Pedestrian and cyclist target filtering method, device and equipment, storage medium and vehicle

Abstract

This disclosure relates to a method, apparatus, device, storage medium, and vehicle for filtering pedestrian and cyclist targets. It involves acquiring images captured by an acquisition device and the calibration parameters of the acquisition device at the time of image acquisition, obtaining image recognition results, including multiple pedestrian detection boxes and multiple cyclist detection boxes, and acquiring matching results between the multiple cyclist detection boxes and the multiple pedestrian detection boxes. The matching results include pedestrian targets and cyclist targets. The location and distance information of each target in the matching results are determined based on the calibration parameters of the acquisition device at the time of image acquisition. Based on the matching results, the confidence level of the detection boxes corresponding to each target in the matching results, and the location and distance information of each target in the matching results, the targets in the matching results are filtered out. This can filter out targets that are not of interest or are useless, reduce the computational load on the vehicle, improve the vehicle's operating speed, and thus reduce the difficulty of subsequent target tracking and speed / distance measurement.

Description

Technical Field

[0001] This disclosure relates to the field of autonomous driving technology, and in particular to a method, apparatus, device, storage medium, and vehicle for filtering pedestrian and cyclist targets. Background Technology

[0002] With the development of autonomous driving technology, autonomous vehicles will gradually be promoted in people's daily lives. In order to reduce the incidence of safety accidents and alleviate driver fatigue, Advanced Driving Assistance Systems (ADAS) have become a research hotspot.

[0003] In car driving scenarios, pedestrians and cyclists are two important targets that ADAS needs to identify quickly and accurately.

[0004] In related technologies, many pedestrians or cyclists identified are irrelevant or useless targets, increasing the computational load on the vehicle during subsequent processing, affecting its operating speed, and thus increasing the difficulty of subsequent target tracking, speed measurement, and distance measurement. Therefore, there is an urgent need for a method to filter out irrelevant or useless targets. Summary of the Invention

[0005] To address the aforementioned technical problems, this disclosure provides a method, apparatus, device, storage medium, and vehicle for filtering pedestrian and cyclist targets.

[0006] In a first aspect, embodiments of this disclosure provide a method for filtering pedestrian and cyclist targets, the method comprising:

[0007] Acquire images captured by the acquisition device and the calibration parameters of the acquisition device when the images were acquired;

[0008] Obtain the recognition results of the image, which include multiple pedestrian detection boxes and multiple cyclist detection boxes;

[0009] Obtain the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets;

[0010] The location and distance information of each target in the matching result are determined based on the calibration parameters of the acquisition device when the image is acquired.

[0011] Based on the matching results, the confidence of the detection box corresponding to each target in the matching results, and the location and distance information of each target in the matching results, the targets in the matching results are filtered.

[0012] Secondly, embodiments of this disclosure provide a pedestrian / cyclist target filtering device, comprising:

[0013] The first acquisition module is used to acquire the image acquired by the acquisition device and the calibration parameters of the acquisition device when the image was acquired.

[0014] The second acquisition module is used to acquire the recognition results of the image, the recognition results including multiple pedestrian detection boxes and multiple cyclist detection boxes;

[0015] The third acquisition module is used to acquire the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets;

[0016] The determination module is used to determine the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image is acquired;

[0017] The filtering module is used to filter the targets in the matching result based on the matching result, the confidence of the detection box corresponding to each target in the matching result, and the position information and distance information of each target in the matching result.

[0018] Thirdly, embodiments of this disclosure provide an electronic device, including:

[0019] Memory;

[0020] Processor; and

[0021] Computer programs;

[0022] The computer program is stored in the memory and configured to be executed by the processor to implement the method as described in the first aspect.

[0023] Fourthly, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method described in the first aspect.

[0024] Fifthly, embodiments of this disclosure also provide a computer program product, which includes a computer program or instructions that, when executed by a processor, implement the pedestrian and cyclist target filtering method as described above.

[0025] Sixthly, embodiments of this disclosure also provide a vehicle, including:

[0026] Memory;

[0027] Processor; and

[0028] Computer programs;

[0029] The computer program is stored in the memory and configured to be executed by the processor to implement the pedestrian and cyclist target filtering method as described above.

[0030] The pedestrian and cyclist target filtering method, apparatus, device, storage medium, and vehicle provided in this disclosure acquire images captured by an acquisition device and calibration parameters of the acquisition device when the images were acquired. The recognition results of the images are obtained, including multiple pedestrian detection boxes and multiple cyclist detection boxes. Matching results between the multiple cyclist detection boxes and the multiple pedestrian detection boxes are obtained, including pedestrian and cyclist targets. The position and distance information of each target in the matching results are determined based on the calibration parameters of the acquisition device when the images were acquired. Based on the matching results, the confidence level of the detection boxes corresponding to each target in the matching results, and the position and distance information of each target in the matching results, the targets in the matching results are filtered out. This can filter out targets that are unnecessary or useless, thereby reducing interfering targets, reducing the computational load on the vehicle, increasing the vehicle's operating speed, and thus reducing the difficulty of subsequent target tracking and speed / distance measurement. Attached Figure Description

[0031] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0032] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0033] Figure 1 A flowchart of a pedestrian and cyclist target filtering method provided in this embodiment of the disclosure;

[0034] Figure 2 A flowchart of a pedestrian and cyclist target filtering method provided in another embodiment of this disclosure;

[0035] Figure 3 A flowchart of a pedestrian and cyclist target filtering method provided in another embodiment of this disclosure;

[0036] Figure 4 A flowchart of a pedestrian and cyclist target filtering method provided in another embodiment of this disclosure;

[0037] Figure 5 A schematic diagram of the pedestrian and cyclist target filtering device provided in an embodiment of this disclosure;

[0038] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0039] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0040] Numerous specific details are set forth in the following description to provide a thorough understanding of this disclosure; however, this disclosure may also be implemented in other ways different from those described herein. Clearly, the embodiments described in the specification are only a part of, and not all, of the embodiments of this disclosure. The specific embodiments described herein are merely illustrative of the invention and not intended to limit it. All other embodiments obtained by those skilled in the art based on the described embodiments of the invention are within the scope of protection of this invention.

[0041] This disclosure provides a method for filtering pedestrian and cyclist targets, which will be described below with reference to specific embodiments.

[0042] Figure 1 This is a flowchart of a pedestrian and cyclist target filtering method provided in an embodiment of this disclosure. This method can be applied to vehicle-mounted terminals and can be used in scenarios where unwanted or useless targets are filtered from identified pedestrian or cyclist targets. It can reduce the computational load on the vehicle-mounted terminal, increase its operating speed, and thus reduce the difficulty of subsequent target tracking and speed / distance measurement. It is understood that the pedestrian and cyclist target filtering method provided in this disclosure can also be applied to other scenarios.

[0043] The following is about Figure 1 The pedestrian and cyclist target filtering method shown is introduced below. The specific steps of this method are as follows:

[0044] S101. Acquire the image acquired by the acquisition device and the calibration parameters of the acquisition device when the image was acquired.

[0045] The vehicle-mounted terminal acquires images captured by the acquisition device. For example, if the vehicle is equipped with a forward-looking camera or a surround-view camera, the images can be acquired by the forward-looking camera or the surround-view camera. The vehicle-mounted terminal can acquire the images captured by the acquisition device and the calibration parameters of the acquisition device when the images were acquired.

[0046] S102. Obtain the recognition result of the image, the recognition result including multiple pedestrian detection boxes and multiple cyclist detection boxes.

[0047] After acquiring multiple images, the vehicle terminal obtains the recognition results of the images, which include multiple pedestrian detection boxes and multiple cyclist detection boxes.

[0048] In some embodiments, the vehicle-mounted terminal recognizes the image to obtain a recognition result. Further, the vehicle-mounted terminal acquires the recognition result of the image, which includes multiple pedestrian detection boxes and multiple cyclist detection boxes.

[0049] In this embodiment, the vehicle-mounted terminal recognizes the image to obtain the recognition result. For each pedestrian, a pedestrian detection box and its confidence level are identified; for each cyclist, a pedestrian detection box, a cyclist detection box, their confidence levels are identified. The detection boxes and their confidence levels correspond one-to-one; that is, the detection boxes and their confidence levels appear in pairs.

[0050] In some embodiments, an image can be input into a convolutional neural network to obtain multiple pedestrian detection boxes, the confidence score of each pedestrian detection box, multiple cyclist detection boxes, and the confidence score of each cyclist detection box.

[0051] S103. Obtain the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets.

[0052] After obtaining the recognition result of the image, the matching result of the multiple cyclist detection boxes and the multiple pedestrian detection boxes is obtained, and the matching result includes pedestrian targets and cyclist targets.

[0053] In some embodiments, the vehicle-mounted terminal matches the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes to obtain a matching result, which includes pedestrian targets and cyclist targets. Since a single cyclist image will identify two detection boxes: one for pedestrians and one for cyclists, matching the pedestrian and cyclist detection boxes is necessary to reduce interfering targets and facilitate subsequent target tracking and speed / distance measurement. Further, the vehicle-mounted terminal acquires the matching result of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, which includes pedestrian targets and cyclist targets.

[0054] S104. Determine the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image is acquired.

[0055] After obtaining the matching results, the location and distance information of each target in the matching results are determined according to the calibration parameters of the acquisition device when the image was acquired. Specifically, the location and distance information of each pedestrian target and each cyclist target can be determined according to the calibration parameters of the acquisition device when the image was acquired. In some embodiments, the location and distance information of each target can be determined according to the intrinsic and extrinsic parameters of the acquisition device.

[0056] S105. Filter the targets in the matching results based on the matching results, the confidence of the detection box corresponding to each target in the matching results, and the position and distance information of each target in the matching results.

[0057] In this step, the vehicle-mounted terminal filters the targets in the matching results based on the target type, the confidence level of the detection box corresponding to each target, and the location and distance information of each target. Target filtering can be performed from multiple perspectives, including target type, the confidence level of the detection box corresponding to each target, and the location and distance information of each target, to further reduce interfering targets and facilitate subsequent target tracking and speed / distance measurement.

[0058] This embodiment of the disclosure acquires images captured by an acquisition device and the calibration parameters of the acquisition device when the images were acquired, and obtains the recognition results of the images. The recognition results include multiple pedestrian detection boxes and multiple cyclist detection boxes. The matching results of the multiple cyclist detection boxes and the multiple pedestrian detection boxes are obtained. The matching results include pedestrian targets and cyclist targets. The position information and distance information of each target in the matching results are determined according to the calibration parameters of the acquisition device when the images were acquired. Based on the matching results, the confidence of the detection boxes corresponding to each target in the matching results, and the position information and distance information of each target in the matching results, the targets in the matching results are filtered out. Targets that are not of interest or are useless can be filtered out, thereby reducing interfering targets, reducing the computational load on the vehicle, increasing the running speed of the vehicle, and thus reducing the difficulty of subsequent target tracking and speed and distance measurement.

[0059] Figure 2 Here is a flowchart of a pedestrian and cyclist target filtering method provided in another embodiment of this disclosure, as shown below. Figure 2 As shown, the method includes the following steps:

[0060] S201. Obtain the image acquired by the acquisition device and the calibration parameters of the acquisition device when the image was acquired.

[0061] Specifically, the implementation process and principle of S201 and S101 are the same, and will not be repeated here.

[0062] S202. Obtain the recognition result of the image, the recognition result including multiple pedestrian detection boxes and multiple cyclist detection boxes.

[0063] Specifically, the implementation process and principle of S202 and S102 are the same, and will not be repeated here.

[0064] S203. Match the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes.

[0065] After obtaining multiple cyclist detection boxes and multiple pedestrian detection boxes, the vehicle-mounted terminal matches the multiple cyclist detection boxes and the multiple pedestrian detection boxes. In some embodiments, the vehicle-mounted terminal matches the cyclist detection box and the target pedestrian detection box based on the maximum height of the detection boxes between the pedestrian detection box and the cyclist detection box, the ratio of their heights, the difference between their bottom edge ordinates, and the difference between their bottom edge center point abscissas.

[0066] S204. If the cyclist detection box and the pedestrian detection box match successfully, then the matching result is determined to be the cyclist target.

[0067] If the cyclist detection box and the pedestrian detection box match successfully, the vehicle terminal determines that the matching result is the cyclist target.

[0068] S205. If the pedestrian detection box is not matched, then the matching result is determined to be a pedestrian target.

[0069] If the pedestrian detection box is not matched, the vehicle terminal determines that the matching result is a pedestrian target.

[0070] S206. Based on the calibration parameters of the acquisition device and the pixel coordinates of the midpoint of the bottom edge of the detection frame corresponding to the target, the midpoint of the bottom edge of the detection frame corresponding to the target is transformed into the vehicle coordinate system to obtain the coordinate information of the target in the vehicle coordinate system.

[0071] For example, the calibration parameters of the acquisition device can be used to determine the transformation relationship between pixel coordinates and the vehicle coordinate system. Furthermore, the midpoint of the bottom edge of the detection box corresponding to the target (i.e., the position of the target in the pixel coordinate system) can be transformed to the vehicle coordinate system to obtain the coordinate information of the target in the vehicle coordinate system. The vehicle coordinate system is a coordinate system with the vehicle's location as the origin.

[0072] S207. Calculate the position information and distance information of the target based on the coordinate information of the target in the vehicle coordinate system.

[0073] After obtaining the target's coordinates in the vehicle coordinate system, the vehicle's location is determined, with the origin as the coordinate system. Based on the coordinates of two points, the position and distance between them can be calculated. In other words, the target's position and distance information can be calculated based on its coordinates in the vehicle coordinate system.

[0074] S208. Based on the matching result, the confidence level of the detection box corresponding to each target in the matching result, and the position and distance information of each target in the matching result, determine whether the target is a target to be filtered.

[0075] In this step, the vehicle terminal determines whether the target is a target to be filtered based on multiple factors, including the matching result, the confidence level of the detection box corresponding to each target in the matching result, and the position and distance information of each target in the matching result.

[0076] In some embodiments, the targets to be filtered include edge targets, invalid targets, incomplete targets, and distant targets; correspondingly, determining whether a target is a target to be filtered includes: determining whether the target is an edge target; or determining whether the target is an invalid target; or determining whether the target is an incomplete target; or determining whether the target is a distant target.

[0077] Specifically, the vehicle terminal determines whether the target is an edge target, an invalid target, an incomplete target, or a distant target. If any one of these determinations is successful, the target is identified as a target to be filtered.

[0078] S209. If the target is a target to be filtered, then the target to be filtered is filtered.

[0079] If the vehicle terminal determines that the target is a target to be filtered, then the target to be filtered is filtered to reduce the amount of computation on the vehicle and increase the running speed of the vehicle, thereby reducing the difficulty of subsequent target tracking and speed and distance measurement.

[0080] This embodiment of the disclosure obtains the recognition result of the image by acquiring an image captured by an acquisition device and the calibration parameters of the acquisition device when the image was acquired. The recognition result includes multiple pedestrian detection boxes and multiple cyclist detection boxes. The multiple cyclist detection boxes and the multiple pedestrian detection boxes are matched. If the cyclist detection box and the pedestrian detection box match successfully, the matching result is determined to be a cyclist target; if the pedestrian detection box is not matched, the matching result is determined to be a pedestrian target. Further, based on the calibration parameters of the acquisition device and the pixel coordinates of the midpoint of the bottom edge of the detection box corresponding to the target, the midpoint of the bottom edge of the detection box corresponding to the target is transformed to the vehicle coordinate system to obtain the coordinate information of the target in the vehicle coordinate system. The position information and distance information of the target are calculated based on the coordinate information of the target in the vehicle coordinate system. Then, based on the matching result, the confidence of the detection box corresponding to each target in the matching result, and the position information and distance information of each target in the matching result, it is determined whether the target is a target to be filtered. The targets to be filtered include edge targets, invalid targets, incomplete targets, and distant targets. If the target is a target to be filtered, then the target to be filtered is filtered. Since the determination of whether a target is a target to be filtered is based on the matching results, the confidence level of the detection box corresponding to each target in the matching results, and the position and distance information of each target in the matching results, if the target is a target to be filtered, then the target to be filtered is filtered. This makes target filtering more comprehensive and flexible, thereby reducing interfering targets, reducing the computational load on the vehicle, and increasing the operating speed of the vehicle, thus reducing the difficulty of subsequent target tracking and speed / distance measurement.

[0081] Figure 3 Here is a flowchart of a pedestrian and cyclist target filtering method provided in another embodiment of this disclosure, as shown below. Figure 3 As shown, the method includes the following steps:

[0082] S301. Acquire the image acquired by the acquisition device and the calibration parameters of the acquisition device when the image was acquired.

[0083] Specifically, the implementation process and principle of S301 and S101 are the same, and will not be repeated here.

[0084] S302. Obtain the recognition result of the image, the recognition result including multiple pedestrian detection boxes and multiple cyclist detection boxes.

[0085] Specifically, the implementation process and principle of S302 and S102 are the same, and will not be repeated here.

[0086] S303. Obtain the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets.

[0087] Specifically, the implementation process and principle of S303 and S103 are the same, and will not be repeated here.

[0088] S304. Based on the preset height of the target, the height of the detection box corresponding to the target, the focal length of the acquisition device, and the calibration parameters of the acquisition device, a transformation matrix from the image coordinate system to the vehicle coordinate system is obtained.

[0089] Let the target p have coordinates (X, Y, Z) in the vehicle coordinate system and (X0, Y0) in the image coordinate system. Let the focal length of the acquisition device be f, where Z is the preset height of the target and Z0 is the height of the detection box corresponding to the target. Based on the pinhole imaging model, using the principle of triangular similarity, we can obtain Z / f = X / X0 = Y / Y0. Further, combined with the calibration parameters of the acquisition device, we can obtain the transformation matrix from the image coordinate system to the vehicle coordinate system.

[0090] It should be noted that the detection box can be a rectangular detection box, with the height of the detection box being the length of the horizontal side of the rectangle and the width of the detection box being the length of the vertical side of the rectangle.

[0091] S305. Based on the transformation matrix from the image coordinate system to the vehicle coordinate system and the coordinates of the midpoint of the bottom edge of the detection box corresponding to the target in the image coordinate system, the coordinate information of the target in the vehicle coordinate system is obtained.

[0092] After obtaining the transformation matrix from the image coordinate system to the vehicle coordinate system, the midpoint of the bottom edge of the detection box corresponding to the target is transformed from the image coordinate system to the vehicle coordinate system based on the transformation matrix, thereby obtaining the coordinate information of the target in the vehicle coordinate system.

[0093] S306. Calculate the position information and distance information of the target based on the coordinate information of the target in the vehicle coordinate system.

[0094] After obtaining the target's coordinates in the vehicle coordinate system, the vehicle's location is determined, with the origin as the coordinate system. Based on the coordinates of two points, the position and distance between them can be calculated. In other words, the target's position and distance information can be calculated based on its coordinates in the vehicle coordinate system.

[0095] S307. Based on the matching result, the confidence level of the detection box corresponding to each target in the matching result, and the position information and distance information of each target in the matching result, determine whether the target is an edge target. If so, execute S312; otherwise, execute S308.

[0096] In this step, the vehicle-mounted terminal determines whether the target is an edge target based on multiple factors, including the matching result, the confidence level of the detection box corresponding to each target in the matching result, and the position and distance information of each target in the matching result. If the target is an edge target, step S312 is executed; if the target is not an edge target, steps S308 and the steps following S308 are executed.

[0097] S308. Determine whether the target is invalid. If so, execute S312; otherwise, execute S309.

[0098] If the target is not an edge target, then determine whether the target is an invalid target. If the target is an invalid target, then execute S312; if the target is not an invalid target, then execute S309 and the steps following S309.

[0099] S309. Determine whether the target is an incomplete target. If so, execute S312; otherwise, execute S310.

[0100] If the target is not an invalid target, then determine whether the target is an incomplete target. If the target is an incomplete target, then execute S312; if the target is not an incomplete target, then execute S310 and the steps following S310.

[0101] It should be noted that S307, S308, and S309 are parallel and there is no specific order of judgment. The embodiments disclosed in this disclosure are for illustrative purposes only and do not limit the order of judgment. For example, one could first determine whether the target is an invalid target, then determine whether the target is an incomplete target, and finally determine whether the target is a marginal target. Alternatively, one could first determine whether the target is an invalid target, then determine whether the target is a marginal target, and finally determine whether the target is an incomplete target, and so on.

[0102] S310. Determine whether the target is a distant target. If yes, execute S312; otherwise, execute S311.

[0103] If the target is not an incomplete target, then determine whether the target is a distant target. If the target is a distant target, then execute S312; if the target is not a distant target, then execute S311.

[0104] S311. Do not filter the target.

[0105] If the vehicle terminal determines that the target is not a distant target, then the target will not be filtered.

[0106] S312. Filter the target to be filtered.

[0107] The vehicle-mounted terminal filters the targets to be filtered. Specifically, the vehicle-mounted terminal filters edge targets, invalid targets, incomplete targets, and distant targets to reduce the computational load on the vehicle and increase the operating speed of the vehicle, thereby reducing the difficulty of subsequent target tracking and speed and distance measurement.

[0108] This embodiment of the disclosure obtains the recognition result of the image by acquiring an image captured by an acquisition device and the calibration parameters of the acquisition device when the image was acquired. The recognition result includes multiple pedestrian detection boxes and multiple cyclist detection boxes. Matching results between the multiple cyclist detection boxes and the multiple pedestrian detection boxes are obtained, and the matching results include pedestrian targets and cyclist targets. Further, based on the preset height of the target, the height of the detection box corresponding to the target, the focal length of the acquisition device, and the calibration parameters of the acquisition device, a transformation matrix from the image coordinate system to the vehicle coordinate system is obtained. Based on the transformation matrix and the coordinates of the midpoint of the bottom edge of the detection box corresponding to the target in the image coordinate system, the coordinate information of the target in the vehicle coordinate system is obtained. The position information and distance information of the target are calculated based on the coordinate information of the target in the vehicle coordinate system. Then, based on the matching results, the confidence of the detection box corresponding to each target in the matching results, and the position and distance information of each target in the matching results, it is determined whether the target is an edge target, whether the target is an invalid target, whether the target is an incomplete target, and whether the target is a distant target, and the edge target, the invalid target, the incomplete target, and the distant target are filtered. This disclosure sets up multiple judgment criteria. Based on the matching result, the confidence level of the detection box corresponding to each target in the matching result, and the position and distance information of each target in the matching result, it determines whether the target is an edge target, an invalid target, an incomplete target, or a distant target. It filters edge targets, invalid targets, incomplete targets, and distant targets, making the target filtering more comprehensive and flexible. It can filter out edge targets, invalid targets, incomplete targets, and distant targets, thereby reducing interfering targets, reducing the computational load on the vehicle, and increasing the running speed of the vehicle, thus reducing the difficulty of subsequent target tracking and speed and distance measurement.

[0109] Figure 4 Here is a flowchart of a pedestrian and cyclist target filtering method provided in another embodiment of this disclosure, as shown below. Figure 4 As shown, the method includes the following steps:

[0110] S401. Acquire the image acquired by the acquisition device and the calibration parameters of the acquisition device when the image was acquired.

[0111] Specifically, the implementation process and principle of S401 and S101 are the same, and will not be repeated here.

[0112] S402. Obtain the recognition result of the image, the recognition result including multiple pedestrian detection boxes and multiple cyclist detection boxes.

[0113] Specifically, the implementation process and principle of S402 and S102 are the same, and will not be repeated here.

[0114] S403. Obtain the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets.

[0115] Specifically, the implementation process and principle of S403 and S103 are the same, and will not be repeated here.

[0116] S404. Determine the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image is acquired.

[0117] Specifically, the implementation process and principle of S404 and S104 are the same, and will not be repeated here.

[0118] S405. If the detection box corresponding to the target in the matching result is located at the edge of the image, then the target is determined to be an edge target.

[0119] If the detection box corresponding to the target in the matching result is located at the edge of the image, then the target is determined to be an edge target. The edge location can be set by the user and is not limited. For example, when more than 80% of the area of ​​the detection box corresponding to the target is located at the left or right edge of the image, the target is determined to be an edge target.

[0120] S406. If the ordinate of the bottom edge of the detection box corresponding to the target in the matching result is less than the ordinate of the intersection of the left and right edges of the road in the image, then the target is determined to be an invalid target.

[0121] For a road in an image, the left and right edges of the road will intersect at a point at the far end. If the ordinate of the bottom edge of the detection box corresponding to the target in the matching result is less than the ordinate of the intersection point of the left and right edges of the road in the image, it means that the target is in the air or far away from the vehicle, and the target is determined to be an invalid target.

[0122] S407. For pedestrian targets in the matching results, if the confidence of the detection box corresponding to the pedestrian target is less than the preset pedestrian confidence threshold, then the pedestrian target is determined to be an invalid target.

[0123] For pedestrian targets in the matching results, if the confidence level of the detection box corresponding to the pedestrian target is less than the preset pedestrian confidence level threshold, the vehicle terminal determines that the pedestrian target is an invalid target.

[0124] S408. For the cyclist target in the matching result, if the distance information of the cyclist target is less than a preset distance threshold and the confidence of the detection box corresponding to the cyclist target is less than a preset cyclist confidence threshold, then the cyclist target is determined to be an incomplete target.

[0125] For the cyclist target in the matching results, if the distance information of the cyclist target is less than a preset distance threshold and the confidence of the detection box corresponding to the cyclist target is less than a preset cyclist confidence threshold, then the vehicle terminal determines that the cyclist target is an incomplete target.

[0126] It should be noted that S405, S406, S407, and S408 are parallel and have no specific order. They can be interchanged at will. The embodiments disclosed herein are for illustrative purposes only and do not limit the order of the steps.

[0127] S409. Based on the distance information of other targets in the matching result, sort the other targets, determine a preset number of the closest targets from the other targets, and determine the remaining targets as distant targets. The other targets are targets other than edge targets, invalid targets, and incomplete targets.

[0128] In this step, the vehicle-mounted terminal sorts the other targets based on their distance information in the matching results, identifies a preset number of the closest targets, and classifies the remaining targets as distant targets. These other targets are those excluding edge targets, invalid targets, and incomplete targets. Specifically, it determines whether the number of other targets is greater than a preset number. If the number is greater than the preset number, the other targets are sorted based on their distance information in the matching results, and the preset number of the closest targets are identified, while the remaining targets are classified as distant targets. When the number of other targets is less than or equal to the preset number, it is not necessary to identify distant targets.

[0129] S410. Filter the target to be filtered.

[0130] Specifically, the implementation process and principle of S410 and S312 are the same, and will not be repeated here.

[0131] This embodiment of the disclosure obtains the recognition result of the image by acquiring the image acquired by the acquisition device and the calibration parameters of the acquisition device when the image was acquired. The recognition result includes multiple pedestrian detection boxes and multiple cyclist detection boxes. The matching result of the multiple cyclist detection boxes and the multiple pedestrian detection boxes is obtained. The matching result includes pedestrian targets and cyclist targets. The position information and distance information of each target in the matching result are determined according to the calibration parameters of the acquisition device when the image was acquired. If the detection box corresponding to the target in the matching result is located at the edge of the image, the target is determined to be an edge target. If the ordinate of the bottom edge of the detection box corresponding to the target in the matching result is less than the ordinate of the intersection of the left and right edges of the road in the image, the target is determined to be an invalid target. For pedestrian targets in the matching result, if the confidence of the detection box corresponding to the pedestrian target is less than a preset pedestrian confidence threshold, the pedestrian target is determined to be an invalid target. For cycling targets in the matching result, if the distance information of the cycling target is less than a preset distance threshold and the confidence of the detection box corresponding to the cycling target is less than a preset cyclist confidence threshold, the cycling target is determined to be an incomplete target. Based on the distance information of other targets in the matching result, the other targets are sorted, and a preset number of the closest targets are determined from the other targets. The remaining targets are determined to be distant targets. The other targets are targets other than edge targets, invalid targets, and incomplete targets. Edge targets, invalid targets, incomplete targets, and distant targets are then filtered. This embodiment of the disclosure employs corresponding filtering methods for different targets in the matching results, making the target filtering more comprehensive and flexible. It can filter out edge targets, invalid targets, incomplete targets, and distant targets, thereby reducing interfering targets, further reducing the computational load on the vehicle, and increasing the running speed of the vehicle, thus reducing the difficulty of subsequent target tracking and speed and distance measurement.

[0132] Figure 5 This is a schematic diagram of the pedestrian and cyclist target filtering device provided in this embodiment. The pedestrian and cyclist target filtering device can be a vehicle-mounted terminal as described in the above embodiment, or it can be a component or assembly within the vehicle-mounted terminal. The pedestrian and cyclist target filtering device provided in this embodiment can execute the processing flow provided in the pedestrian and cyclist target filtering method embodiment, such as... Figure 5As shown, the pedestrian and cyclist target filtering device 50 includes: a first acquisition module 51, a second acquisition module 52, a third acquisition module 53, a determination module 54, and a filtering module 55; wherein, the first acquisition module 51 is used to acquire an image acquired by an acquisition device and the calibration parameters of the acquisition device when the image is acquired; the second acquisition module 52 is used to acquire the recognition result of the image, the recognition result including multiple pedestrian detection boxes and multiple cyclist detection boxes; the third acquisition module 53 is used to acquire the matching result of the multiple cyclist detection boxes and the multiple pedestrian detection boxes, the matching result including pedestrian targets and cyclist targets; the determination module 54 is used to determine the position information and distance information of each target in the matching result according to the calibration parameters of the acquisition device when the image is acquired; the filtering module 55 is used to filter the targets in the matching result according to the matching result, the confidence of the detection box corresponding to each target in the matching result, and the position information and distance information of each target in the matching result.

[0133] Optionally, when the third acquisition module 53 acquires the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, it is specifically used to: match the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes; if the cyclist detection box and the pedestrian detection box are successfully matched, the matching result is determined to be a cyclist target; if the pedestrian detection box is not matched, the matching result is determined to be a pedestrian target.

[0134] Optionally, when the determining module 54 determines the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image is acquired, it is specifically used to: transform the midpoint of the bottom edge of the detection box corresponding to the target to the vehicle coordinate system based on the calibration parameters of the acquisition device and the pixel coordinates of the midpoint of the bottom edge of the detection box corresponding to the target, to obtain the coordinate information of the target in the vehicle coordinate system; and calculate the position and distance information of the target based on the coordinate information of the target in the vehicle coordinate system.

[0135] Optionally, when the determining module 54 determines the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image is acquired, it is specifically used to: obtain a transformation matrix from the image coordinate system to the vehicle coordinate system based on the preset height of the target, the height of the detection box corresponding to the target, the focal length of the acquisition device, and the calibration parameters of the acquisition device; obtain the coordinate information of the target in the vehicle coordinate system based on the transformation matrix from the image coordinate system to the vehicle coordinate system and the coordinates of the midpoint of the bottom edge of the detection box corresponding to the target in the image coordinate system; and calculate the position and distance information of the target based on the coordinate information of the target in the vehicle coordinate system.

[0136] Optionally, when the filtering module 55 filters targets in the matching result based on the matching result, the confidence level of the detection box corresponding to each target in the matching result, and the position and distance information of each target in the matching result, it is specifically used to: determine whether the target is a target to be filtered based on the matching result, the confidence level of the detection box corresponding to each target in the matching result, and the position and distance information of each target in the matching result; if the target is a target to be filtered, then the target to be filtered is filtered.

[0137] Optionally, the targets to be filtered include edge targets, invalid targets, incomplete targets, and distant targets;

[0138] Accordingly, when the filtering module 55 determines whether the target is a target to be filtered, it is specifically used to: determine whether the target is an edge target; or determine whether the target is an invalid target; or determine whether the target is an incomplete target; or determine whether the target is a distant target.

[0139] Optionally, when the filtering module 55 determines whether the target is an edge target, it is specifically used to: if the detection box corresponding to the target in the matching result is at the edge position of the image, then the target is determined to be an edge target.

[0140] Optionally, when the filtering module 55 determines whether the target is an invalid target, it is specifically used to: if the ordinate of the bottom edge of the detection box corresponding to the target in the matching result is less than the ordinate of the intersection of the left and right edges of the road in the image, then the target is determined to be an invalid target; for pedestrian targets in the matching result, if the confidence of the detection box corresponding to the pedestrian target is less than a preset pedestrian confidence threshold, then the pedestrian target is determined to be an invalid target.

[0141] Optionally, when the filtering module 55 determines whether the target is an incomplete target, it is specifically used to: for the cyclist target in the matching result, if the distance information of the cyclist target is less than a preset distance threshold and the confidence of the detection box corresponding to the cyclist target is less than a preset cyclist confidence threshold, then the cyclist target is determined to be an incomplete target.

[0142] Optionally, when the filtering module 55 determines whether the target is a distant target, it is specifically used to: sort the other targets based on the distance information of other targets in the matching result, determine a preset number of the closest targets from the other targets, and determine the remaining targets as distant targets, wherein the other targets are targets other than edge targets, invalid targets, and incomplete targets.

[0143] Figure 5The pedestrian and cyclist target filtering device of the illustrated embodiment can be used to implement the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and will not be repeated here.

[0144] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this disclosure. See below for details. Figure 6 It shows a schematic diagram of a structure suitable for implementing the electronic device 600 in the embodiments of this disclosure. Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0145] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603 to implement the pedestrian and cyclist target filtering method as described in the embodiments of this disclosure. The RAM 603 also stores various programs and data required for the operation of the electronic device 600. The processing device 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.

[0146] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic device 600 to exchange data with other devices wirelessly or via wired communication. Although Figure 6 An electronic device 600 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0147] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts, thereby implementing the pedestrian and cyclist target filtering method as described above. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, it performs the functions defined in the methods of embodiments of this disclosure.

[0148] This disclosure also provides a vehicle, including: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the pedestrian and cyclist target filtering method as described above.

[0149] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0150] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0151] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0152] The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to:

[0153] Acquire images captured by the acquisition device and the calibration parameters of the acquisition device when the images were acquired;

[0154] Obtain the recognition results of the image, which include multiple pedestrian detection boxes and multiple cyclist detection boxes;

[0155] Obtain the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets;

[0156] The location and distance information of each target in the matching result are determined based on the calibration parameters of the acquisition device when the image is acquired.

[0157] The targets in the matching results are filtered based on the matching results, the confidence of the detection box corresponding to each target in the matching results, and the location and distance information of each target in the matching results.

[0158] Optionally, when one or more of the above-described procedures are executed by the electronic device, the electronic device may also execute other steps described in the above embodiments.

[0159] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0160] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0161] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0162] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0163] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. 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 fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0164] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0165] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0166] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for filtering pedestrian and cyclist targets, characterized in that, include: Acquire images captured by the acquisition device and the calibration parameters of the acquisition device when the images were acquired; Obtain the recognition results of the image, which include multiple pedestrian detection boxes and multiple cyclist detection boxes; Obtain the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets; The location and distance information of each target in the matching result are determined based on the calibration parameters of the acquisition device when the image is acquired. Based on the matching results, the confidence of the detection box corresponding to each target in the matching results, and the location and distance information of each target in the matching results, the targets in the matching results are filtered; The step of determining the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image was acquired includes: By transforming the target from the pixel coordinate system to the vehicle coordinate system, the coordinate information of the target in the vehicle coordinate system is obtained; The target's position and distance information are calculated based on its coordinates in the vehicle coordinate system.

2. The method according to claim 1, characterized in that, The step of obtaining the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes includes: Match the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes; If the cyclist detection box and the pedestrian detection box match successfully, then the matching result is determined to be the cyclist target; If the pedestrian detection box is not matched, the matching result is determined to be a pedestrian target.

3. The method according to claim 1, characterized in that, The step of determining the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image was acquired includes: Based on the calibration parameters of the acquisition device and the pixel coordinates of the midpoint of the bottom edge of the detection box corresponding to the target, the midpoint of the bottom edge of the detection box corresponding to the target is transformed into the vehicle coordinate system to obtain the coordinate information of the target in the vehicle coordinate system. The target's position and distance information are calculated based on its coordinates in the vehicle coordinate system.

4. The method according to claim 1, characterized in that, The step of determining the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image was acquired includes: Based on the preset height of the target, the height of the detection box corresponding to the target, the focal length of the acquisition device, and the calibration parameters of the acquisition device, a transformation matrix from the image coordinate system to the vehicle coordinate system is obtained; Based on the transformation matrix from the image coordinate system to the vehicle coordinate system and the coordinates of the midpoint of the bottom edge of the detection box corresponding to the target in the image coordinate system, the coordinate information of the target in the vehicle coordinate system is obtained. The target's position and distance information are calculated based on its coordinates in the vehicle coordinate system.

5. The method according to claim 1, characterized in that, The step of filtering targets in the matching result based on the matching result, the confidence score of the detection box corresponding to each target in the matching result, and the position and distance information of each target in the matching result includes: Based on the matching results, the confidence of the detection box corresponding to each target in the matching results, and the position and distance information of each target in the matching results, it is determined whether the target is a target to be filtered. If the target is a target to be filtered, then the target to be filtered is filtered.

6. The method according to claim 5, characterized in that, The targets to be filtered include edge targets, invalid targets, incomplete targets, and distant targets; Accordingly, determining whether the target is a target to be filtered includes: Determine whether the target is an edge target; or Determine whether the target is an invalid target; or Determine whether the target is an incomplete target; or Determine whether the target is a distant target.

7. The method according to claim 6, characterized in that, The determination of whether the target is an edge target includes: If the detection box corresponding to the target in the matching result is located at the edge of the image, then the target is determined to be an edge target.

8. The method according to claim 6, characterized in that, The determination of whether the target is an invalid target includes: If the ordinate of the bottom edge of the detection box corresponding to the target in the matching result is less than the ordinate of the intersection of the left and right edges of the road in the image, then the target is determined to be an invalid target. For pedestrian targets in the matching results, if the confidence level of the detection box corresponding to the pedestrian target is less than the preset pedestrian confidence level threshold, then the pedestrian target is determined to be an invalid target.

9. The method according to claim 6, characterized in that, The determination of whether the target is an incomplete target includes: For the cyclist target in the matching results, if the distance information of the cyclist target is less than a preset distance threshold and the confidence of the detection box corresponding to the cyclist target is less than a preset cyclist confidence threshold, then the cyclist target is determined to be an incomplete target.

10. The method according to claim 6, characterized in that, The determination of whether the target is a distant target includes: Based on the distance information of other targets in the matching results, the other targets are sorted, and a preset number of the closest targets are determined from the other targets. The remaining targets are determined as distant targets. The other targets are targets other than edge targets, invalid targets, and incomplete targets.

11. A pedestrian / cyclist target filtering device, characterized in that, include: The first acquisition module is used to acquire the image acquired by the acquisition device and the calibration parameters of the acquisition device when the image was acquired. The second acquisition module is used to acquire the recognition results of the image, the recognition results including multiple pedestrian detection boxes and multiple cyclist detection boxes; The third acquisition module is used to acquire the matching results of the plurality of cyclist detection boxes and the plurality of pedestrian detection boxes, wherein the matching results include pedestrian targets and cyclist targets; The determination module is used to determine the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image is acquired; The filtering module is used to filter the targets in the matching result based on the matching result, the confidence of the detection box corresponding to each target in the matching result, and the position information and distance information of each target in the matching result. When the determining module determines the position and distance information of each target in the matching result based on the calibration parameters of the acquisition device when the image is acquired, it is specifically used for: By transforming the target from the pixel coordinate system to the vehicle coordinate system, the coordinate information of the target in the vehicle coordinate system is obtained; The target's position and distance information are calculated based on its coordinates in the vehicle coordinate system.

12. An electronic device, characterized in that, include: Memory; processor; as well as Computer programs; The computer program is stored in the memory and configured to be executed by the processor to implement the method as described in any one of claims 1-10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-10.

14. A vehicle, characterized in that, Includes the pedestrian and cyclist target filtering device as described in claim 11.

Images