A parking space fusion method, system, storage medium and vehicle for an APA parking scenario

Abstract

This invention discloses a parking space fusion method for APA parking scenarios, comprising the steps of: using an onboard ultrasonic sensor to identify the current parking space and determine the effective ultrasonic parking space; using an onboard surround-view camera to identify the parking space and determine the effective visual parking space; and fusing the ultrasonic and visual parking spaces based on the overlap ratio, graphic distance, and shape similarity relationship between the ultrasonic and visual parking spaces to obtain and output the fused parking space for the current parking space. This invention also discloses a corresponding system, storage medium, and vehicle. Implementing embodiments of this invention can quickly obtain accurate fused parking spaces, improving the success rate and efficiency of parking.

Description

Technical Field

[0001] This invention relates to the field of automatic parking assist (APA) technology, and in particular to a parking space fusion method, system, storage medium and vehicle for APA parking scenarios. Background Technology

[0002] Currently, the main parking space fusion methods in APA parking scenarios are: using Around View Monitor (AVM) sensors (hereinafter referred to as surround view sensors) to perceive the parking environment and search for visual parking spaces, and using Ultrasonic Radar (USS) sensors to perceive the parking environment and search for ultrasonic parking spaces. Then, the two parking space information are fused to obtain fused parking space information.

[0003] However, the logic of the fusion methods used in existing technologies is relatively simple. For example, in some cases, when fusing visual parking spaces and ultrasonic parking spaces, the method simply compares the magnitude of their parking space coordinate information to select one as the final output fused parking space, without fully utilizing the information of their overlapping parts, and without considering the error of ultrasonic parking spaces.

[0004] Meanwhile, in some cases, ultrasonic parking space detection is used first to determine valid parking spaces. However, the number of ultrasonic information points is small, and the detection of the surrounding boundary is not stable enough, making it impossible to accurately identify parking spaces. The error fluctuation range is about 30cm. At the same time, ultrasonic detection distance is limited compared to visual detection. This means that if ultrasonic parking space detection is used first to determine valid parking spaces, some valid parking spaces will be filtered out.

[0005] Therefore, the accuracy of existing technologies for fusing ultrasonic parking space and visual parking space is not good, which will affect the success rate and efficiency of parking. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a parking space fusion method, system, storage medium and vehicle for APA parking scenarios, which can quickly obtain accurate fused parking spaces and improve the success rate and efficiency of parking.

[0007] To address the aforementioned technical problems, as one aspect of the present invention, a parking space fusion method for APA parking scenarios is provided, which includes at least the following steps:

[0008] The vehicle uses onboard ultrasonic sensors to identify available parking spaces and determine valid ultrasonic parking spaces.

[0009] The vehicle-mounted surround-view camera is used to identify the parking spaces and determine the effective visual parking spaces.

[0010] Based on the overlap ratio, graphic distance, and shape similarity between the ultrasonic parking space and the visual parking space, the ultrasonic parking space and the visual parking space are fused to obtain and output the fused parking space of the current parking space.

[0011] The step of using an onboard ultrasonic sensor to identify the current parking space and determine the effective ultrasonic parking space further includes:

[0012] The vehicle uses an onboard ultrasonic sensor to identify the current parking space and obtain the coordinate information of the current parking space. The coordinate information includes at least the coordinate information of each vertex and the length and width information.

[0013] When the size of the parking space meets the preset minimum parking size and there are no obstacles in it, the identified parking space is determined as a valid ultrasonic parking space.

[0014] The step of using an onboard surround-view camera to identify parking spaces and determine effective visual parking spaces further includes:

[0015] The vehicle-mounted surround-view camera is used to identify the parking space and obtain the coordinate information of the current parking space. The coordinate information includes at least the coordinate information of each vertex and the length and width information.

[0016] When the size of the parking space meets the preset minimum parking size and there are no obstacles in it, the identified parking space is determined as a valid visual parking space.

[0017] The step of fusing the ultrasonic parking space and the visual parking space based on the overlap ratio, graphic distance, and shape similarity relationship between the two, to obtain and output the fused parking space for the current parking space, further includes:

[0018] Based on the coordinate information of the effective ultrasonic parking space and visual parking space, the relative position information of the two and their respective effective parking areas are calculated. The relative position information includes: the distance between the center points of the two and the diagonal of the minimum bounding box of the two.

[0019] Based on the relative position information and their respective effective parking areas, the correction ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space is calculated using the improved IOU algorithm formula.

[0020] The correction ratio is compared with a pre-calibrated threshold. If the correction ratio is greater than the threshold, either the ultrasonic parking space or the visual parking space is taken as the current fused parking space; otherwise, the overlapping area of ​​the ultrasonic parking space and the visual parking space is taken as the current fused parking space.

[0021] Output the type and location information of the current merged parking space.

[0022] The step of calculating the ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space using the improved IOU algorithm formula based on the relative position information and their respective effective parking areas further includes:

[0023] The corrected ratio CIOU is calculated using the following improved IOU algorithm formula:

[0024]

[0025]

[0026]

[0027]

[0028] Where ρ(P1, P2) is the Euclidean distance between the center points of the ultrasonic parking space P1 and the visual parking space P2; c is the diagonal length of the minimum bounding box containing P1 and P2; S1 is the effective parking area of ​​the ultrasonic parking space; S2 is the effective parking area of ​​the visual parking space; α is the weighting function; v is the function that measures the consistency of the aspect ratio; P1W and P1H are the width and height of the ultrasonic parking space P1, respectively; and P2W and P2H are the width and height of the ultrasonic parking space P2, respectively.

[0029] Accordingly, another aspect of the present invention also provides a parking space fusion system for APA parking scenarios, which includes at least:

[0030] The ultrasonic parking space recognition unit is used to identify the parking space using the vehicle's onboard ultrasonic sensor and determine the valid ultrasonic parking space.

[0031] A visual parking space recognition unit is used to identify the parking spaces using an onboard surround-view camera to determine valid visual parking space spaces.

[0032] The fusion processing unit is used to perform fusion processing on the ultrasonic parking space and the visual parking space based on the overlap ratio, graphic distance, and shape similarity relationship between the two, to obtain and output the fused parking space of the current parking space.

[0033] The fusion processing unit further includes:

[0034] The parameter acquisition unit is used to calculate the relative position information and the effective parking area of ​​the ultrasonic parking space and the visual parking space based on the effective ultrasonic parking space and the visual parking space. The relative position information includes: the distance between the center points of the two and the diagonal of the minimum bounding box of the two.

[0035] The calculation unit is used to calculate the correction ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space based on the relative position information and their respective effective parking areas using an improved IOU (intersection and union ratio) algorithm.

[0036] The comparison processing unit is used to compare the correction ratio with a pre-calibrated threshold. When the correction ratio is greater than the threshold, either the ultrasonic parking space or the visual parking space is taken as the current fused parking space; otherwise, the overlapping area of ​​the ultrasonic parking space and the visual parking space is taken as the current fused parking space.

[0037] The output unit is used to output the type and location information of the current fused parking space.

[0038] The calculation unit calculates the corrected ratio CIOU using the following improved IOU algorithm formula:

[0039]

[0040]

[0041]

[0042]

[0043] Where ρ(P1, P2) is the Euclidean distance between the center points of the ultrasonic parking space P1 and the visual parking space P2; c is the diagonal length of the minimum bounding box containing P1 and P2; S1 is the effective parking area of ​​the ultrasonic parking space; S2 is the effective parking area of ​​the visual parking space; α is the weighting function; v is the function that measures the consistency of the aspect ratio; P1W and P1H are the width and height of the ultrasonic parking space P1, respectively; and P2W and P2H are the width and height of the ultrasonic parking space P2, respectively.

[0044] Accordingly, in another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method as described above.

[0045] Accordingly, another aspect of the present invention also provides a vehicle having at least an APA parking device, wherein the system described above is deployed in the APA parking device.

[0046] Implementing the embodiments of the present invention has the following beneficial effects:

[0047] This invention provides a parking space fusion method, system, storage medium, and vehicle for APA parking scenarios. It identifies ultrasonic parking space and visual parking space using ultrasonic and visual sensors, respectively. Then, an improved IOU algorithm is used to comprehensively consider the overlap ratio, graphic distance, and shape similarity between ultrasonic and line parking spaces to determine the appropriate parking space fusion strategy, ultimately obtaining the final fused parking space information. This method combines the advantages of ultrasonic and visual recognition, resulting in a more accurate and stable fusion strategy, thereby improving the success rate of APA parking.

[0048] The improved IOU algorithm used in this embodiment of the invention incorporates factors such as the graphic distance and shape similarity of the two parking space spaces into the existing IOU algorithm. It can accurately fit and calculate quantitative data on the overlap between the ultrasonic parking space space and the visual parking space space, thereby making the final fused parking space more accurate, robust, and generalizable.

[0049] The method of this invention can be applied to various APA-integrated parking scenarios, thereby effectively improving the success rate of parking space detection and parking efficiency, while avoiding the waste of parking space resources. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, obtaining other drawings based on these drawings without creative effort still falls within the scope of the present invention.

[0051] Figure 1 This is a schematic diagram of the main flow of an embodiment of a parking space fusion method for APA parking scenarios provided by the present invention;

[0052] Figure 2 This invention relates to a schematic diagram illustrating the relative relationship between ultrasonic parking space and visual parking space in an APA parking scenario.

[0053] Figure 3 for Figure 2 A schematic diagram of the corresponding parameters;

[0054] Figure 4 This is a schematic diagram of a parking space fusion system for APA parking scenarios provided by the present invention.

[0055] Figure 5 for Figure 4 A schematic diagram of the fusion processing unit. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.

[0057] like Figure 1 The diagram shows the main flow of an embodiment of a parking space fusion method for APA parking scenarios provided by the present invention; combined with... Figures 2 to 3 As shown, in this embodiment, the method includes at least the following steps:

[0058] Step S10: Use the vehicle's onboard ultrasonic sensor to identify the current parking space and determine the effective ultrasonic parking space.

[0059] In a specific example, step S10 further includes:

[0060] The vehicle uses an onboard ultrasonic sensor to identify the current parking space and obtain the coordinate information of the current parking space. The coordinate information includes at least the coordinate information of each vertex and the length and width information. It can be understood that the parking space type can also be identified in this step, such as horizontal parking space, vertical parking space and angled parking space.

[0061] When the size of the parking space meets the preset minimum parking size and there are no obstacles in it, the identified parking space is determined as a valid ultrasonic parking space.

[0062] For example Figure 2 This illustrates a scenario where a horizontal parking space is identified. In this example, an onboard ultrasonic sensor identifies the current parking space, determining its spatial parameters (such as width, depth, and any obstacles). The identified ultrasonic parking space P1 includes four vertices A, B, M, and D, with coordinates A(X, Y, M) and D(X, Y, D). a Y a ), B(X) b Y b ), M(X) m Y m ), D(X d Y dIn the specific calculation process, the coordinates of points M and D are generally calculated from points A and B. Having obtained the coordinates of these four vertices, the parking space area S1 of P1, as well as the specific values ​​of its width (P1W) and height (P1H), can then be calculated for subsequent fusion processing.

[0063] Step S11: Use the vehicle-mounted surround view camera to identify the parking space and determine the effective visual parking space.

[0064] In a specific example, step S11 further includes:

[0065] The vehicle-mounted surround-view camera is used to identify the parking space and obtain the coordinate information of the current parking space. The coordinate information includes at least the coordinate information of each vertex and the length and width information.

[0066] When the size of the parking space meets the preset minimum parking size and there are no obstacles in it, the identified parking space is determined as a valid visual parking space.

[0067] Continue to refer to Figure 2 For example, an in-vehicle surround-view camera is used to identify the parking spaces, such as identifying image features like the corner points of the parking spaces. The identified visual parking space P2 includes four vertices E, F, G, and H, with coordinates E(X) and H(H). e Y e ), F(X) f Y f ), G(X) g Y g ), H(X) h Y h In the specific calculation process, the coordinates of points G and H are generally calculated from points E and F. Having obtained the coordinates of these four vertices, the parking space area S2 of P2, as well as the specific values ​​of its width (P2W) and height (P2H), can then be calculated for subsequent fusion processing.

[0068] Step S12: Based on the overlap ratio, graphic distance, and shape similarity relationship between the ultrasonic parking space and the visual parking space, perform fusion processing on the ultrasonic parking space and the visual parking space to obtain and output the fused parking space of the current parking space.

[0069] In a specific example, step S12 further includes:

[0070] Step S120: Based on the coordinate information of the effective ultrasonic parking space and visual parking space, calculate the relative position information of the two spaces and their respective effective parking areas. The relative position information includes: the distance between the center points of the two spaces and the diagonal length of the minimum outer frame of the two spaces.

[0071] Among them, such as Figure 2 As shown, the distance between the center points of the two spaces is the Euclidean distance between the center point o' of the ultrasonic parking space P1 and the center point o of the visual parking space P2; the minimum bounding box of the two spaces is the space enclosed by points A', B, D, and M', and the length of its diagonal is the length between points M' and B; at the same time, in this example, area S1 is the area of ​​rectangle ABDM, P1W is the length of AB, and P1H is the length of AM; while area S2 is the area of ​​rectangle EFHG, P2W is the length of EF, and P2H is the length of EG.

[0072] Step S121: Based on the relative position information and their respective effective parking areas, calculate the correction ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space using the improved IOU algorithm formula.

[0073] In one example, the corrected ratio CIOU can be calculated using the following improved IOU algorithm formula:

[0074]

[0075]

[0076]

[0077]

[0078] Where ρ(P1, P2) is the Euclidean distance between the center points of the ultrasonic parking space P1 and the visual parking space P2; c is the diagonal length of the minimum bounding box containing P1 and P2 (i.e., the length of M'B); S1 is the effective parking area of ​​the ultrasonic parking space; S2 is the effective parking area of ​​the visual parking space; α is the weighting function; v is the function that measures the consistency of the aspect ratio; P1W and P1H are the width and height of the ultrasonic parking space P1, respectively, and P2W and P2H are the width and height of the ultrasonic parking space P2, respectively; IOU is the ratio of the intersection and union of the parking areas of the ultrasonic parking space and the visual parking space. Figure 3 In the equation, S1∩S2 is the area of ​​the space enclosed by points E', F, H, and G'; while S1∪S2 is the area of ​​the space enclosed by points E, E', A, B, D, M, G', and G.

[0079] Step S122: Compare the correction ratio with a pre-calibrated threshold. When the correction ratio is greater than the threshold, it indicates a high degree of overlap between the two. In this case, either the ultrasonic parking space or the visual parking space is taken as the current fused parking space. (Refer to...) Figure 2 In the scenario shown, P1 (ABDM) or P2 (EGFH) can be used as the current merged parking space.

[0080] Otherwise, it indicates that the two are significantly different and the visual parking space needs to be corrected based on the ultrasonic parking space space, and the corrected visual parking space will be used as the current fused parking space.

[0081] Specifically, the overlapping area of ​​the ultrasonic parking space and the visual parking space can be used as the current fused parking space, referring to... Figure 2 As shown, the space enclosed by points E', F, H, and G' is taken as the current merged parking space;

[0082] It is understandable that the threshold here can be obtained through pre-calibration. For example, in one case, several different thresholds can be determined based on prior knowledge. The accuracy of APA parking success can be tested under different thresholds, and the average of the thresholds corresponding to the top few (such as the top five) accuracy rates can be obtained to obtain the threshold required in this invention.

[0083] Step S123: Output the type and location information of the current fused parking space.

[0084] It is understood that in this invention, the modified ratio CIOU is obtained by improving upon IOU. The improvement lies in introducing the diagonal distance of the minimum bounding box, adding the distance between the center points of the ultrasonic parking space and the visual parking space, and considering the consistency of their aspect ratios. A larger CIOU value indicates a higher degree of overlap between the two frames. By introducing the CIOU value, the degree of overlap between the ultrasonic and visual parking spaces can be accurately determined. In this invention, by incorporating CIOU into the calculation of overlap in fused parking spaces within a parking scenario, the overlap between the ultrasonic and visual parking spaces can be determined more accurately. Furthermore, CIOU has strong generalization capabilities and can simultaneously cover fused parking scenarios involving multiple types of parking spaces.

[0085] like Figure 4 The diagram shown illustrates a structural schematic of an embodiment of a parking space fusion system for APA parking scenarios provided by the present invention. (In conjunction with...) Figure 5 As shown, in this embodiment, the system 1 includes at least:

[0086] The ultrasonic parking space recognition unit 10 is used to identify the parking space of the current parking space using the vehicle-mounted ultrasonic sensor and determine the effective ultrasonic parking space.

[0087] The visual parking space recognition unit 11 is used to identify the parking space using the vehicle-mounted surround view camera and determine the effective visual parking space.

[0088] The fusion processing unit 12 is used to perform fusion processing on the ultrasonic parking space and the visual parking space according to the overlap ratio, graphic distance and shape similarity relationship between the ultrasonic parking space and the visual parking space, to obtain the fused parking space of the current parking space and output it.

[0089] like Figure 5 As shown, the fusion processing unit 12 further includes:

[0090] The parameter acquisition unit 120 is used to calculate the relative position information of each space and the effective parking area of ​​each space based on the coordinate information of the effective ultrasonic parking space and visual parking space. The relative position information includes: the distance between the center points of the two spaces and the diagonal distance between the minimum outer frames of the two spaces.

[0091] The calculation unit 121 is used to calculate the correction ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space based on the relative position information and their respective effective parking areas using the improved IOU algorithm formula.

[0092] The calculation unit calculates the corrected ratio CIOU using the following improved IOU algorithm formula:

[0093]

[0094]

[0095]

[0096]

[0097] Where ρ(P1, P2) is the Euclidean distance between the center points of the ultrasonic parking space P1 and the visual parking space P2; c is the diagonal length of the minimum bounding box containing P1 and P2; S1 is the effective parking area of ​​the ultrasonic parking space; S2 is the effective parking area of ​​the visual parking space; α is the weighting function; v is the function that measures the consistency of the aspect ratio; P1W and P1H are the width and height of the ultrasonic parking space P1, respectively; and P2W and P2H are the width and height of the ultrasonic parking space P2, respectively.

[0098] The comparison processing unit 122 is used to compare the correction ratio with a pre-calibrated threshold. When the correction ratio is greater than the threshold, either the ultrasonic parking space or the visual parking space is taken as the current fused parking space; otherwise, the overlapping area of ​​the ultrasonic parking space and the visual parking space is taken as the current fused parking space.

[0099] Output unit 123 is used to output the type and location information of the current fused parking space.

[0100] For more details, please refer to and combine with the above. Figures 1 to 3 The description of that will not be repeated here.

[0101] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements as follows: Figures 1 to 3 The steps of the described method. For more details, please refer to and combine with the foregoing descriptions. Figures 1 to 3 The description of that will not be repeated here.

[0102] In another aspect, the present invention provides a vehicle having at least one APA parking device, wherein the APA parking device is configured with such... Figures 4 to 5 The described parking space fusion system is for APA parking scenarios. For more details, please refer to and combine with the aforementioned... Figures 4 to 5 The description of that will not be repeated here.

[0103] Implementing the embodiments of the present invention has the following beneficial effects:

[0104] This invention provides a parking space fusion method, system, storage medium, and vehicle for APA parking scenarios. It identifies ultrasonic parking space and visual parking space using ultrasonic and visual sensors, respectively. Then, an improved IOU algorithm is used to comprehensively consider the overlap ratio, graphic distance, and shape similarity between ultrasonic and line parking spaces to determine the appropriate parking space fusion strategy, ultimately obtaining the final fused parking space information. This method combines the advantages of ultrasonic and visual recognition, resulting in a more accurate and stable fusion strategy, thereby improving the success rate of APA parking.

[0105] The improved IOU algorithm used in this embodiment of the invention incorporates factors such as the graphic distance and shape similarity of the two parking space spaces into the existing IOU algorithm. It can accurately fit and calculate quantitative data on the overlap between the ultrasonic parking space space and the visual parking space space, thereby making the final fused parking space more accurate, robust, and generalizable.

[0106] The method of this invention can be applied to various APA-integrated parking scenarios, thereby effectively improving the success rate of parking space detection and parking efficiency, while avoiding the waste of parking space resources.

[0107] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0109] The above description is merely a preferred embodiment of the present invention and should not be construed as limiting the scope of the invention. Therefore, any equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A parking space fusion method for APA parking scenarios, characterized in that, It should include at least the following steps: The vehicle uses onboard ultrasonic sensors to identify available parking spaces and determine valid ultrasonic parking spaces. The vehicle-mounted surround-view camera is used to identify the parking spaces and determine the effective visual parking spaces. Based on the overlap ratio, graphic distance, and shape similarity between the ultrasonic parking space and the visual parking space, the ultrasonic parking space and the visual parking space are fused to obtain and output the fused parking space for the current parking space. This step further includes the following steps: Based on the coordinate information of the effective ultrasonic parking space and visual parking space, the relative position information of the two spaces and their respective effective parking areas are calculated. The relative position information includes: the distance between the center points of the two spaces and the diagonal length of the minimum bounding box of the two spaces. Based on the relative position information and their respective effective parking areas, the correction ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space is calculated using the improved IOU algorithm formula. The correction ratio is compared with a pre-calibrated threshold. If the correction ratio is greater than the threshold, either the ultrasonic parking space or the visual parking space is taken as the current fused parking space; otherwise, the overlapping area of ​​the ultrasonic parking space and the visual parking space is taken as the current fused parking space. Output the type and location information of the current merged parking space.

2. The method of claim 1, wherein, The step of using an onboard ultrasonic sensor to identify the current parking space and determine the effective ultrasonic parking space further includes: The vehicle uses an onboard ultrasonic sensor to identify the current parking space and obtain the coordinate information of the current parking space. The coordinate information includes at least the coordinate information of each vertex of the parking space and its length and width. When the size of the parking space meets the preset minimum parking size and there are no obstacles in it, the identified parking space is determined as a valid ultrasonic parking space.

3. The method of claim 1, wherein, The step of using an onboard surround-view camera to identify parking spaces and determine effective visual parking spaces further includes: The vehicle-mounted surround-view camera is used to identify the parking space and obtain the coordinate information of the current parking space. The coordinate information includes at least the coordinate information of each vertex of the parking space and its length and width. When the size of the parking space meets the preset minimum parking size and there are no obstacles in it, the identified parking space is determined as a valid visual parking space.

4. The method according to any one of claims 1 to 3, characterized in that, The step of calculating the correction ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space using the improved IOU algorithm formula based on the relative position information and their respective effective parking areas further includes: The corrected ratio CIOU is calculated using the following improved IOU algorithm formula: ； in, S1 is the Euclidean distance between the center points of the ultrasonic parking space P1 and the visual parking space P2; c is the diagonal length of the smallest bounding box containing P1 and P2; S1 is the effective parking area of ​​the ultrasonic parking space; S2 is the effective parking area of ​​the visual parking space. is the weighting function, v is the function that measures the consistency of the aspect ratio; P1W and P1H are the width and height of the ultrasonic parking space P1, respectively, and P2W and P2H are the width and height of the ultrasonic parking space P2, respectively.

5. A parking space fusion system for an APA parking scenario, characterized by, At least including: The ultrasonic parking space recognition unit is used to identify the parking space using the vehicle's onboard ultrasonic sensor and determine the valid ultrasonic parking space. A visual parking space recognition unit is used to identify the parking spaces using an onboard surround-view camera to determine valid visual parking space spaces. The fusion processing unit is used to perform fusion processing on the ultrasonic parking space and the visual parking space according to the overlap ratio, graphic distance and shape similarity relationship between the two, to obtain the fused parking space of the current parking space and output it. The fusion processing unit further includes: The parameter acquisition unit is used to calculate the relative position information of each space and the effective parking area of ​​each space based on the coordinate information of the effective ultrasonic parking space and visual parking space. The relative position information includes: the distance between the center points of the two spaces and the diagonal distance between the minimum outer frames of the two spaces. The calculation unit is used to calculate the correction ratio of the intersection area and the merged area between the ultrasonic parking space and the visual parking space based on the relative position information and their respective effective parking areas using the improved IOU algorithm formula. The comparison processing unit is used to compare the correction ratio with a pre-calibrated threshold. When the correction ratio is greater than the threshold, either the ultrasonic parking space or the visual parking space is taken as the current fused parking space; otherwise, the overlapping area of ​​the ultrasonic parking space and the visual parking space is taken as the current fused parking space. The output unit is used to output the type and location information of the current fused parking space.

6. The system of claim 5, wherein, The calculation unit calculates the corrected ratio CIOU using the following improved IOU algorithm formula: ； in, S1 is the Euclidean distance between the center points of the ultrasonic parking space P1 and the visual parking space P2; c is the diagonal length of the smallest bounding box containing P1 and P2; S1 is the effective parking area of ​​the ultrasonic parking space; S2 is the effective parking area of ​​the visual parking space. is the weighting function, v is the function that measures the consistency of the aspect ratio; P1W and P1H are the width and height of the ultrasonic parking space P1, respectively, and P2W and P2H are the width and height of the ultrasonic parking space P2, respectively.

7. A computer readable storage medium having stored thereon a computer program, characterized in that When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 4.

8. A vehicle having at least one APA parking device thereon, wherein the APA parking device is equipped with a parking space fusion system for APA parking scenarios as described in any one of claims 5 to 6.

Images