calculated; by comparing the target confidence level with a confidence level threshold, a confidence level comparison result is determined; based on the confidence level comparison result, the first detection information, and the second detection information, the parking space information of the vehicle to be detected is determined. Different first and second sensors are determined for different environmental conditions corresponding to different areas to be detected, thus refining the accuracy of sensor allocation.

[0080] Optionally, the parking space information of the vehicle to be detected is determined based on the confidence comparison result, the first detection information, and the second detection information, including: when the confidence comparison result is determined to be that the target confidence is greater than the confidence threshold, the first detection information is determined as the parking space information of the vehicle to be detected; when the confidence comparison result is determined to be that the target confidence is equal to the confidence threshold, an alarm message is generated based on the first detection information and the second detection information, and the real-time scene image of the area to be detected of the vehicle to be detected, the initial parking scene type, and the initial sensor data are reacquired; when the confidence comparison result is determined to be that the target confidence is less than the confidence threshold, the parking space information of the vehicle to be detected is determined to be that there is no valid parking space.

[0081] Specifically, if the fusion reliability (target confidence level) of the first and second sensors is significantly higher than the preset confidence threshold, it indicates high consistency of detection results and low error risk. In this case, the first detection information collected by the first sensor is used to determine the parking space information. If the fusion reliability (target confidence level) of the first and second sensors is equal to the preset confidence threshold, it indicates a potential error risk. An alarm message is generated, clearly stating "Detection accuracy is critical, reconfirmation is required," along with the conflict point between the first and second detection information (e.g., "Camera identifies parking space width as 2.4 meters, radar detects as 2.2 meters, error is close to the threshold"). The vehicle can be controlled to move a preset distance towards the parking space for re-detection. If the fusion reliability (target confidence level) of the first and second sensors is less than the preset confidence threshold, it indicates low consistency of detection results. The system clearly returns no valid parking space.

[0082] If the confidence comparison result indicates that the target confidence level is greater than the confidence threshold, then the first detection information is determined as the parking space information of the vehicle to be detected. If the confidence comparison result indicates that the target confidence level is equal to the confidence threshold, then an alarm is generated based on the first and second detection information, and the real-time scene image of the area to be detected, the initial parking scene type, and the initial sensor data of the vehicle to be detected are reacquired. If the confidence comparison result indicates that the target confidence level is less than the confidence threshold, then the parking space information of the vehicle to be detected is determined to be that there is no valid parking space. Different operations are performed for different confidence comparison results, which refines the steps for obtaining parking space information and improves the accuracy of obtaining parking space information.

[0083] Example 3

[0084] Figure 3 This is a schematic diagram of a parking space information determination device provided in Embodiment 3 of the present invention. This embodiment of the present invention is applicable to situations requiring parking space information determination. The device can execute a parking space information determination method and can be implemented in hardware and / or software.

[0085] See Figure 3 The parking space information determination device shown includes: an initial data acquisition module 301, a tag recognition module 302, a target data acquisition module 303, a sensor determination module 304, and a parking space information determination module 305, wherein...

[0086] The initial data acquisition module 301 is used to acquire real-time scene images of the area to be detected of the vehicle to be detected, the initial parking scene type, and the initial sensor data. The initial sensor data includes: at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor.

[0087] The label recognition module 302 is used to recognize at least one scene label from a real-time scene image;

[0088] The target data acquisition module 303 is used to update the initial sensor data according to each scene label and determine the target sensor data;

[0089] The sensor determination module 304 is used to determine the first sensor and the second sensor of the vehicle to be detected based on the target sensing data.

[0090] The parking space information determination module 305 is used to determine the parking space information of the vehicle to be detected based on the detection of the area to be detected by the first sensor and the second sensor.

[0091] The technical solution of this invention acquires real-time scene images of the area to be detected for a vehicle, initial parking scene type, and initial sensor data. The initial sensor data includes at least one vehicle sensor and the sensor weights corresponding to each vehicle sensor. At least one scene label is obtained by recognizing the real-time scene image. The initial sensor data is updated according to each scene label to determine target sensor data. The first and second sensors of the vehicle to be detected are determined according to the target sensor data. The parking space information of the vehicle to be detected is determined based on the detection of the area to be detected by the first and second sensors. Different sensor allocation schemes can be executed for different scenarios, refining the steps for acquiring parking space information and improving the accuracy of determining parking space information.

[0092] Optionally, the target data acquisition module 303 includes:

[0093] The type determination unit is used to identify scene tags for each scene tag and determine the tag type corresponding to the scene tag;

[0094] The association information determination unit is used to find the association information corresponding to the tag type based on the tag type. The association information includes: the associated sensor and the update level of the associated sensor.

[0095] The data update unit is used to update the initial sensor data based on the associated information and determine the target sensor data.

[0096] Optional, data update unit, specifically used for:

[0097] Determine the update step size based on the associated sensors in the associated information;

[0098] The weight offset is determined based on the update step size and the update level corresponding to the associated sensor.

[0099] Based on the search of associated sensors in the initial sensor data, the target sensor and its corresponding sensor weight are determined.

[0100] The target sensor data is determined by updating the sensor weights corresponding to the target sensor based on the weight offset.

[0101] Optionally, the initial data acquisition module 301 is specifically used for:

[0102] Get the current light intensity and light intensity threshold corresponding to the current time;

[0103] Compare the current light intensity with the light intensity threshold to determine the intensity comparison result;

[0104] When the intensity comparison result is determined to be that the current light intensity is less than the light intensity threshold, the infrared scene image of the area to be detected of the vehicle to be detected is determined as the real-time scene image of the area to be detected of the vehicle to be detected.

[0105] When the intensity comparison result is determined to be that the current light intensity is greater than or equal to the light intensity threshold, the visible light scene image of the area to be detected of the vehicle to be detected is determined as the real-time scene image of the area to be detected of the vehicle to be detected.

[0106] Optionally, the tag recognition module 302 is specifically used for:

[0107] For real-time scene image recognition, determine image feature data, which includes pixel features, weather features, and road condition features;

[0108] Obtain at least one sample feature data and the corresponding label for each sample feature data;

[0109] Based on the matching of image feature data with the feature data of each sample, at least one label is determined;

[0110] Each label is identified as a scene label.

[0111] Optionally, the parking space information determination module 305 includes:

[0112] The first information determining unit is used to obtain first detection information and a first confidence level based on the detection of the area to be detected by the first sensor;

[0113] The second information determination unit is used to obtain second detection information and a second confidence level based on the detection of the area to be detected by the second sensor;

[0114] The target information determination unit is used to calculate the target confidence level based on the target sensing data, the first confidence level, and the second confidence level.

[0115] The result determination unit is used to determine the confidence comparison result based on the comparison between the target confidence level and the confidence threshold.

[0116] The parking space data determination unit is used to determine the parking space information of the vehicle to be detected based on the confidence comparison result, the first detection information, and the second detection information.

[0117] Optional, parking space data determination unit, specifically used for:

[0118] When the confidence comparison result is determined to be that the target confidence is greater than the confidence threshold, the first detection information is determined as the parking space information of the vehicle to be detected.

[0119] When the confidence comparison result is determined to be that the target confidence is equal to the confidence threshold, an alarm message is generated based on the first detection information and the second detection information, and the real-time scene image of the detection area of ​​the vehicle to be detected, the initial parking scene type and the initial sensor data are reacquired.

[0120] If the confidence comparison result shows that the target confidence level is less than the confidence threshold, then the parking space information of the vehicle to be detected is determined to be that there are no valid parking spaces.

[0121] The parking space information determination device provided in this embodiment of the invention can execute the parking space information determination method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of executing the parking space information determination method.

[0122] Example 4

[0123] Figure 4 A schematic diagram of the structure of a parking space information determination device 400 that can be used to implement an embodiment of the present invention is shown.

[0124] like Figure 4 As shown, the parking space information determination device 400 includes at least one processor 401 and a memory, such as a read-only memory (ROM) 402 and a random access memory (RAM) 403, communicatively connected to the at least one processor 401. The memory stores computer programs executable by the at least one processor. The processor 401 can perform various appropriate actions and processes based on the computer program stored in the ROM 402 or loaded from storage unit 408 into the RAM 403. The RAM 403 can also store various programs and data required for the operation of the parking space information determination device 400. The processor 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0125] Multiple components in the parking space information determination device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, optical disk, etc.; and a communication unit 409, such as a network card, modem, wireless communication transceiver, etc. The communication unit 409 allows the parking space information determination device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0126] Processor 401 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 401 performs the various methods and processes described above, such as the parking space information determination method.

[0127] In some embodiments, the parking space information determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the parking space information determination device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by processor 401, one or more steps of the parking space information determination method described above may be performed. Alternatively, in other embodiments, processor 401 may be configured to perform the parking space information determination method by any other suitable means (e.g., by means of firmware).

[0128] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0129] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0130] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0131] To provide user interaction, the systems and techniques described herein can be implemented on a parking space information determining device, which includes: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the parking space information determining device. Other types of devices can also be used to provide user interaction; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0132] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0133] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability.

[0134] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0135] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A parking space information determination method characterized by comprising: The method comprises: acquiring a real-time scene image of a to-be-detected area of a to-be-detected vehicle, an initial parking scene type, and initial sensor data, wherein the initial sensor data comprises at least one vehicle sensor and a sensor weight corresponding to each vehicle sensor; identifying at least one scene label from the real-time scene image; updating the initial sensor data according to each scene label to determine target sensor data; determining a first sensor and a second sensor of the to-be-detected vehicle according to the target sensor data; determining parking space information of the to-be-detected vehicle by detecting the to-be-detected area using the first sensor and the second sensor.

2. The method of claim 1, wherein, The updating of the initial sensor data according to each scene label to determine target sensor data comprises: identifying the scene label for each scene label to determine a label type corresponding to the scene label; obtaining associated information corresponding to the label type according to the label type, wherein the associated information comprises an associated sensor and an update level corresponding to the associated sensor; updating the initial sensor data according to the associated information to determine target sensor data.

3. The method of claim 2, wherein, The updating of the initial sensor data according to the associated information to determine target sensor data comprises: determining an update step according to the associated sensor in the associated information; determining a weight offset according to the update step and the update level corresponding to the associated sensor; determining a target sensor and a sensor weight corresponding to the target sensor by searching for the associated sensor in the initial sensor data; updating the sensor weight corresponding to the target sensor according to the weight offset to determine target sensor data.

4. The method of claim 1, wherein, The acquisition of the real-time scene image of the to-be-detected area of the to-be-detected vehicle comprises: acquiring a current light intensity corresponding to a current time and a light intensity threshold; comparing the current light intensity with the light intensity threshold to determine an intensity comparison result; when it is determined that the intensity comparison result is that the current light intensity is less than the light intensity threshold, determining an infrared scene image of the to-be-detected area of the to-be-detected vehicle as the real-time scene image of the to-be-detected area of the to-be-detected vehicle; when it is determined that the intensity comparison result is that the current light intensity is greater than or equal to the light intensity threshold, determining a visible light scene image of the to-be-detected area of the to-be-detected vehicle as the real-time scene image of the to-be-detected area of the to-be-detected vehicle.

5. The method of claim 1, wherein, The identification of at least one scene label from the real-time scene image comprises: identifying image feature data from the real-time scene image, wherein the image feature data comprises pixel features, weather features, and road condition features; acquiring at least one sample feature data and a label identifier corresponding to each sample feature data; determining at least one label identifier according to a match between the image feature data and each sample feature data; determining each label identifier as a scene label.

6. The method of claim 1, wherein, The determination of parking space information of the to-be-detected vehicle by detecting the to-be-detected area using the first sensor and the second sensor comprises: According to the detection of the first sensor on the to-be-detected area, first detection information and a first confidence level are obtained; According to the detection of the second sensor on the to-be-detected area, second detection information and a second confidence level are obtained; According to the target sensor data, the first confidence level and the second confidence level, a target confidence level is calculated; According to the comparison of the target confidence level and the confidence level threshold, a confidence level comparison result is determined; According to the confidence level comparison result, the first detection information and the second detection information, parking space information of the to-be-detected vehicle is determined.

7. The method of claim 6, wherein, The determination of the parking space information of the to-be-detected vehicle according to the confidence level comparison result, the first detection information and the second detection information comprises: When it is determined that the confidence level comparison result is that the target confidence level is greater than the confidence level threshold, the first detection information is determined as the parking space information of the to-be-detected vehicle; When it is determined that the confidence level comparison result is that the target confidence level is equal to the confidence level threshold, alarm information is generated according to the first detection information and the second detection information, and real-time scene images of a to-be-detected area of a to-be-detected vehicle, an initial parking scene type and initial sensor data are re-acquired; When it is determined that the confidence level comparison result is that the target confidence level is less than the confidence level threshold, it is determined that the parking space information of the to-be-detected vehicle is no valid parking space.

8. A parking space information determination apparatus characterized by comprising: The device comprises: An initial data acquisition module, configured to acquire real-time scene images of a to-be-detected area of a to-be-detected vehicle, an initial parking scene type and initial sensor data, wherein the initial sensor data comprises at least one vehicle sensor and a sensor weight corresponding to each vehicle sensor; A label identification module, configured to identify at least one scene label from the real-time scene images; A target data acquisition module, configured to update the initial sensor data according to each scene label to determine target sensor data; A sensor determination module, configured to determine a first sensor and a second sensor of the to-be-detected vehicle according to the target sensor data; A parking space information determination module, configured to determine parking space information of the to-be-detected vehicle according to the detection of the first sensor and the second sensor on the to-be-detected area.

9. A parking space information determination device, characterized by comprising: The parking space information determination device comprises: At least one processor; and A memory in communication connection with the at least one processor; wherein The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the parking space information determination method in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions, and the computer instructions are used to enable the processor to execute the parking space information determination method in any one of claims 1-7 when executed.

Images