Effective parking space identification method, device, equipment and program product

Abstract

The invention discloses an effective parking space identification method, device and equipment and a program product, and relates to the technical field of parking assistance. The method comprises the following steps: acquiring ultrasonic data and a panoramic image around a target vehicle; performing spatial feature extraction on the ultrasonic data, and identifying distance signal neutral gear features in the ultrasonic data; performing feature extraction on the panoramic image, and identifying parking space marking features and obstacle features in the panoramic image; and detecting whether an effective parking space exists around the target vehicle or not according to the distance signal neutral gear characteristics, the parking space marking line characteristics and the obstacle characteristics. According to the technical scheme of the embodiment of the invention, the reliability of effective parking space identification is improved, so that the driving performance of the parking auxiliary system is improved.

Description

Technical Field

[0001] This invention relates to the field of parking assistance technology, and in particular to an effective parking space identification method, device, equipment, and program product. Background Technology

[0002] Parking assistance systems (APA), as a key function to improve driving convenience and safety, have become a focus of technological research and development for OEMs and suppliers. Its core technology lies in using onboard sensors to perceive the environment in real time and accurately identify available parking spaces.

[0003] Currently, publicly available parking assistance systems primarily utilize ultrasonic sensors mounted on the sides of the vehicle to detect the distance to surrounding obstacles while the vehicle is cruising at low speeds. When a continuous and sufficiently long gap in distance is detected, it is identified as a potential parking space. This solution is low-cost and technologically mature, but it relies solely on distance as a single physical quantity, making it susceptible to environmental interference.

[0004] In conclusion, there is an urgent need for a reliable and effective parking space recognition method to improve the driving performance of parking assistance systems. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and program product for identifying effective parking spaces, which improves the reliability of effective parking space identification and thus enhances the driving performance of parking assistance systems.

[0006] According to one aspect of the present invention, an effective parking space identification method is provided, the method comprising:

[0007] Acquire ultrasonic data and panoramic images around the target vehicle;

[0008] Spatial feature extraction is performed on the ultrasonic data to identify distance signal gap features in the ultrasonic data;

[0009] Feature extraction is performed on the panoramic image to identify parking space markings and obstacle features in the panoramic image;

[0010] Based on the distance signal gap feature, the parking space marking feature, and the obstacle feature, detect whether there are valid parking spaces around the target vehicle.

[0011] According to another aspect of the present invention, an effective parking space identification device is provided, the device comprising:

[0012] The target vehicle surrounding data acquisition module is used to acquire ultrasonic data and panoramic images around the target vehicle;

[0013] An ultrasonic data feature extraction module is used to extract spatial features from the ultrasonic data and identify distance signal gap features in the ultrasonic data.

[0014] A panoramic image feature extraction module is used to extract features from the panoramic image and identify parking space marking features and obstacle features in the panoramic image;

[0015] The effective parking space detection module is used to detect whether there are effective parking spaces around the target vehicle based on the distance signal gap feature, the parking space marking feature, and the obstacle feature.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the effective parking space identification method according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the effective parking space identification method according to any embodiment of the present invention.

[0021] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the effective parking space identification method according to any embodiment of the present invention.

[0022] The technical solution of this invention introduces spatial distance features to extract spatial features from ultrasonic data, identifying distance signal gaps in the ultrasonic data. It also introduces visual image features to extract features from panoramic images, identifying parking space markings and obstacle features. Based on these features, it detects whether there are valid parking spaces around the target vehicle. By introducing visual features, it overcomes the inherent limitations of ultrasonic sensors, such as their inability to identify parking space markings and susceptibility to environmental interference. Furthermore, by introducing spatial distance features, it overcomes the environmental adaptability issues of surround-view camera systems, which are easily affected by inclement weather or ambient light. By fusing spatial distance and visual image features, it constructs a more complete discrimination system, transforming the criteria for valid parking space identification from one-dimensional to multi-dimensional. This significantly improves the accuracy of valid parking space identification and judgment results. Furthermore, it allows for complementary advantages between two sensors, maintaining basic judgment capabilities even when one sensor's performance temporarily declines, enhancing robustness and reliability of valid parking space identification. It achieves automated and real-time multi-feature fusion decision-making, requiring no driver intervention. It automatically completes data collection, feature extraction, fusion judgment, and result output, making operation extremely simple and providing a smooth user experience. This improves the efficiency of valid parking space identification and ease of use. In addition, it fully utilizes the ultrasonic sensors and surround-view camera system already standard on vehicles, eliminating the need for new hardware sensors. Functional leaps can be achieved solely through algorithm optimization and system integration, reducing the cost of valid parking space identification and demonstrating high cost-effectiveness and mass production feasibility.

[0023] 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

[0024] 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.

[0025] Figure 1 This is a flowchart of an effective parking space identification method provided in Embodiment 1 of the present invention;

[0026] Figure 2 This is a flowchart of an effective parking space identification method provided in Embodiment 2 of the present invention;

[0027] Figure 3 This is a schematic diagram of the structure of an effective parking space identification device according to Embodiment 2 of the present invention;

[0028] Figure 4 This is a flowchart of an effective parking space identification method provided in Embodiment 2 of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of an effective parking space identification device according to Embodiment 3 of the present invention;

[0030] Figure 6 This is a schematic diagram of the structure of an electronic device that implements the effective parking space identification method of this invention. Detailed Implementation

[0031] 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.

[0032] 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.

[0033] Example 1

[0034] Figure 1 This is a flowchart illustrating a valid parking space identification method provided in Embodiment 1 of the present invention. This embodiment of the invention is applicable to situations where valid parking spaces are identified using a multi-feature fusion approach. The method can be executed by a valid parking space identification device, which can be implemented in hardware and / or software and can be configured in an electronic device that performs valid parking space identification functionality.

[0035] See Figure 1 The effective parking space identification method shown includes:

[0036] S101. Acquire ultrasonic data and panoramic images around the target vehicle.

[0037] The target vehicle can be a vehicle that is currently parking. The area surrounding the target vehicle can be the area covered by the target vehicle's ultrasonic sensors and panoramic imaging system. The ultrasonic data can be ultrasonic data around the target vehicle collected by the target vehicle's ultrasonic sensors. The ultrasonic data is used to characterize the spatial distance characteristics between the target vehicle and surrounding objects. For example, the ultrasonic data can be a continuous, time-varying distance point cloud signal. The panoramic image can be a panoramic image of the area around the target vehicle collected by the target vehicle's surround-view camera system. The panoramic image is used to characterize the visual image characteristics of objects in the target vehicle's surrounding environment.

[0038] Specifically, ultrasonic sensors and their driving circuits, positioned on the sides of the target vehicle, can emit ultrasonic waves and receive the echoes. By measuring the time difference, the distance between the ultrasonic sensors and obstacles around the target vehicle can be calculated, thus obtaining the ultrasonic data of the target vehicle. Alternatively, a panoramic image can be obtained by acquiring real-time image data from the target vehicle's surroundings using an existing surround-view camera system (AVM). For example, a surround-view camera typically includes four fisheye cameras: a front-facing camera, a rear-facing camera, and four cameras positioned under the left and right rearview mirrors.

[0039] S102. Extract spatial features from ultrasonic data and identify distance signal gap features in the ultrasonic data.

[0040] The distance signal neutral point feature can be used to characterize a continuous idle area detected by the ultrasonic sensors of a target vehicle. For example, the distance signal neutral point feature can be a continuous distance signal neutral point that exceeds a preset length threshold. The preset length threshold can be a pre-defined lower limit value for the length used to characterize the distance signal neutral point. For example, the preset length threshold is 1.8 times the vehicle length. For example, the distance signal neutral point feature can include the neutral point length and neutral point width, etc.

[0041] Specifically, gap identification and processing algorithms, such as edge detection algorithms or cluster analysis algorithms, are used to extract spatial features from ultrasonic data, identify continuous gaps in the distance signal that exceed a preset length threshold, and obtain distance signal gap features.

[0042] Optionally, the ultrasonic data can be filtered and denoised before spatial feature extraction to improve the accuracy of distance signal gap identification.

[0043] S103. Extract features from the panoramic image and identify parking space markings and obstacle features in the panoramic image.

[0044] Parking space marking features can be used to represent parking space markings on the ground around a target vehicle. For example, parking space marking features can be used to characterize parking lines on the ground around the target vehicle that are parallel or perpendicular to the vehicle's driving direction. For example, parking space marking features can include the marking length and width, etc. Obstacle features can be used to characterize obstacles around the target vehicle. For example, obstacle features can include vehicle tires and posts, etc. For example, obstacle features can include the type and location of obstacles around the target vehicle.

[0045] Specifically, image recognition algorithms can be used to extract features from panoramic images and identify parking space markings and obstacle features within them. For example, image recognition algorithms may include SIFT (Scale-Invariant Feature Transform) algorithms or template matching algorithms.

[0046] Optionally, before feature extraction from the panoramic image, the panoramic image can be preprocessed, such as by distortion correction and top-view transformation.

[0047] In an optional embodiment of the present invention, feature extraction is performed on the panoramic image to identify parking space marking features and obstacle features in the panoramic image, including: using an image recognition model to extract features from the panoramic image, identifying the integrity of parking space markings, the first position of vertical obstacles, and the second position of unwanted obstacles in the panoramic image; and calculating the first number of vertical obstacles and the second number of unwanted obstacles.

[0048] Image recognition models can be used to extract features from the entire image to obtain the completeness of parking space markings, the first position of vertical obstacles, and the second position of unwanted obstacles in a panoramic image. For example, image recognition models may include histogram of oriented gradients (HOR) and support vector machines (or lightweight convolutional neural networks). The completeness of parking space markings can be used to characterize the degree of integrity of the parking space markings. Vertical obstacles can be obstacles with vertical features around the target vehicle. Unwanted obstacles can be non-standard obstacles around the target vehicle. The first position can be the location of the vertical obstacle. The second position can be the location of the unwanted obstacle. The first quantity can be the number of vertical obstacles in the panoramic image. The second quantity can be the number of unwanted obstacles in the panoramic image.

[0049] In an optional embodiment of the present invention, vertical obstacles include vehicle tires, wall corners, and pillars; undesirable obstacles include parking locks and traffic cones. This solution, by specifying vertical obstacles as vehicle tires, wall corners, and pillars, and undesirable obstacles as parking locks and traffic cones, improves the typicality of vertical and undesirable obstacles, thereby increasing the efficiency of identifying effective parking spaces.

[0050] Specifically, the panoramic image is input into a pre-trained image recognition model to extract features from the panoramic image and identify the integrity of parking markings, the first position of vertical obstacles, and the second position of unwanted obstacles. The identified vertical obstacles and unwanted obstacles can be statistically analyzed to determine the first number of vertical obstacles and the second number of unwanted obstacles.

[0051] This scheme introduces an image recognition model to extract features from panoramic images, improving the efficiency and accuracy of panoramic image recognition. By identifying the integrity of parking space markings, the first position of vertical obstacles, and the second position of unwanted obstacles in panoramic images, the scheme calculates the first number of vertical obstacles and the second number of unwanted obstacles. The inclusion of vertical obstacles and unwanted obstacles further improves the typicality and extraction efficiency of obstacle features.

[0052] S104. Based on the characteristics of the distance signal gap, the characteristics of the parking space markings, and the characteristics of obstacles, detect whether there are valid parking spaces around the target vehicle.

[0053] A valid parking space is a parking space where the target vehicle can park.

[0054] Specifically, a pre-trained parking space recognition model can be used to detect features such as distance signal gaps, parking space markings, and obstacles to determine whether there are valid parking spaces around the target vehicle. For example, the parking space recognition model can be a convolutional neural network model.

[0055] Optionally, preset decision rules can be used to detect features such as distance signal gaps, parking space markings, and obstacles to determine whether there are valid parking spaces around the target vehicle.

[0056] In an optional embodiment of the present invention, spatial features are extracted from the ultrasonic data to identify the gap length and gap width of the distance signal gap in the ultrasonic data; correspondingly, based on the distance signal gap features, parking space marking features, and obstacle features, it is detected whether there are valid parking spaces around the target vehicle, including: detecting whether there are valid parking spaces around the target vehicle based on the gap length and gap width of the distance signal gap, parking space marking features, and obstacle features.

[0057] The gap length can be the length of the distance signal gap in the ultrasonic signal. The gap width can be the width of the distance signal gap in the ultrasonic signal.

[0058] Specifically, a pre-trained parking space recognition model can be used to detect the length and width of the neutral space, the features of the parking space markings, and the features of obstacles to determine whether there are valid parking spaces around the target vehicle.

[0059] Optionally, preset decision rules can be used to detect the neutral length, neutral width, parking space marking features, and obstacle features to determine whether there are valid parking spaces around the target vehicle.

[0060] This solution further enhances the typicality of distance signal gap features by specifying the gap length and gap width, thereby improving the efficiency of identifying effective parking spaces.

[0061] The technical solution of this invention introduces spatial distance features to extract spatial features from ultrasonic data, identifying distance signal gaps in the ultrasonic data. It also introduces visual image features to extract features from panoramic images, identifying parking space markings and obstacle features. Based on these features, it detects whether there are valid parking spaces around the target vehicle. By introducing visual features, it overcomes the inherent limitations of ultrasonic sensors, such as their inability to identify parking space markings and susceptibility to environmental interference. Furthermore, by introducing spatial distance features, it overcomes the environmental adaptability issues of surround-view camera systems, which are easily affected by inclement weather or ambient light. By fusing spatial distance and visual image features, it constructs a more complete discrimination system, transforming the criteria for valid parking space identification from one-dimensional to multi-dimensional. This significantly improves the accuracy of valid parking space identification and judgment results. Furthermore, it allows for complementary advantages between two sensors, maintaining basic judgment capabilities even when one sensor's performance temporarily declines, enhancing robustness and reliability of valid parking space identification. It achieves automated and real-time multi-feature fusion decision-making, requiring no driver intervention. It automatically completes data collection, feature extraction, fusion judgment, and result output, making operation extremely simple and providing a smooth user experience. This improves the efficiency of valid parking space identification and ease of use. In addition, it fully utilizes the ultrasonic sensors and surround-view camera system already standard on vehicles, eliminating the need for new hardware sensors. Functional leaps can be achieved solely through algorithm optimization and system integration, reducing the cost of valid parking space identification and demonstrating high cost-effectiveness and mass production feasibility.

[0062] Example 2

[0063] Figure 2This is a flowchart of an effective parking space identification method provided in Embodiment 2 of the present invention. Based on the above embodiments, this embodiment of the present invention specifies "detecting whether there is an effective parking space around a target vehicle based on distance signal gap features, parking space marking features, and obstacle features" as "performing multi-feature fusion of distance signal gap features, parking space marking features, and obstacle features to obtain a fused feature score; detecting whether there is an effective parking space around the target vehicle based on the fused feature score." This can improve the fusion efficiency of multi-scale feature fusion, thereby improving the detection efficiency of effective parking spaces. It should be noted that parts not described in detail in this embodiment of the present invention can be referred to in the descriptions of other embodiments.

[0064] See Figure 2 The effective parking space identification method shown includes:

[0065] S201. Acquire ultrasonic data and panoramic images around the target vehicle.

[0066] S202. Extract spatial features from ultrasonic data and identify distance signal gap features in the ultrasonic data.

[0067] S203. Extract features from the panoramic image and identify parking space markings and obstacle features in the panoramic image.

[0068] S204. Perform multi-feature fusion on the distance signal gap feature, parking space marking feature and obstacle feature to obtain the fused feature score.

[0069] Feature fusion scoring can be used to comprehensively quantify the spatial geometry and visual semantic information around a target vehicle.

[0070] Specifically, a pre-trained fusion feature scoring detection model can be used to fuse multiple features, including distance signal gap features, parking line features, and obstacle features, and output a fusion feature score. For example, the fusion feature scoring detection model can be any form of classifier, such as a pre-trained logistic regression, decision tree, or a simple if-then-else rule set.

[0071] In an optional embodiment of the present invention, multi-feature fusion is performed on the distance signal gap feature, parking space marking feature and obstacle feature to obtain a fused feature score, including: obtaining a first weight coefficient, a second weight coefficient and a third weight coefficient; using the first weight coefficient, the second weight coefficient and the third weight coefficient, the distance signal gap feature, parking space marking feature and obstacle feature are weighted and summed to obtain a fused feature score.

[0072] The first weighting coefficient corresponds to the distance signal gap characteristic. The second weighting coefficient corresponds to the parking space marking characteristic. The third weighting coefficient corresponds to the obstacle characteristic. The first, second, and third weighting coefficients can be preset and adjusted by technicians based on experimental tests.

[0073] Specifically, a fusion feature score can be obtained by weighting and summing the distance signal gap features, parking space marking features, and obstacle features using a fusion feature score calculation formula:

[0074] ;

[0075] In the formula, S is the fusion feature score; a is the first weight coefficient; b represents the distance signal gap characteristic; b is the second weighting coefficient. The parking space marking features are represented by c, which is the third weighting coefficient. These are characteristics of obstacles.

[0076] This scheme uses a first weighting coefficient, a second weighting coefficient, and a third weighting coefficient to perform a weighted summation of distance signal gap features, parking space marking features, and obstacle features to obtain a fusion feature score, thereby improving the computational efficiency of the fusion feature score and thus improving the identification efficiency of effective parking spaces.

[0077] S205. Based on the fusion feature score, detect whether there are valid parking spaces around the target vehicle.

[0078] Specifically, a preset scoring threshold can be obtained. The fused feature score is compared with the preset scoring threshold. If the fused feature score is greater than or equal to the preset scoring threshold, it is determined that the target vehicle is located in a valid parking space; if the fused feature score is less than the preset scoring threshold, it is determined that the target vehicle is located in a valid parking space.

[0079] The technical solution of this invention fuses multiple features, including distance signal gap features, parking space marking features, and obstacle features, to obtain a fused feature score. Based on the fused feature score, it detects whether there are valid parking spaces around the target vehicle. This can improve the fusion efficiency of multi-scale feature fusion, thereby improving the detection efficiency of valid parking spaces.

[0080] Figure 3 This is a schematic diagram of an effective parking space recognition system. (Example) Figure 3 As shown, the system includes a first sensing module, a second sensing module, and a data processing module (i.e., this device).

[0081] The first sensing module consists of ultrasonic sensors and their driving circuitry positioned on the side of the target vehicle. Its function is to emit ultrasonic waves and receive the echoes, calculating the distance to obstacles by measuring the time difference. This module outputs a continuous, time-varying distance point cloud signal.

[0082] The second perception module is the vehicle's existing surround-view camera system (AVM). It typically includes four fisheye cameras: a front-facing camera, a rear-facing camera, and cameras positioned under the left and right side mirrors. Its function is to collect real-time image data of the vehicle's surroundings.

[0083] The data processing module is the core of this system. It can be a dedicated processing unit (such as a DSP (Digital Signal Processor), GPU (Graphics Processing Unit), or NPU (Neural Network Processing Unit)) within the APA (Auto Parking Assist Controller), and its signal input terminal is connected to the two sensing modules mentioned above through an in-vehicle network (such as CAN (Controller Area Network) or Ethernet).

[0084] Figure 4 This is a flowchart of an effective parking space identification method. For example... Figure 4 As shown, the data processing module is configured to perform the following operations:

[0085] S401. Simultaneously acquire raw ultrasound data and AVM images.

[0086] The raw ultrasonic data refers to the ultrasonic data surrounding the target vehicle. The AVM image is a panoramic image of the area around the target vehicle.

[0087] S402, Extract spatial dimension features.

[0088] Specifically, the ultrasonic signal (i.e., ultrasonic data) is processed. First, filtering and noise reduction are performed. Then, a continuous distance signal gap exceeding a preset length threshold (e.g., 1.8 times the vehicle length) is identified, and the gap length is calculated. ) and gap depth ( The neutral length and neutral depth can be used to characterize the distance between the target vehicle and surrounding obstacles. For example, the neutral length can be the average length of the distance signal neutral. The neutral depth can be the average depth of the distance signal neutral.

[0089] S403, Extract visual dimension features.

[0090] Specifically, the AVM image undergoes preprocessing (such as distortion correction and top-view transformation). Then, a pre-trained image recognition model (e.g., based on HOG+SVM or a lightweight CNN network) is used to identify and extract specific features from the AVM image. These features include: parking space marking features, vertical obstacle features, and unwanted obstacle features. Among these, parking space marking features are used to characterize parallel or perpendicular parking lines on the ground. Specifically, parking space marking features can be categorized as the completeness of the parking space markings (…). Vertical obstacle features are used to characterize objects with vertical features, such as vehicle tires, wall corners, and pillars, framed within a panoramic image. Specifically, vertical obstacle features include the location and number of vertical obstacles. Undesired obstacle features are used to characterize non-standard obstacles such as ground locks and cones.

[0091] S404, weighted fusion calculation.

[0092] Specifically, the data processing module includes a decision logic unit (i.e., a fusion feature score calculation unit) that stores preset weight coefficients (e.g., first weight coefficient, second weight coefficient, and third weight coefficient) and corresponding preset decision rules. This unit receives all feature values ​​(including distance signal gap features, parking space marking features, and vertical obstacle features) from the feature extraction unit (including the ultrasonic data feature extraction module and the panoramic image feature extraction module), and substitutes them into a weighted fusion formula (i.e., the fusion feature score calculation formula) for calculation, outputting a comprehensive confidence score (S), i.e., the fusion feature score.

[0093] For example, the weighted fusion formula can be represented by the following formula:

[0094] ;

[0095] Where S is the confidence score; These are weighting coefficients that have been calibrated through numerous experiments; This is the length of the gap; This refers to the gap depth; Features of parking space markings; This is a characteristic of vertical obstacles.

[0096] This fusion calculation process is the key to this solution. It is no longer a simple judgment of "a vacant space means a parking space", but comprehensively considers spatial geometry and visual semantic information. A real parking space usually shows as "a long vacant space + clear parking lines + no obstructive paths or obstacles within the parking space", and its fusion feature score S will be very high. While an interfering vacant space (such as a small gap between two vehicles) may have "a long vacant space but no parking lines", or a parking space with a ground lock may have "enough space but interference from unexpected obstacles", and its fusion feature score S will be very low, thus directly achieving the effect of improving accuracy and reducing false alarms.

[0097] S405. Compare the calculated confidence score S with a preset scoring threshold T.

[0098] S406. If S ≥ T, it is determined that there is a valid parking space, and the result (including the position and type of the valid parking space) is output to the APA controller to trigger subsequent parking planning.

[0099] S407. If S < T, it is determined that there is no valid parking space, and the system can choose to ignore it or prompt the driver.

[0100] Exemplarily, assume that the target vehicle slowly drives past a gap (not a parking space) between two parked vehicles.

[0101] The first perception module detects a continuous distance signal vacant space feature that is 5 meters long (i.e., very large), initially meeting the parking space length requirement.

[0102] The panoramic image captured by the surround-view camera of the second perception module, after visual dimension feature extraction, finds that there are no parking markings on the ground ( ).

[0103] Although there are vehicles on both sides (with vertical obstacle features), a ground lock is identified as a non-standard obstacle at the end of the distance signal vacant space.

[0104] The decision logic unit performs weighted fusion calculation:

[0105] ;

[0106] In the formula, S is the confidence score; is the weight coefficient calibrated through a large number of experiments; is the product; "5" is the vacant space length; is the vacant space depth; "0" is the parking marking feature; is the vertical obstacle feature; is the non-standard obstacle feature.

[0107] The final calculated confidence score S, lacking visual feature support and containing negative features, will be lower than the preset scoring threshold T. Therefore, the gap in the distance signal is ultimately determined to be an invalid parking space, successfully avoiding a false identification.

[0108] In existing technologies, although ultrasonic waves can be used for initial screening, followed by visual information to assist in verifying the identified parking spaces, this approach often employs simple AND / OR logic or sequential judgment. For example, when a non-parking space (such as the gap between two cars) is identified by ultrasonic waves, if the visual verification module fails to clearly capture situations like the absence of parking lines due to camera angles or lighting conditions, the system may still misclassify it as a parking space. Moreover, in inclement weather, the signal-to-noise ratio of ultrasonic sensors decreases, and the image quality of the camera also degrades significantly. In this case, both sensor sources may become unreliable, leading to overall system performance degradation or even failure. Furthermore, for parking spaces without vertical obstacles, the response characteristics of ultrasonic sensors are very weak, making independent identification almost impossible, necessitating the use of vision as the primary sensor. However, once vision fails due to environmental factors such as lighting and weather conditions, the entire system's ability to identify parking spaces is immediately lost.

[0109] This solution effectively distinguishes between real parking spaces and interfering gaps, significantly improving the recognition rate of spatial and line parking spaces in various complex environments, while greatly reducing the false alarm rate. This enhances the reliability and user trust of the APA system. By introducing spatial distance features, spatial feature extraction is performed on ultrasonic data to identify distance signal gaps. Visual image features are also introduced to extract features from panoramic images, identifying parking space markings and obstacle features. Based on these features, the system detects whether there are valid parking spaces around the target vehicle. Introducing visual features overcomes the inherent limitations of ultrasonic sensors, such as their inability to recognize parking space markings and susceptibility to environmental interference. Furthermore, introducing spatial distance features overcomes the environmental adaptability issues of surround-view camera systems, such as susceptibility to adverse weather or ambient light. By fusing spatial distance and visual image features, a comprehensive system is formed. A more complete discrimination system transforms the criteria for identifying valid parking spaces from one-dimensional to multi-dimensional, significantly improving the accuracy of identification and judgment. Furthermore, it allows for complementary use of two sensors, maintaining basic judgment capabilities even when one sensor temporarily degrades, enhancing robustness and reliability. It achieves automated and real-time multi-feature fusion decision-making, requiring no driver intervention. The system automatically completes data collection, feature extraction, fusion judgment, and result output, offering extremely simple operation and a smooth user experience, improving both the efficiency of valid parking space identification and ease of use. Moreover, it fully utilizes the vehicle's standard ultrasonic sensors and surround-view camera system, eliminating the need for new hardware sensors. Functional leaps are achieved solely through algorithm optimization and system integration, reducing the cost of valid parking space identification and demonstrating high cost-effectiveness and mass production feasibility.

[0110] Example 3

[0111] Figure 5 This is a schematic diagram of a valid parking space identification device provided in Embodiment 3 of the present invention. This embodiment of the invention is applicable to situations where valid parking spaces are identified using a multi-feature fusion method. The device can execute a valid parking space identification method and can be implemented in hardware and / or software. The device can be configured in an electronic device that carries the valid parking space identification function.

[0112] See Figure 5The effective parking space identification device shown includes: a target vehicle surrounding data acquisition module 501, an ultrasonic data feature extraction module 502, a panoramic image feature extraction module 503, and an effective parking space detection module 504. Specifically, the target vehicle surrounding data acquisition module 501 acquires ultrasonic data and a panoramic image around the target vehicle; the ultrasonic data feature extraction module 502 extracts spatial features from the ultrasonic data and identifies distance signal gap features in the ultrasonic data; the panoramic image feature extraction module 503 extracts features from the panoramic image and identifies parking space marking features and obstacle features in the panoramic image; and the effective parking space detection module 504 detects whether there are effective parking spaces around the target vehicle based on the distance signal gap features, the parking space marking features, and the obstacle features.

[0113] The technical solution of this invention introduces spatial distance features to extract spatial features from ultrasonic data, identifying distance signal gaps in the ultrasonic data. It also introduces visual image features to extract features from panoramic images, identifying parking space markings and obstacle features. Based on these features, it detects whether there are valid parking spaces around the target vehicle. By introducing visual features, it overcomes the inherent limitations of ultrasonic sensors, such as their inability to identify parking space markings and susceptibility to environmental interference. Furthermore, by introducing spatial distance features, it overcomes the environmental adaptability issues of surround-view camera systems, which are easily affected by inclement weather or ambient light. By fusing spatial distance and visual image features, it constructs a more complete discrimination system, transforming the criteria for valid parking space identification from one-dimensional to multi-dimensional. This significantly improves the accuracy of valid parking space identification and judgment results. Furthermore, it allows for complementary advantages between two sensors, maintaining basic judgment capabilities even when one sensor's performance temporarily declines, enhancing robustness and reliability of valid parking space identification. It achieves automated and real-time multi-feature fusion decision-making, requiring no driver intervention. It automatically completes data collection, feature extraction, fusion judgment, and result output, making operation extremely simple and providing a smooth user experience. This improves the efficiency of valid parking space identification and ease of use. In addition, it fully utilizes the ultrasonic sensors and surround-view camera system already standard on vehicles, eliminating the need for new hardware sensors. Functional leaps can be achieved solely through algorithm optimization and system integration, reducing the cost of valid parking space identification and demonstrating high cost-effectiveness and mass production feasibility.

[0114] In an optional embodiment of the present invention, the effective parking space detection module 504 includes: a fusion feature scoring calculation unit, used to perform multi-feature fusion on the distance signal gap feature, the parking space marking feature and the obstacle feature to obtain a fusion feature score; and a first effective parking space detection unit, used to detect whether there is an effective parking space around the target vehicle based on the fusion feature score.

[0115] In an optional embodiment of the present invention, the fusion feature scoring calculation unit includes: a weight coefficient acquisition subunit, used to acquire a first weight coefficient, a second weight coefficient, and a third weight coefficient; wherein the first weight coefficient corresponds to the distance signal gap feature; the second weight coefficient corresponds to the parking space marking feature; and the third weight coefficient corresponds to the obstacle feature; the fusion feature scoring calculation subunit is used to use the first weight coefficient, the second weight coefficient, and the third weight coefficient to perform a weighted summation of the distance signal gap feature, the parking space marking feature, and the obstacle feature to obtain a fusion feature score.

[0116] In an optional embodiment of the present invention, the ultrasonic data feature extraction module 502 includes: an ultrasonic data feature extraction unit, used to extract spatial features from the ultrasonic data and identify the gap length and gap width of the distance signal gap in the ultrasonic data; correspondingly, the effective parking space detection module 504 includes: a second effective parking space detection unit, used to detect whether there is an effective parking space around the target vehicle based on the gap length and gap width of the distance signal gap, the parking space marking features, and the obstacle features.

[0117] In an optional embodiment of the present invention, the panoramic image feature extraction module 503 includes: a first panoramic image feature extraction unit, used to extract features from the panoramic image using an image recognition model, and to identify the integrity of parking space markings, the first position of vertical obstacles, and the second position of unwanted obstacles in the panoramic image; and a second panoramic image feature extraction unit, used to calculate the first number of vertical obstacles and the second number of unwanted obstacles.

[0118] In an alternative embodiment of the invention, the vertical obstacles include vehicle tires, wall corners, and pillars; the undesirable obstacles include ground locks and traffic cones.

[0119] In an optional embodiment of the present invention, after detecting whether there is a valid parking space around the target vehicle based on the distance signal gap feature, the parking space marking feature, and the obstacle feature, the device further includes: a parking planning triggering module, used to feed back the valid parking space to the parking assistance main controller and trigger parking planning when there is a valid parking space around the target vehicle; and a prompt information feedback module, used to provide prompt information to the driver of the vehicle when there is a valid parking space around the target vehicle.

[0120] The effective parking space identification device provided in this embodiment of the invention can execute the effective parking space identification method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0121] In the technical solutions of this invention, the acquisition, storage, and application of ultrasonic data, panoramic images, first weighting coefficients, second weighting coefficients, and third weighting coefficients around the target vehicle all comply with relevant laws and regulations and do not violate public order and good morals.

[0122] Example 4

[0123] Figure 6 A schematic diagram of an electronic device 600 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0124] like Figure 6 As shown, the electronic device 600 includes at least one processor 601 and a memory, such as a read-only memory (ROM) 602 or a random access memory (RAM) 603, communicatively connected to the at least one processor 601. The memory stores computer programs executable by the at least one processor. The processor 601 can perform various appropriate actions and processes based on the computer program stored in the ROM 602 or loaded into the RAM 603 from storage unit 608. The RAM 603 may also store various programs and data required for the operation of the electronic device 600. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0125] Multiple components in electronic device 600 are connected to I / O interface 605, including: input unit 606, such as keyboard, mouse, etc.; output unit 607, such as various types of displays, speakers, etc.; storage unit 608, such as disk, optical disk, etc.; and communication unit 609, such as network card, modem, wireless transceiver, etc. Communication unit 609 allows electronic device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0126] Processor 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 601 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, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 601 performs the various methods and processes described above, such as the effective parking space recognition method.

[0127] In some embodiments, the valid parking space identification method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded into and / or installed on electronic device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by processor 601, one or more steps of the valid parking space identification method described above may be performed. Alternatively, in other embodiments, processor 601 may be configured to perform the valid parking space identification 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 interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: 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 electronic device. Other types of devices can also be used to provide interaction with the user; 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. An effective parking space identification method, characterized in that, The method includes: Acquire ultrasonic data and panoramic images around the target vehicle; Spatial feature extraction is performed on the ultrasonic data to identify distance signal gap features in the ultrasonic data; Feature extraction is performed on the panoramic image to identify parking space markings and obstacle features in the panoramic image; Based on the distance signal gap feature, the parking space marking feature, and the obstacle feature, detect whether there are valid parking spaces around the target vehicle.

2. The method according to claim 1, characterized in that, The step of detecting whether there are valid parking spaces around the target vehicle based on the distance signal gap feature, the parking space marking feature, and the obstacle feature includes: Multi-feature fusion is performed on the distance signal gap feature, the parking space marking feature, and the obstacle feature to obtain a fused feature score; Based on the fusion feature score, it is determined whether there are any available parking spaces around the target vehicle.

3. The method according to claim 2, characterized in that, The process of fusing multiple features—the distance signal gap feature, the parking space marking feature, and the obstacle feature—to obtain a fused feature score includes: Obtain a first weighting coefficient, a second weighting coefficient, and a third weighting coefficient; wherein, the first weighting coefficient corresponds to the distance signal gap feature; the second weighting coefficient corresponds to the parking space marking feature; and the third weighting coefficient corresponds to the obstacle feature; The distance signal gap feature, the parking space marking feature, and the obstacle feature are weighted and summed using the first weighting coefficient, the second weighting coefficient, and the third weighting coefficient to obtain a fusion feature score.

4. The method according to claim 1, characterized in that, The step of extracting spatial features from the ultrasonic data and identifying distance signal gap features in the ultrasonic data includes: Spatial feature extraction is performed on the ultrasonic data to identify the gap length and gap width of the distance signal gap in the ultrasonic data; Accordingly, detecting whether there are valid parking spaces around the target vehicle based on the distance signal gap feature, the parking space marking feature, and the obstacle feature includes: Based on the distance signal gap length, gap width, parking space marking features, and obstacle features, the system detects whether there are valid parking spaces around the target vehicle.

5. The method according to claim 1, characterized in that, The step of extracting features from the panoramic image and identifying parking space markings and obstacle features in the panoramic image includes: An image recognition model is used to extract features from the panoramic image and identify the integrity of parking space markings, the first position of vertical obstacles, and the second position of unwanted obstacles in the panoramic image. Calculate the first number of vertical obstacles and the second number of undesirable obstacles.

6. The method according to claim 5, characterized in that, The vertical obstacles include vehicle tires, wall corners, and pillars; the undesirable obstacles include ground locks and traffic cones.

7. The method according to claim 1, characterized in that, After detecting whether there are valid parking spaces around the target vehicle based on the distance signal gap feature, the parking space marking feature, and the obstacle feature, the method further includes: When there are available parking spaces around the target vehicle, the available parking spaces are fed back to the parking assistance main controller to trigger parking planning; When there are available parking spaces around the target vehicle, a notification message is sent to the driver.

8. An effective parking space identification device, characterized in that, The device includes: The target vehicle surrounding data acquisition module is used to acquire ultrasonic data and panoramic images around the target vehicle; An ultrasonic data feature extraction module is used to extract spatial features from the ultrasonic data and identify distance signal gap features in the ultrasonic data. A panoramic image feature extraction module is used to extract features from the panoramic image and identify parking space marking features and obstacle features in the panoramic image; The effective parking space detection module is used to detect whether there are effective parking spaces around the target vehicle based on the distance signal gap feature, the parking space marking feature, and the obstacle feature.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the effective parking space identification method according to any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the valid parking space identification method according to any one of claims 1-7.

Images