Target detection method and device, electronic equipment and storage medium
By calculating the obstacle envelope and occlusion ratio, effective obstacles are screened out, solving the problem of low detection accuracy of occluded obstacles and improving detection accuracy and safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING CO WHEELS TECH CO LTD
- Filing Date
- 2023-03-16
- Publication Date
- 2026-08-04
AI Technical Summary
In existing technologies, target detection methods based on bird's-eye view perception have low accuracy in detecting occluded obstacles, leading to incorrect obstacle attribute and state detection, which may cause incorrect vehicle decisions.
By calculating the envelope formed by obstacles and the occlusion ratio, effective obstacles are screened out, improving detection accuracy and reducing the probability of incorrect decisions.
It improves the accuracy of obstacle detection, reduces incorrect vehicle decisions caused by erroneous detection results, and enhances the user's driving experience and safety.
Smart Images
Figure CN118671730B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of computer technology, and in particular to a target detection method, apparatus, electronic device, and storage medium. Background Technology
[0002] Currently, obstacle detection methods mostly rely on sensors to collect sensor data, and then use this data to detect and identify obstacles. However, as the number of obstacles around the sensor increases, the sensor cannot directly collect sensor data for obstacles that are obscured by other obstacles, resulting in a relatively low detection accuracy.
[0003] For example, target detection methods based on Bird's Eye View (BEV) have relatively low accuracy when dealing with occluded obstacles. Because the vehicle is positioned behind other obstacles from the BEV perspective, occlusion is confirmed, preventing the accurate detection of a significant amount of obstacle information. This leads to incorrect obstacle attribute and state detection results, which can cause problems in subsequent applications, such as incorrect braking decisions. Therefore, accurate obstacle detection is a pressing issue that needs to be addressed.
[0004] In addition, polar coordinates are commonly used in existing identification processes, and will be briefly introduced here. Polar coordinates: In a plane, a fixed point O is chosen, called the pole. A ray Ox is drawn, called the polar axis. A unit of length and a positive direction of angle are selected (usually counterclockwise). For any point M in the plane, ρ represents the length of line segment OM (sometimes also r), and θ represents the angle from Ox to OM. ρ is called the polar radius of point M, and θ is called the polar angle of point M. The ordered pair (ρ, θ) is called the polar coordinates of point M. The coordinate system established in this way is called the polar coordinate system. Summary of the Invention
[0005] To address the aforementioned technical issues, this disclosure provides a target detection method, apparatus, electronic device, and storage medium, which can improve obstacle detection accuracy and reduce the probability of problems occurring in subsequent application scenarios to a certain extent.
[0006] In a first aspect, embodiments of this disclosure provide a target detection method, including:
[0007] The sensor data collected by the sensor device is acquired, and the sensor data is processed for obstacle identification to obtain the target information of each obstacle.
[0008] The first intersection point between each obstacle and each ray in the target area is calculated based on the target information, wherein the target area is a region constructed by emitting multiple rays from the sensing device as the central starting point;
[0009] An obstacle envelope centered on the sensing device is formed based on the first intersection point of each obstacle;
[0010] The occlusion ratio of the obstacle is calculated based on the obstacle envelope and the first intersection point. The effectiveness of the obstacle is determined based on the relationship between the occlusion ratio and the first preset ratio, and an obstacle detection result including effective obstacles is obtained.
[0011] Optionally, the target information includes edge information.
[0012] Optionally, calculating the first intersection point between the obstacle and each ray in the target region based on the target information includes:
[0013] In the polar coordinate system, rays are emitted from the sensing device as the central starting point and distributed at preset angular intervals to construct the target area.
[0014] Based on the edge information, calculate the first intersection point between the edge of each obstacle and each ray in the target area.
[0015] Optionally, forming an obstacle envelope centered on the sensing device based on the first intersection point of each obstacle includes:
[0016] The first intersection points on each ray in the target area are sorted, and the second intersection point on each ray that is closest to the sensing device is obtained;
[0017] The second intersection points on each ray are connected sequentially to form an obstacle envelope centered on the sensing device.
[0018] Optionally, the target information includes size information.
[0019] Optionally, before sequentially connecting the second intersection points on each ray to form an obstacle envelope centered on the sensing device, the following steps are included:
[0020] If the obstacle corresponding to the second intersection point contains a target obstacle whose size information is smaller than the preset size, then
[0021] Determine the target ray containing the second intersection point corresponding to the target obstacle;
[0022] Determine the first intersection point on the target ray, excluding all intersection points between the target obstacle and the target ray;
[0023] The second intersection point on each of the other rays besides the target ray and the first intersection point are connected in sequence to form an obstacle envelope centered on the sensing device.
[0024] Optionally, after identifying a target obstacle with a size smaller than a preset size among the obstacles corresponding to the second intersection point, the method further includes:
[0025] The target obstacle is set to a penetrable attribute, wherein the penetrable attribute means that the target obstacle does not obstruct other obstacles;
[0026] The target obstacle is determined to be a valid obstacle based on the penetrability attribute.
[0027] Optionally, the step of calculating the occlusion ratio of the obstacle based on the obstacle envelope and the first intersection point, determining the effectiveness of the obstacle based on the relationship between the occlusion ratio and a first preset ratio, and obtaining an obstacle detection result including effective obstacles includes:
[0028] The occlusion ratio of the obstacle is calculated based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle;
[0029] If the obtained attribute of the obstacle is not a penetrable attribute, and the occlusion ratio is less than the first preset ratio, then the obstacle is determined to be a valid obstacle.
[0030] Obstacle detection information is generated based on the target information of all valid obstacles.
[0031] Optionally, calculating the occlusion ratio of the obstacle based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle includes:
[0032] The visibility ratio of the obstacle is calculated based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle;
[0033] The occlusion ratio of the obstacle is obtained based on the difference between the second preset ratio and the visible ratio.
[0034] Secondly, embodiments of this disclosure provide a target detection device, comprising:
[0035] The acquisition module is used to acquire the sensing data collected by the sensing device, and to perform obstacle identification processing on the sensing data to obtain the target information of each obstacle;
[0036] The calculation module is used to calculate the first intersection point between each obstacle and each ray in the target area based on the target information, wherein the target area is a region constructed by emitting multiple rays from the sensing device as the center starting point;
[0037] A forming module is used to form an obstacle envelope centered on the sensing device based on the first intersection point of each obstacle;
[0038] The generation module is used to calculate the occlusion ratio of the obstacle based on the obstacle envelope and the first intersection point, determine the validity of the obstacle based on the relationship between the occlusion ratio and the first preset ratio, and obtain the obstacle detection result including the valid obstacle.
[0039] Thirdly, embodiments of this disclosure provide an electronic device, including:
[0040] Memory;
[0041] Processor; and
[0042] Computer programs;
[0043] The computer program is stored in memory and configured to be executed by the processor to implement the target detection method described above.
[0044] Fourthly, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the target detection method described above.
[0045] This disclosure provides a target detection method, comprising: acquiring sensing data collected by a sensing device, and performing obstacle identification processing on the sensing data to obtain target information of each obstacle; calculating the first intersection point between each obstacle and each ray in the target area based on the target information, wherein the target area is a region constructed by emitting multiple rays centered on the sensing device; forming an obstacle envelope centered on the sensing device based on the first intersection points of each obstacle; calculating the occlusion ratio of the obstacle based on the obstacle envelope and the first intersection points; determining the validity of the obstacle based on the relationship between the occlusion ratio and a first preset ratio; and obtaining an obstacle detection result including valid obstacles. The method provided by this disclosure can filter out invalid obstacles whose detection results are prone to errors due to severe occlusion, outputting detection results including valid obstacles, improving detection accuracy, and to a certain extent reducing problems in subsequent application scenarios. Attached Figure Description
[0046] 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.
[0047] 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, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0048] Figure 1 A schematic diagram illustrating an application scenario provided by an embodiment of this disclosure;
[0049] Figure 2 A schematic flowchart of a target detection method provided in an embodiment of this disclosure;
[0050] Figure 3 A schematic diagram of a target area provided in an embodiment of this disclosure;
[0051] Figure 4 A schematic flowchart of a target detection method provided in an embodiment of this disclosure;
[0052] Figure 5 A schematic diagram of an obstacle envelope provided in an embodiment of this disclosure;
[0053] Figure 6 A schematic flowchart of a target detection method provided in an embodiment of this disclosure;
[0054] Figure 7 This is a schematic diagram of the structure of a target detection device provided in an embodiment of the present disclosure;
[0055] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation
[0056] 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.
[0057] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.
[0058] To address the aforementioned technical problems, this disclosure provides a target detection method. From a BEV perspective, by calculating the envelope formed by obstacles and the occlusion ratio of each obstacle, invalid obstacles detected by the perception algorithm that could cause false braking are filtered out, and the detection result containing valid obstacles is output. This improves detection accuracy and, to a certain extent, reduces erroneous decisions made by the vehicle due to detection results containing invalid obstacles, thereby reducing the probability of safety issues in subsequent application scenarios. Simultaneously, while ensuring safety, it enhances the user's driving experience. Detailed descriptions are provided through one or more of the following embodiments.
[0059] Specifically, target detection methods can be performed by vehicles or servers. For example, in one application scenario, such as... Figure 1 As shown, vehicle 11 transmits the sensor data collected by the sensing device to server 12. Server 12 determines the validity of obstacles based on the acquired sensor data, obtains detection results including valid obstacle target information, and returns the detection results to vehicle 11. Vehicle 11 makes control decisions based on the detection results.
[0060] In another application scenario, vehicle 11 determines the effectiveness of obstacles based on the sensing data collected by the sensing device, obtains detection results including information on effective obstacle targets, and makes control decisions based on the detection results.
[0061] It is understood that the target detection method provided in this disclosure is not limited to the possible scenarios described above. The following detailed description uses vehicle 11 performing the target detection method as an example.
[0062] Figure 2 This is a flowchart illustrating a target detection method provided in an embodiment of the present disclosure, applied to the aforementioned vehicle 11, specifically including the following: Figure 2 The following steps S210 to S240 are shown:
[0063] S210. Acquire the sensing data collected by the sensing device, and perform obstacle identification processing on the sensing data to obtain the target information of each obstacle.
[0064] The target information includes edge information and size information.
[0065] Understandably, the sensing device can be a sensor, such as a lidar unit or a camera. The deployment method can involve installing at least one sensing device on a vehicle or other equipment. This device can be used to collect environmental information around the vehicle or other equipment to generate sensing data, which can be the aforementioned bird's-eye view. Subsequently, existing perception algorithms can be used to process the sensing data for obstacle identification, obtaining target information for each obstacle. The specific perception algorithm is not limited. The target information includes edge information and size information. Edge information includes the obstacle's form, such as linear, rectangular, or polygonal shapes.
[0066] S220. Calculate the first intersection point between each obstacle and each ray in the target area based on the target information.
[0067] The target area is a region constructed by emitting multiple rays from the sensing device as the central starting point.
[0068] Optionally, the calculation of the first intersection point in S220 above can be achieved through the following steps:
[0069] In the polar coordinate system, with the sensing device as the central starting point, rays are emitted at preset angular intervals to construct the target area; based on the edge information, the first intersection point between the edge of each obstacle and each ray in the target area is calculated.
[0070] Understandably, in a polar coordinate system, rays are emitted from the sensing device as the central starting point and distributed at preset angular intervals to form a target area. The target area can also be understood as a polar coordinate canvas, which includes a center and multiple rays emanating from the center. The preset angles can be determined by the user according to their needs.
[0071] Understandably, if a vehicle is equipped with multiple sensors, a target area needs to be constructed for each sensor. Different sensors collect different data, leading to different target information for obstacles. Therefore, for each sensor, the effectiveness of obstacles must be determined, invalid obstacles must be filtered out, and detection results including valid obstacles must be obtained. Finally, the detection results from different sensors are comprehensively processed to facilitate correct control decisions by the vehicle. The following embodiment uses a single sensor as an example.
[0072] Understandably, after the target area is constructed, the intersection points of the edges of each obstacle with each ray in the target area are calculated based on the edge information of each obstacle. These intersection points are recorded as the first intersection points. An obstacle may have a first intersection point with different rays, or it may have multiple first intersection points with a single ray. It is also possible that due to a large preset angle, the obstacle may not have a first intersection point with any ray. Therefore, the preset angle can be set according to different application requirements to ensure that each obstacle has at least one first intersection point with at least one ray.
[0073] For example, see Figure 3 , Figure 3 This is a schematic diagram of a target area provided in an embodiment of the present disclosure. Figure 3 The target area includes multiple rays spaced apart around the sensing device and a rectangular obstacle 1. The edge of the rectangular obstacle 1 and three rays all have a first intersection point. These three rays are denoted as rays 1 to 3. Rays 1 and 3 each have one first intersection point, and ray 2 has two first intersection points. The rectangular obstacle 1 and each ray in the target area have four first intersection points. Figure 3 The four black dots shown represent different obstacles with varying numbers of first intersection points.
[0074] S230. An obstacle envelope centered on the sensing device is formed based on the first intersection point of each obstacle.
[0075] Understandably, based on the above S220, an obstacle envelope centered on the sensing device is formed according to the first intersection of all obstacles. The obstacle envelope is formed on the basis of the target area and is a line formed by connecting some of the first intersections in sequence.
[0076] S240. Calculate the occlusion ratio of the obstacle based on the obstacle envelope and the first intersection point, determine the effectiveness of the obstacle based on the relationship between the occlusion ratio and the first preset ratio, and obtain the obstacle detection result including the effective obstacle.
[0077] Understandably, based on the above and S220 and S230, for each obstacle, the occlusion ratio of the obstacle is calculated based on the first intersection point of the obstacle falling on the obstacle envelope and all first intersection points. The occlusion ratio refers to the proportion of the part of the obstacle that cannot be directly collected by the sensing device (the occluded part) to the entire obstacle. The validity of the obstacle is determined based on the relationship between the occlusion ratio and the first preset ratio. The validity of the obstacle refers to the occlusion status of the obstacle. If the entire obstacle or most of the obstacles are occluded, the obstacle is very likely to be an invalid obstacle. The target information of invalid obstacles identified by the perception algorithm has relatively low accuracy and is prone to erroneous detection results. Therefore, the target information of invalid obstacles can be filtered out. If the obstacle is not occluded or only a small part is occluded, the obstacle is very likely to be a valid obstacle. The target information of valid obstacles identified by the perception algorithm has relatively high accuracy. Therefore, the target information of valid obstacles is output.
[0078] This disclosure provides a target detection method. It acquires sensing data collected by a sensing device and performs obstacle recognition processing on the sensing data to obtain target information for each obstacle. This target information is the recognition result output from a normal obstacle recognition process, which may contain errors due to severe occlusion. In this case, the method calculates the first intersection point between each obstacle and each ray in the target area based on the target information of all obstacles. Then, based on some of the first intersection points, an obstacle envelope centered on the sensing device is formed on the target area. The occlusion ratio of the obstacle is calculated based on the first intersection points falling on the obstacle envelope and all the first intersection points. The validity of the obstacle is determined based on the relationship between the occlusion ratio and a first preset ratio. Target information of invalid obstacles that are severely occluded and prone to errors is filtered out, resulting in a final obstacle detection result including valid obstacle target information. By filtering out invalid obstacle target information and retaining valid obstacle target information, the detection accuracy can be effectively improved. This can reduce problems in subsequent application scenarios to a certain extent, such as reducing erroneous decisions made by vehicles due to inaccurate detection, thus improving the user's driving experience while ensuring safety.
[0079] Based on the above embodiments, see Figure 4 , Figure 4 This is a flowchart illustrating a target detection method provided in an embodiment of the present disclosure. Optionally, the formation of the obstacle envelope in step S230 specifically includes, as follows: Figure 4 The following steps S410 to S420 are shown:
[0080] S410. Sort the first intersection points on each ray in the target area and obtain the second intersection point on each ray that is closest to the sensing device.
[0081] Understandably, in each ray in the target area, a ray to be processed is determined to have at least one first intersection point. The at least one first intersection point on each ray to be processed is sorted according to its distance from the sensing device (center). The first intersection point on each ray to be processed that is closest to the sensing device is obtained, and the closest first intersection point is taken as the second intersection point on the corresponding ray.
[0082] S420. The second intersection points on each ray are connected sequentially to form an obstacle envelope centered on the sensing device.
[0083] Understandably, based on the above S410, in the target area, the second intersection points on each ray are connected in sequence to form an obstacle envelope centered on the sensing device. That is, the line formed by connecting all the second intersection points in a preset order is used as the envelope. For example, first select the second intersection point on any ray, and then select the second intersection points on adjacent rays in a clockwise order and connect them. The line obtained after traversing all rays in a preset order can be used as the envelope. Optionally, if there is no first intersection point on the ray, it is impossible to obtain a second intersection point on the ray. In this case, the sensing device (center) can be used as the second intersection point on the ray for connection.
[0084] Optionally, the formation of the obstacle envelope in S420 is achieved through the following steps:
[0085] If there is a target obstacle with a size smaller than a preset size among the obstacles corresponding to the second intersection point, then the target ray corresponding to the second intersection point of the target obstacle is determined; the first first intersection point is determined on the target ray, excluding the intersection point with the edge of the target obstacle; the first first intersection point is taken as the second intersection point of the target ray; wherein the first first intersection point is the intersection point on the target ray that is closest to the sensing device, excluding the intersection point with the edge of the target obstacle.
[0086] Understandably, before obtaining the final obstacle envelope, an initial envelope is formed based on the second intersection point determined in S410. Subsequently, based on the size information, it is determined whether any target obstacle smaller than a preset size exists among all obstacles corresponding to the second intersection points. The preset size can be the size of a preset rectangle with both length and width of 1 meter. If no target obstacle smaller than the preset size (length and width both less than 1 meter) exists, the initial envelope is directly used as the final obstacle envelope. If a target obstacle smaller than the preset size exists, meaning the preset rectangle can completely cover the target obstacle, in this case, the target obstacle will not obstruct the sensor's data acquisition of other obstacles behind it, or the target obstacle's obstruction of other obstacles is limited and will not significantly affect the detection results of other obstacles. In this case, the second intersection point of the target obstacle on the initial envelope is adjusted. Specifically, the target ray corresponding to the second intersection point of the target obstacle is determined. On the sorted target ray, the first intersection point (excluding intersections with the edge of the target obstacle) is identified. This first intersection point is the intersection of the edge of other obstacles with the target ray and is closest to the sensing device. This first intersection point is then adjusted to become the second intersection point on the target ray. In other words, after obtaining the initial envelope, the second intersection point on the envelope is adjusted for the presence of small obstacles (target obstacles). Moving backward along the target ray, the first intersection point on the sorted target ray, excluding all intersections of small obstacles on the target ray, is taken as the new second intersection point on the target ray. Subsequently, the new second intersection point on the target ray is connected sequentially with the second intersection points on the other rays (excluding the target ray) to form a new envelope. The presence of small obstacles on the envelope is then checked again. If no small obstacles are found, the new envelope is taken as the final obstacle envelope.
[0087] For example, see Figure 5 , Figure 5 This is a schematic diagram of an obstacle envelope provided in an embodiment of the present disclosure. Figure 5This includes an initial envelope 510 and a final obstacle envelope 520. The initial envelope 510, after adjusting the second intersection point, yields the final obstacle envelope 520. The initial envelope 510 only shows the sensor 516, part of the obstacle content, and some intersection points between the edges of the obstacles and the target ray 512. The initial envelope 510 contains a target obstacle 511 smaller than a preset size. The target obstacle 511 and the target ray 512 have a second intersection point 513 and a first intersection point 514. Excluding the second intersection point 513 and the first intersection point 514 corresponding to the target obstacle 511 on the target ray 512, the first intersection point on the target ray 512 is the first intersection point 515. The first intersection point 515 is taken as the new second intersection point on the target ray 512. A new envelope is formed based on the first intersection point 515 and the second intersection points on the remaining rays. If the new envelope does not contain any small obstacle intersection points, it is taken as the final obstacle envelope. Figure 5 As shown in the final obstacle envelope 520, at this point, the target obstacle 511 is ahead of the final obstacle envelope 520.
[0088] Optionally, after identifying the target obstacle, the following steps may also be included:
[0089] The target obstacle is set to a penetrable attribute, wherein the penetrable attribute means that the target obstacle does not obstruct other obstacles; the target obstacle is determined to be a valid obstacle based on the penetrable attribute.
[0090] Understandably, based on the above S410, if there is a target obstacle smaller than a preset size among the obstacles corresponding to the second intersection point, then the attribute of all intersection points of the target obstacle is set to the penetrable attribute, and the attribute of the target obstacle is also set to the penetrable attribute. The penetrable attribute means that the target obstacle does not obstruct other obstacles, or the target obstacle only slightly obstructs other obstacles. Subsequently, when determining the validity of obstacles, target obstacles with the penetrable attribute can be directly determined as valid obstacles, meaning that the detection results for the target obstacles are relatively accurate and can be output to the vehicle for reference.
[0091] This disclosure provides a target detection method. First intersection points on each ray are sorted, and second intersection points on each ray closest to the sensing device are obtained. It is determined whether there is a target obstacle smaller than a preset size among the obstacles corresponding to the second intersection point. If so, the target ray containing the second intersection point corresponding to the target obstacle is determined. This adjusts the intersection point of the target obstacle on the envelope, meaning the target obstacle obstructs other obstacles less, thus having a smaller impact on the accuracy of other obstacle identification results. Therefore, a new second intersection point is determined by moving backward along the target ray, thereby updating the envelope and facilitating accurate subsequent determination of the effectiveness of each obstacle.
[0092] Based on the above embodiments, Figure 6 This is a flowchart illustrating a target detection method provided in an embodiment of the present disclosure. The above-described S420 step, determining the effectiveness of an obstacle, specifically includes, as follows: Figure 6 The following steps S610 to S630 are shown:
[0093] S610. Calculate the occlusion ratio of the obstacle based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection point of the obstacle.
[0094] Optionally, the calculation of the occlusion ratio in S610 above is achieved through the following steps:
[0095] The visibility ratio of the obstacle is calculated based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection point of the obstacle; the occlusion ratio of the obstacle is obtained based on the difference between the second preset ratio and the visibility ratio.
[0096] For each obstacle, determine the number of second intersection points of the obstacle falling on the envelope and the number of rays that intersect with each ray in the target area (i.e., the number of rays containing the first intersection points of the obstacle). Use the ratio of the number of second intersection points of the obstacle falling on the envelope to the number of rays that intersect with each ray in the target area as the visibility ratio, thus obtaining the visibility ratio of the obstacle; for example, ... Figure 3As shown, obstacle 1 has four first intersection points with rays 1, 2, and 3, respectively, denoted as 210 to 240. Therefore, the number of rays with intersection points between the obstacle and each ray in the target area is 3. Among them, obstacle 1 has a second intersection point with ray 1 that falls on the envelope (i.e., first intersection point 230), obstacle 1 has a second intersection point with ray 2 that falls on the envelope (i.e., first intersection point 220), and obstacle 1 has a second intersection point with ray 3 that falls on the envelope (i.e., first intersection point 210). Therefore, the number of second intersection points of the obstacle that fall on the envelope is 3. Thus, the ratio of the number of second intersection points of obstacle 1 that fall on the envelope to the number of rays with intersection points between obstacle 1 and each ray in the target area is 1, that is, the visibility ratio of obstacle 1 is 1.
[0097] Understandably, the occlusion ratio of an obstacle is obtained by calculating the difference between the second preset ratio and the visible ratio. Based on the above example, preferably, the second preset ratio is 1, and the difference between the second preset ratio and the visible ratio is 1-1=0, that is, the occlusion ratio is 0, which means that obstacle 1 is not occluded and is completely visible to the sensing device. An occlusion ratio of 1 means that the obstacle is completely occluded and is not visible to the sensing device. The larger the occlusion ratio, the lower the visibility to the sensing device. An occlusion ratio between 0 and 1 means that it is partially occluded, that is, partially visible. The specific division of the occlusion ratio can be determined by the user according to their needs.
[0098] S620. If the obtained attribute of the obstacle is not a penetrable attribute, and the occlusion ratio is less than a first preset ratio, then the obstacle is determined to be a valid obstacle.
[0099] Understandably, based on the above S610, the attributes of each obstacle are obtained. If the attribute of the obstacle is not a penetrable attribute and the occlusion ratio of the obstacle is less than the first preset ratio, then the obstacle is determined as a valid obstacle. If the attribute of the obstacle is a penetrable attribute, the obstacle is directly determined as a valid obstacle without considering the occlusion ratio. If the attribute of the obstacle is not a penetrable attribute and the occlusion ratio of the obstacle is greater than or equal to the first preset ratio, then the obstacle is determined as an invalid obstacle.
[0100] S630. Generate obstacle detection information based on the target information of all valid obstacles.
[0101] Understandably, obstacle detection information is generated based on the target information of all valid obstacles. This obstacle detection information is the detection result of the sensing data collected by a sensing device.
[0102] This disclosure provides a target detection method. For each obstacle, the visible ratio of the obstacle is calculated by using the second intersection point falling on the envelope line and removing all first intersection points on the same ray except for the first first intersection point. The occlusion ratio of the obstacle is calculated based on the difference between the second preset ratio and the visible ratio. Subsequently, based on the occlusion ratio and attributes of the obstacle, valid obstacles are determined. Valid obstacles have less occlusion, allowing the sensing device to obtain a large amount of information, thus resulting in higher detection accuracy. All obstacles behind the envelope line are filtered out as invalid obstacles. Invalid obstacles have an occlusion ratio close to 1 and their attributes are not penetrable. Invalid obstacles have more occlusion, preventing the sensing device from obtaining a large amount of information, resulting in lower detection accuracy and a high likelihood of erroneous detection results. Therefore, outputting detection results that include valid obstacle target information can improve detection accuracy, reduce the possibility of incorrect decisions made by the vehicle due to inaccurate detection, and improve the user's driving experience while ensuring safety.
[0103] Figure 7 This is a schematic diagram of the target detection device provided in an embodiment of the present disclosure. The target detection device provided in this embodiment of the present disclosure can execute the processing flow provided in the above-described target detection method embodiment, and is applied to the above-described vehicle 11, such as... Figure 7 As shown, the target detection device 700 includes:
[0104] The acquisition module 710 is used to acquire the sensing data collected by the sensing device, and to perform obstacle recognition processing on the sensing data to obtain the target information of each obstacle.
[0105] The calculation module 720 is used to calculate the first intersection point between each obstacle and each ray in the target area based on the target information, wherein the target area is a region constructed by emitting multiple rays from the sensing device as the center starting point;
[0106] Forming module 730 is used to form an obstacle envelope centered on the sensing device based on the first intersection point of each obstacle;
[0107] The generation module 740 is used to calculate the occlusion ratio of the obstacle based on the obstacle envelope and the first intersection point, determine the validity of the obstacle based on the relationship between the occlusion ratio and the first preset ratio, and obtain an obstacle detection result including valid obstacles.
[0108] The target information includes edge information.
[0109] Optionally, the computing module 720 is used for:
[0110] In the polar coordinate system, rays are emitted from the sensing device as the central starting point and distributed at preset angular intervals to construct the target area.
[0111] Based on the edge information, calculate the first intersection point between the edge of each obstacle and each ray in the target area.
[0112] Optionally, forming module 730 is used for:
[0113] The first intersection points on each ray in the target area are sorted, and the second intersection point on each ray that is closest to the sensing device is obtained;
[0114] The second intersection points on each ray are connected sequentially to form an obstacle envelope centered on the sensing device.
[0115] The target information includes size information.
[0116] Optionally, device 700 is also used for:
[0117] If there is a target obstacle with a size smaller than a preset size among the obstacles corresponding to the second intersection point, then the target ray corresponding to the second intersection point of the target obstacle is determined;
[0118] On the target ray, a first intersection point is determined, excluding the intersection point with the edge of the target obstacle, wherein the first intersection point is the intersection point on the target ray that is closest to the sensing device, excluding the intersection point with the edge of the target obstacle;
[0119] The first intersection point is determined as the second intersection point on the target ray.
[0120] Optionally, device 700 is also used for:
[0121] The target obstacle is set to a penetrable attribute, wherein the penetrable attribute means that the target obstacle does not obstruct other obstacles;
[0122] The target obstacle is determined to be a valid obstacle based on the penetrability attribute.
[0123] Optionally, the generation module 740 is used for:
[0124] The occlusion ratio of the obstacle is calculated based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle;
[0125] If the obtained attribute of the obstacle is not a penetrable attribute, and the occlusion ratio is less than the first preset ratio, then the obstacle is determined to be a valid obstacle.
[0126] Obstacle detection information is generated based on the target information of all valid obstacles.
[0127] Optionally, the generation module 740 is used for:
[0128] The visibility ratio of the obstacle is calculated based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle;
[0129] The occlusion ratio of the obstacle is obtained based on the difference between the second preset ratio and the visible ratio.
[0130] Figure 7 The target detection device shown in the embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and will not be repeated here.
[0131] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. See below for details. Figure 8 The diagram illustrates a structural schematic suitable for implementing the electronic device 800 in the embodiments of this disclosure. The electronic device 800 in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), wearable electronic devices, etc., as well as fixed terminals such as digital TVs, desktop computers, smart home devices, etc. Figure 8 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.
[0132] like Figure 8 As shown, the electronic device 800 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 808 into a random access memory (RAM) 803 to implement the target detection method as described in the embodiments of this disclosure. The RAM 803 also stores various programs and data required for the operation of the electronic device 800. The processing device 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.
[0133] Typically, the following devices can be connected to I / O interface 805: input devices 806 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 807 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 808 including, for example, magnetic tapes, hard disks, etc.; and communication devices 809. Communication device 809 allows electronic device 800 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 7 An electronic device 800 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.
[0134] 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 target detection method as described above. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 809, or installed from a storage device 808, or installed from a ROM 802. When the computer program is executed by the processing device 801, it performs the functions defined in the methods of embodiments of this disclosure.
[0135] 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.
[0136] 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.
[0137] 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.
[0138] Optionally, when one or more of the above-described procedures are executed by the electronic device, the electronic device may also perform other steps described in the above embodiments.
[0139] 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).
[0140] 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.
[0141] 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.
[0142] 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.
[0143] 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.
[0144] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or gateway that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or gateway. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or gateway that includes said element.
[0145] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A target detection method characterized by, include: The sensor data collected by the sensor device is acquired, and the sensor data is processed for obstacle identification to obtain the target information of each obstacle. The first intersection point between each obstacle and each ray in the target area is calculated based on the target information, wherein the target area is a region constructed by emitting multiple rays from the sensing device as the central starting point; Forming an obstacle envelope centered on the sensing device based on the first intersection points of each obstacle includes: sorting the first intersection points on each ray in the target area, obtaining the second intersection point on each ray that is closest to the sensing device; and sequentially connecting the second intersection points on each ray to form an obstacle envelope centered on the sensing device. The occlusion ratio of the obstacle is calculated based on the obstacle envelope and the first intersection point. The effectiveness of the obstacle is determined based on the relationship between the occlusion ratio and the first preset ratio, and an obstacle detection result including effective obstacles is obtained.
2. The method according to claim 1, characterized in that, The target information includes edge information, and the step of calculating the first intersection point between the obstacle and each ray in the target region based on the target information includes: In the polar coordinate system, rays are emitted from the sensing device as the central starting point and distributed at preset angular intervals to construct the target area. Based on the edge information, calculate the first intersection point between the edge of each obstacle and each ray in the target area.
3. The method according to claim 1, characterized in that, The target information includes size information. Before sequentially connecting the second intersection points on each ray to form an obstacle envelope centered on the sensing device, the process includes: If there is a target obstacle with a size smaller than a preset size among the obstacles corresponding to the second intersection point, then the target ray corresponding to the second intersection point of the target obstacle is determined; On the target ray, a first intersection point is determined, excluding the intersection point with the edge of the target obstacle, wherein the first intersection point is the intersection point on the target ray that is closest to the sensing device, excluding the intersection point with the edge of the target obstacle; The first intersection point is determined as the second intersection point on the target ray.
4. The method according to claim 3, characterized in that, After identifying a target obstacle with a size smaller than a preset size among the obstacles corresponding to the second intersection point, the method further includes: The target obstacle is set to a penetrable attribute, wherein the penetrable attribute means that the target obstacle does not obstruct other obstacles; The target obstacle is determined to be a valid obstacle based on the penetrability attribute.
5. The method according to claim 1, characterized in that, The step of calculating the occlusion ratio of the obstacle based on the obstacle envelope and the first intersection point, determining the effectiveness of the obstacle based on the relationship between the occlusion ratio and a first preset ratio, and obtaining an obstacle detection result including effective obstacles includes: The occlusion ratio of the obstacle is calculated based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle; If the obtained attribute of the obstacle is not a penetrable attribute, and the occlusion ratio is less than the first preset ratio, then the obstacle is determined to be a valid obstacle. Obstacle detection information is generated based on the target information of all valid obstacles.
6. The method according to claim 5, characterized in that, The calculation of the occlusion ratio of the obstacle based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle includes: The visibility ratio of the obstacle is calculated based on the number of second intersection points corresponding to the obstacle and the number of rays corresponding to the first intersection points of the obstacle; The occlusion ratio of the obstacle is obtained based on the difference between the second preset ratio and the visible ratio.
7. A target detection device, characterized in that, include: The acquisition module is used to acquire the sensing data collected by the sensing device, and to perform obstacle identification processing on the sensing data to obtain the target information of each obstacle; The calculation module is used to calculate the first intersection point between each obstacle and each ray in the target area based on the target information, wherein the target area is a region constructed by emitting multiple rays from the sensing device as the center starting point; A forming module is used to form an obstacle envelope centered on the sensing device based on the first intersection point of each obstacle; The generation module is used to calculate the occlusion ratio of the obstacle based on the obstacle envelope and the first intersection point, determine the validity of the obstacle based on the relationship between the occlusion ratio and a first preset ratio, and obtain an obstacle detection result including valid obstacles. The forming module is used for: The first intersection points on each ray in the target area are sorted to obtain the second intersection point on each ray that is closest to the sensing device; the second intersection points on each ray are connected in sequence to form an obstacle envelope centered on the sensing device.
8. 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 target detection method as described in any one of claims 1 to 6.
9. 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 steps of the target detection method as described in any one of claims 1 to 6.