Target area positioning method and target area positioning device

Abstract

This disclosure provides a method and apparatus for locating a target region, relating to the field of region positioning technology. The method includes: when the probability of a target object appearing in a target environment region is determined to be greater than a third preset threshold, acquiring a first point cloud corresponding to the target object; determining the pose information of the target object based on the first point cloud; determining an initial search position for a target region within the target object based on the pose information; projecting the first point cloud onto a first plane, and determining multiple candidate regions based on the initial search position and the reflection intensity of the point cloud projected onto the first plane; and locating the target region based on the multiple candidate regions and image data. This technical solution can effectively improve positioning efficiency while ensuring the accuracy of target region positioning.

Description

Technical Field

[0001] This disclosure relates to the field of regional positioning technology, and in particular to a method and apparatus for locating a target area. Background Technology

[0002] For locations such as garage entrances and exits, automatic license plate recognition systems are required. Existing automatic license plate recognition systems generally rely on cameras and supplementary lighting to achieve automatic vehicle identification. However, in the process of using automatic license plate recognition in the field, there are inevitably some backlighting scenarios, such as sunlight, other vehicles, or ambient light sources. In order to accurately identify license plate information, it may be necessary to take multiple shots while adjusting the shooting angle, thus resulting in a need to improve the efficiency of vehicle license plate positioning. Summary of the Invention

[0003] The purpose of this disclosure is to provide a method and device for locating a target area, which improves the positioning efficiency to at least a certain extent while ensuring the accuracy of license plate positioning.

[0004] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part by practice of this disclosure.

[0005] According to one aspect of this disclosure, a method for locating a target region is provided. The method includes: acquiring a first point cloud corresponding to a target object; determining the pose information of the target object based on the first point cloud; determining an initial search position for a target region in the target object based on the pose information; projecting the first point cloud onto a first plane, and determining multiple candidate regions based on the initial search position and the reflection intensity of the point cloud projected onto the first plane; and locating the target region based on the multiple candidate regions and image data.

[0006] According to another aspect of this disclosure, a target area positioning device is provided, the device comprising: a point cloud determination module, an attitude information determination module, an initial search position determination module, a candidate area determination module, and a positioning module.

[0007] The aforementioned point cloud determination module is used to acquire a first point cloud corresponding to the target object; the aforementioned posture information determination module is used to determine the posture information of the target object based on the aforementioned first point cloud; the aforementioned initial search position determination module is used to determine an initial search position for the target region in the target object based on the aforementioned posture information; the aforementioned candidate region determination module is used to project the aforementioned first point cloud onto a first plane, and determine multiple candidate regions based on the aforementioned initial search position and the reflection intensity of the point cloud projected onto the aforementioned first plane; and the aforementioned positioning module is used to locate the aforementioned target region based on the aforementioned multiple candidate regions and image data.

[0008] According to another aspect of this disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the target region positioning method as described in the above embodiments.

[0009] According to another aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the target region localization method as described in the above embodiments.

[0010] The target area positioning method and target area positioning device provided in the embodiments of this disclosure have the following technical effects:

[0011] This technical solution first acquires a first point cloud corresponding to the target object. Then, based on the first point cloud, it determines the current pose information of the target object. Further, based on the pose information, it determines the initial search position of the target region within the target object. The first point cloud is projected onto a first plane, and based on the initial search position and the reflection intensity of the point cloud projected onto the first plane, multiple candidate regions are determined. Finally, based on the multiple candidate regions and image data, the target region is located. This technical solution achieves automatic target region location based on the characteristics of LiDAR scanning data, exhibiting high positioning accuracy and thus improving positioning efficiency.

[0012] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0013] Figure 1 This is a schematic diagram illustrating a scenario of a target area positioning scheme in an exemplary embodiment of the present disclosure.

[0014] Figure 2 This diagram illustrates a flowchart of a method for locating a target region in an exemplary embodiment of this disclosure.

[0015] Figure 3 This diagram illustrates a flowchart of a method for determining whether a target object appears in a target environmental region according to an exemplary embodiment of this disclosure.

[0016] Figure 4 A grid diagram reflecting background height information of a target environmental area is shown in an exemplary embodiment of this disclosure.

[0017] Figure 5 This diagram illustrates a flowchart of a method for determining whether a target object appears in a target environmental region in another exemplary embodiment of this disclosure.

[0018] Figure 6 A flowchart illustrating a method for locating a target region in another exemplary embodiment of this disclosure is shown.

[0019] Figure 7 This diagram illustrates a first point cluster in an exemplary embodiment of the present disclosure.

[0020] Figure 8 A schematic diagram illustrating the determination of target object posture information in an exemplary embodiment of this disclosure is shown.

[0021] Figure 9 A schematic diagram illustrating the determination of candidate regions in an exemplary embodiment of this disclosure is shown.

[0022] Figure 10 A schematic diagram of the structure of a target area positioning device according to an embodiment of the present disclosure is shown.

[0023] Figure 11 A schematic diagram of the structure of a target area positioning device according to another embodiment of the present disclosure is shown.

[0024] Figure 12 A schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this disclosure clearer, the embodiments of this disclosure will be described in further detail below with reference to the accompanying drawings.

[0026] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0027] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this disclosure more comprehensive and complete, and to fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, apparatus, steps, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0028] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0029] The following is a detailed description of the target area positioning method embodiments provided in this disclosure:

[0030] in, Figure 1 This is a schematic diagram illustrating a scenario for locating a target area in an exemplary embodiment of this disclosure. (See reference...) Figure 1 The scenario mainly includes: target object 110 (i.e., an object containing the target area, such as a vehicle containing a license plate), target environment area 120, measuring device 130 (such as lidar, camera device (with supplementary lighting components)), network 140, computing device 150, and display device 160.

[0031] The measuring device 130 is used to measure the target environment area 120. In this embodiment, a lidar is used as an example. The lidar scans the target environment area 120 according to a certain period. Since the scanning range required by the application scenario is determined according to the range of the target environment area 120 according to the period, the scanning period and scanning power of the lidar are controlled according to the range of the target environment area 120 and the actual needs, thereby reducing the scanning power consumption of the lidar.

[0032] Furthermore, the measurement results from the measuring device 130 determine whether a target object 110 appears in the target environment area 120. Further, if it is determined that a target object 110 appears in the target environment area 120, the measuring device 130 continues to measure, and the target area is located and its information is identified through processing the measurement results. The above calculation process can be implemented by the computing device 150, and the information identification results of the target area can also be displayed on the display device 160 for user viewing.

[0033] It should be noted that LiDAR scans objects within a relatively close range with high density, making it suitable for acquiring vehicle size information. Furthermore, the reflectivity of LiDAR is related to factors such as the vehicle's exterior material, distance, scanning angle, and emission power. Compared to other vehicle components, the license plate material has a higher reflectivity. Therefore, in the LiDAR's scan point cloud, the license plate area has a higher reflectivity. Thus, this technical solution uses LiDAR reflectivity to locate the license plate area, accelerating license plate recognition efficiency.

[0034] Figure 2 This diagram illustrates a flowchart of a target region localization method in an exemplary embodiment of this disclosure. (See reference...) Figure 2 The method includes:

[0035] S210, Obtain the first point cloud corresponding to the target object;

[0036] S220, determine the attitude information of the target object based on the first point above;

[0037] S230, determine the initial search position of the target region in the target object based on the above attitude information;

[0038] S240, the first point cloud is projected onto a first plane, and multiple candidate regions are determined based on the initial search position and the reflection intensity of the point cloud projected onto the first plane; and,

[0039] S250, based on the above multiple candidate regions and image data, the target region is located.

[0040] Figure 2 In the embodiment shown, the solution first obtains the point cloud (denoted as the first point cloud) corresponding to the target object. Then, based on the first point cloud, the current pose information of the target object is determined, and then the initial search position of the target region within the target object is determined based on the pose information. Determining the initial search position can effectively improve the search efficiency for the target region.

[0041] Furthermore, the aforementioned first point cloud is projected onto the first plane containing the target region, and multiple candidate regions are determined based on the initial search position and the reflection intensity of the point cloud projected onto the first plane. Finally, the target region is located based on the multiple candidate regions and image data. This scheme, by setting multiple candidate regions and combining image data for target region localization, helps to improve the search efficiency and localization accuracy of the target region.

[0042] This technical solution achieves automatic positioning of target area information based on the characteristics of LiDAR scanning data of the target area, with high positioning accuracy, thereby improving positioning efficiency.

[0043] The following will be about Figure 2 The specific implementation methods of each step in the illustrated embodiment are described in detail below:

[0044] In an exemplary embodiment, Figure 3 and Figure 5 Two technical solutions for determining whether a target object 110 is present in the target environment area 120 are shown.

[0045] in, Figure 3 This diagram illustrates a flowchart of a method for determining whether a target object appears in a target environmental region according to an exemplary embodiment of this disclosure. (See reference...) Figure 3 S310-S340 can be used as an embodiment for determining whether the probability of a target object existing in the target environment area is greater than a third preset threshold, and S350-S360 are embodiments after determining that the probability of a target object existing in the target environment area is greater than the third preset threshold.

[0046] In S310, the second point cloud cluster corresponding to the target environment area is obtained.

[0047] In an exemplary embodiment, reference is made to Figure 1 The target environment area 120 can be the area for automatic license plate recognition of vehicles. A measuring device 130 is set up based on the target environment area 120 to ensure that the main scanning points of the measuring device 130 cover the target environment area 120. The range of the target environment area 120 is related to the point cloud density acquired by the lidar in the measuring device 130 and the size characteristics of the target object.

[0048] It should be noted that, assuming no moving objects exist in the target environment area 120, background modeling of the target environment area 120 is performed using LiDAR, i.e., background height information of the area is obtained. For example, there are two exemplary methods for obtaining background height information:

[0049] (1) Static one-time acquisition method: When installing the measuring equipment, the point cloud data without foreground in the target environment area is acquired by LiDAR (which can be further rasterized to obtain the background height information corresponding to each grid). This method is suitable for scenarios where the target environment area is fixed at 120°. (2) Dynamic update method: This method dynamically updates the background by scanning the point cloud in real time with LiDAR. The update frequency can be controlled according to thresholds such as time and / or vertical height. Specifically, for example, an update time can be set, such as updating the grid height information of the target area every ten minutes. It is understandable that a height threshold can also be set, that is, when the area of ​​the grid height information change of the target area is greater than the first preset threshold and reaches the second preset threshold, it is checked whether the change time is continuously greater than the third threshold. When the change time is stably and continuously greater than the third threshold, the environmental background grid information is dynamically updated.

[0050] For example, background modeling is performed on the target environment region 120: such as Figure 4 As shown, background height information of the target environment area is obtained through lidar, where, Figure 4 The grayscale value of the raster is determined based on its background height information B(i,j). Here, B(i,j) is the maximum height among the scan points corresponding to raster (i,j). For example, the larger the value of B(i,j), the larger the grayscale value of the raster. Simultaneously, for ease of data processing, the target environment area is rasterized, and the rasterized background modeling information is represented as B(i,j). Here, B(i,j) is the background height of raster (i,j).

[0051] In this embodiment, within the t-th (positive integer) scan cycle of the lidar, a point cloud corresponding to the target environment region 120 in the t-th scan cycle is acquired (referred to as the "second point cloud" to distinguish it from other "point clouds"). In this embodiment, the second point cloud is used to determine whether a target object currently exists or has appeared within the target environment region, for example, whether a vehicle has entered the target region.

[0052] In S320, the aforementioned second point set is projected onto the second plane, and the projection of the aforementioned second point set onto the aforementioned second plane is rasterized to obtain the first raster set.

[0053] In this embodiment, the aforementioned second point cloud is rasterized within the second plane to determine the foreground using the current height information G(i,j) and background height information B(i,j) within the same raster (i,j). Here, the current height information G(i,j) is the maximum height of the scanned points within the second point cloud corresponding to raster (i,j).

[0054] In an exemplary embodiment, when the target area is a license plate, since license plates are generally located in a vertical plane, the first plane containing the target area is defined as a vertical plane to improve the detection accuracy of the license plate area, while the second plane perpendicular to the first plane is defined as a horizontal plane. It should be noted that the background modeling is also performed within this second plane (horizontal plane), and if rasterization has already been performed during background modeling, the current height information corresponding to any grid cell can be directly obtained based on the second point cloud without needing to rasterize again.

[0055] In S330, the target grid set whose height information change is greater than a first preset threshold is obtained from the first grid set.

[0056] For example, for a grid (i,j) in the first grid set, the height difference between the current height information G(i,j) and the background height information B(i,j) is calculated. If the height difference is greater than a first preset threshold, the grid (i,j) is recorded as the "target grid". Furthermore, the target grids with continuity are recorded as the above-mentioned target grid set.

[0057] In step S340, it is determined whether the area of ​​the target grid set is greater than a second preset threshold. This second preset threshold is related to the size characteristics of the target object. For example, if the projected area of ​​the target object in the second plane is 5 square meters, then the second preset threshold can range from 3 square meters to 5 square meters.

[0058] The first grid set may contain multiple consecutive target grid sets. In this embodiment, the target grid set with the largest area in the first grid set is used as the judgment target.

[0059] If the area of ​​the target grid set is greater than the second preset threshold, it means that the object currently appearing in the target environment area is more likely to be the target object (the probability of the target object appearing in the target environment area is greater than the third preset threshold, which is a value greater than 0.5 and less than 1).

[0060] In an exemplary embodiment, after determining that an object currently appearing within the target environment area has a high probability of being the target object, step S350 is executed: determining that the probability of the target object appearing within the target environment area is greater than a third preset threshold, and activating the camera device to acquire an image of the target environment area. When it is determined that a target object has a high probability of appearing or participating within the target environment area, activating the camera device further determines whether the current object is indeed the target object, improving recognition accuracy. Simultaneously, this avoids acquiring multiple useless images due to continuously keeping the camera device on, and reduces the power consumption problem caused by continuously keeping the camera device on.

[0061] If the area of ​​the target grid is not greater than the second preset threshold, it indicates that the object currently appearing in the target environment area is highly likely not the target object (the probability of the target object currently appearing in the target environment area is not greater than the third preset threshold). Then, S310 is executed again to obtain the second point cloud corresponding to the target environment area. In this embodiment, the area of ​​the target grid set (the target grid set with the largest area in the first point cloud) is used to initially determine whether a target object appears in the target environment area, thereby eliminating interference from interfering objects (such as pedestrians, non-motorized vehicles, animals, etc.). By using a simple foreground target area to filter out interfering objects, the activation of other devices (such as camera devices and computing devices) can be reduced, thereby further reducing the power consumption of the entire system.

[0062] Continue to refer to Figure 3 In S360, the target environment region is identified based on the image to determine whether the target object is present.

[0063] If the target object is not found in the target environment area identified by the image above, then S310 is executed again to obtain the second point cloud corresponding to the target environment area. By repeatedly executing the process of S310-S360, it is determined whether a target object has entered the target environment area.

[0064] If the target object is found to be present in the target environment region based on the image above, and it is determined that the target object has entered the target environment region, then S210 needs to be executed to obtain the first point cloud corresponding to the target object, so as to identify the target region in the target object based on the first point cloud.

[0065] in, Figure 5 This diagram illustrates a flowchart of a method for determining whether a target object appears in a target environmental region according to another exemplary embodiment of this disclosure. (See reference...) Figure 5 S310'-S340' can be used as another embodiment for determining whether the probability of a target object existing in the target environment area is greater than a third preset threshold, and S350-S360 are embodiments after determining that the probability of a target object existing in the target environment area is greater than a third preset threshold.

[0066] As an embodiment of determining whether the probability of a target object existing in the target environment area is greater than a third preset threshold, S310'-S340' is executed:

[0067] In S310', the second point cloud set of the target environment region is obtained, and the target environment region is divided into N regions, where N is a positive integer.

[0068] For example, the specific implementation of obtaining the second point cloud of the target environment area is the same as the specific implementation of S310, and will not be described again here.

[0069] In this embodiment, the target environment area can be divided into multiple regions according to actual needs. For example, the target environment area can be divided into regions based on the projected area of ​​the target object in the second plane. For instance, each divided region is not smaller than the projected area of ​​the target object in the second plane.

[0070] In S320', the average depth information D of the i-th region in the t-th scan cycle is obtained. i(t) And obtain the average depth information D of the i-th region in the (t+1)-th scan cycle. i(t+1) , where i is a positive integer not greater than N, and t is a positive integer. And, in S330', based on the average depth information D... i(t) and the average depth information D i(t+1) Determine the average depth change information of the i-th region.

[0071] For example, if the depth change information at the same location within different scanning cycles of a lidar can reflect whether a moving target exists at that location, then in this embodiment, the depth information of each region in each scanning cycle is calculated. Furthermore, for the same region, the depth change information of that region after consecutive scanning cycles is calculated.

[0072] For ease of calculation, in this embodiment, the average depth information of a region is used to represent the depth information of that region.

[0073] In S340', based on the average depth change information of the i-th region, it is determined whether the probability of the target object appearing in the target environment region is greater than a third preset threshold.

[0074] As a specific implementation of S340': when the number of regions with average depth change information greater than the fourth preset threshold is greater than the second preset threshold, it is determined that the probability of the target object appearing in the target environment region is greater than the third preset threshold.

[0075] In this embodiment, the fourth preset threshold can be determined based on the height of the target object. If the average depth change information is greater than the fourth preset threshold, it indicates that the target object may appear. Furthermore, if the region where the average depth change information is greater than the fourth preset threshold is continuous and has a large area (greater than the second preset threshold, and the second preset threshold is determined based on the projected area of ​​the target object in the second plane), it indicates that the probability of the target object appearing in the target environment region is greater than the third preset threshold.

[0076] As another specific implementation of S340': when the average depth change information of the target area is greater than a fourth preset threshold, it is determined that the probability of the target object appearing in the target environment area is greater than a third preset threshold.

[0077] In this embodiment, when dividing the target environment area, the area of ​​focus is determined as the aforementioned target area, such as a region within a preset distance of the clustering release bar. If the average depth change information of the aforementioned region is greater than a fourth preset threshold, it can be said that the probability of the target object appearing in the target environment area is greater than a third preset threshold.

[0078] As another specific implementation of S340': when the average depth change information of the i-th region is greater than the fourth preset threshold, it is determined that the probability of the target object appearing in the target environment region is greater than the third preset threshold.

[0079] In this embodiment, when dividing the target environment area, each divided area is smaller than the projected area of ​​the target object in the second plane. Furthermore, if the average depth change of a certain area is greater than a fourth preset threshold, it indicates that the probability of the target object appearing in the target environment area is greater than a third preset threshold.

[0080] If it is determined that the probability of the target object appearing in the target environment area is greater than a third preset threshold, then S350 is executed. That is, the camera device is activated to acquire an image of the target environment area, and S360: based on the image, it is identified whether the target object is contained in the target environment area.

[0081] The specific implementation methods of S350 and S360 are as follows: Figure 3 The corresponding embodiments are the same, and will not be repeated here.

[0082] In an exemplary embodiment, Figure 6 This diagram illustrates a flowchart of a method for locating a target region in another exemplary embodiment of this disclosure. The embodiment shown includes... Figure 2 Specific implementation methods for each step.

[0083] To enable faster location of the target area (license plate), this embodiment estimates the pose information of the target object (vehicle) using the first point aggregation described above. Specifically, steps S210 and S220 are executed. (See reference...) Figure 6 The specific implementation of S210 includes S2102-S2108, and the specific implementation of S220 includes S2202-S2208.

[0084] First, update the cloud aggregation mentioned above in steps S2102-S2108. Specifically:

[0085] In S2102, the point cloud corresponding to the above target raster set is determined as the above first point cloud set.

[0086] In an exemplary embodiment, if it is determined in S340 that the area of ​​the target raster set is greater than a second preset threshold, and the image in S360 contains a target object, then the point cloud corresponding to the target raster set is taken as the point cloud set corresponding to the target object, denoted as the "first point cloud set". Figure 7 As shown. In another exemplary embodiment, if it is determined in S340' that the number of regions greater than the fourth preset threshold is greater than the second preset threshold, and the image after S360 contains a target object, then the point cloud of the above-mentioned response region in the latest scanning cycle is taken as the point cloud set corresponding to the target object.

[0087] In an exemplary embodiment, to filter out unwanted point clouds and reduce computational load, this embodiment also determines the centroid corresponding to the target object based on the first point cloud set. The specific steps to be performed include: S2104, in the first point cloud set, obtaining M scanning points closest to the LiDAR scanning center, where M is a positive integer; and S2106, calculating the centroid of the M scanning points and determining the centroid of the M scanning points as the centroid corresponding to the target object.

[0088] Furthermore, after determining the centroid corresponding to the target object, S2108 is executed to filter the scan points in the first point cloud that are more than a fifth preset threshold away from the centroid, thus obtaining an updated first point cloud.

[0089] To accurately estimate the vehicle's attitude, this embodiment filters out scan points outside the range of a fifth preset threshold S at the centroid. In other words, the filtering process yields the minimum envelope point set of the target object. This scheme also includes a method for determining the fifth preset threshold: obtaining the standard size of the target object from a database of standard sizes for multiple objects; and determining the fifth preset threshold based on the standard size of the target object. That is, the fifth preset threshold S is related to the size characteristics of the target object. Since the vehicle's attitude is mostly facing the LiDAR directly, the value of the fifth preset threshold S should not exceed half the maximum size of a typical vehicle.

[0090] After updating the first point of the cloud, the attitude information of the target object is determined through steps S2202-S2208. Specifically:

[0091] In S2202, the updated first point cloud is projected onto the second plane to obtain the first projection point set.

[0092] In this embodiment, the updated first point set is projected onto the second plane (horizontal plane) to obtain a first projection point set. Further, based on the first projection point set, the minimum bounding rectangle of the target object within the second plane (horizontal plane) is determined.

[0093] The step of determining the minimum bounding rectangle of the target object in the second plane includes: S2204, performing dilation and erosion processing on the first projection point set in the manner of a binarized image to obtain a second projection point set; and S2206, determining the minimum axially bounding rectangle of the second projection point set to obtain the minimum bounding rectangle of the target object in the second plane.

[0094] Among them, the aforementioned minimum bounding rectangle (such as Figure 8 The long side of the minimum bounding rectangle (80) corresponds to the horizontal direction of the target object, and the wide side of the minimum bounding rectangle corresponds to the vertical direction of the target object.

[0095] In an exemplary embodiment, after determining the minimum bounding rectangle of the target object in the second plane (horizontal plane), S2208 is executed to determine the posture information of the target object based on the minimum bounding rectangle and the direction of motion of the target object.

[0096] refer to Figure 8 In the target environment area 120 mentioned above, if the geographical orientation is as shown in the figure “N(North, cup)-S(South, south)”, and the direction of movement of the target object is N', then the attitude information of the target object can be determined as the angle α between it and the geographical orientation “N”.

[0097] Continue to refer to Figure 2 In S230: An initial search position for the target region within the target object is determined based on the attitude information. (Exemplary reference...) Figure 6 As a specific implementation of S230, in S2302: the center position of the long side of the minimum bounding rectangle is used as the initial search position of the target area.

[0098] For example, after determining the orientation information of the target object, it can be determined that the bolded position in the smallest bounding rectangle in the horizontal plane corresponds to the license plate of the vehicle (target area), and this position is used as the initial search position of the target area.

[0099] In an exemplary embodiment, since the vehicle license plate is located in the vertical plane (first plane), and the initial search position is located in the horizontal plane (second plane), in S240, the first point cloud (specifically, the updated first point cloud) is projected onto the first plane.

[0100] refer to Figure 6 Then, through steps S2402-S2406, the first point cloud is projected onto the first plane, and multiple candidate regions are determined based on the initial search position and the reflection intensity of the point cloud projected onto the first plane. Specifically:

[0101] In S2402, the aforementioned first point cloud is rasterized in the second plane to obtain a second raster set. In S2404, the aforementioned second raster set is projected in polar coordinates to obtain a third raster set, thereby projecting the aforementioned first point cloud onto the first plane.

[0102] In this embodiment, the first point cloud determined in S210 (or the updated first point cloud determined in S2108) is rasterized in the second plane (horizontal plane) to obtain a second raster set. Further, the second raster set is projected in polar coordinates to obtain a third raster set (e.g., ...). Figure 9 (90) thus, the first point above can be projected onto the first plane (vertical plane).

[0103] In this context, the scan point contained in the (i,j)th grid in the third grid set is the same as the scan point contained in the (i,j)th grid in the second grid set, where i and j are both positive integers.

[0104] In an exemplary embodiment, in order to enable the third grid set in the polar coordinate projection to more robustly locate the license plate (the rest of the target) position, this embodiment also provides noise processing. Specifically, before determining multiple candidate regions, this technical solution further includes: removing outliers and filtering.

[0105] The implementation method for removing outliers is as follows: The centroid of the target object is determined based on the point cloud corresponding to the third grid set; scan points in the third grid set whose distance from the centroid is greater than a sixth preset threshold are identified as outliers; and these outliers are deleted from the scan points corresponding to the third grid set. This reduces interference from outliers on the reflection intensity of the lidar scan points.

[0106] The sixth preset threshold is determined as follows: based on the resolution of the polar coordinate grid, the distance between the centroid and the center of the lidar, and / or the scanning resolution of the lidar, the sixth preset threshold is determined. It is evident that the sixth preset threshold needs to consider two factors: (a) the resolution of the polar coordinate grid; and (b) the distance between the centroid and the center of the lidar, and the lidar's own scanning resolution. Factor (a) determines the distance differences between scan points that may fall within the grid; factor (b) determines the possible distance differences between lidar scan points.

[0107] The filtering process is implemented by performing mean filtering based on the reflection intensity of the point cloud corresponding to the third grid set. Specifically, the target area (license plate) of the target object (vehicle) is the high-reflectivity area within the object. To locate the high-reflectivity area, this embodiment performs mean filtering on the reflection intensity of the point cloud within the third grid set. For example, when performing the mean filtering, the range of the mean filtering needs to be determined. Due to regulations regarding vehicle license plates, the range of the mean filtering is no greater than 1 / 3 of the license plate width.

[0108] In an exemplary embodiment, after performing the above-described denoising process, step S240 is performed: based on the initial search position and the reflection intensity of the point cloud projected onto the first plane, a plurality of candidate regions are determined. These candidate regions include a first candidate region and a second candidate region.

[0109] refer to Figure 6 Specifically: in S2406, starting from the initial search position, multiple first selection boxes are determined in the first direction; the average reflection intensity X of the point cloud corresponding to the s-th first selection box projected onto the aforementioned first plane is calculated. s ; and based on the average value X of the reflection intensity s At least one of the aforementioned first candidate regions is determined from among the multiple first selection boxes. Here, s takes the value of a positive integer.

[0110] For example, from the initial search position (e.g. Figure 8 Zhongru Figure 9 Starting from the bolded position, a first selection box is defined at first preset distances along the first direction. For example, refer to... Figure 9 In the first direction, multiple search columns can be defined (e.g.) Figure 9 In the search column 901, multiple first selection boxes can be determined in the first direction for each search column. Further, the average reflection intensity X of the point cloud corresponding to the s-th (positive integer) first selection box projected onto the aforementioned first plane (vertical plane) is calculated. s ; and based on the average value X of the reflection intensity s At least one of the aforementioned first candidate regions is determined from among the multiple first selection boxes. For example, the average value X of the reflection intensity. s If the intensity value is greater than the preset value, then the first selection box is determined as the first candidate region (e.g., Figure 9 The first candidate region 1, the first candidate region 2, and the first candidate region 3 in search column 901.

[0111] Wherein, the aforementioned first preset distance is not less than the length of the aforementioned target area in the aforementioned first direction. When the aforementioned target area is a license plate in a vehicle, the aforementioned first preset distance is not less than the length of the license plate in the vertical direction (i.e., the width of the vehicle).

[0112] Continue to refer to Figure 6 In S2406', starting from the initial search position, multiple second selection boxes are determined in the second direction; the average reflection intensity Y of the point cloud projected onto the first plane corresponding to the d-th second selection box is calculated. d ; and based on the average value Y of the reflection intensity d At least one of the aforementioned second candidate regions is determined from among the multiple second selection boxes. Here, d is a positive integer, and the first direction is perpendicular to the second direction.

[0113] For example, from the initial search position (e.g. Figure 8 He Ru Figure 9 Starting from the bolded position, a second selection box is defined at second preset distances along the second direction. For example, refer to... Figure 9 In the second direction, multiple search rows can be determined (e.g.) Figure 9 In a search row 902). For each search row, multiple second selection boxes can be determined in the second direction. Further, the average reflection intensity Y of the point cloud corresponding to the d-th (positive integer) second selection box projected onto the aforementioned first plane (vertical plane) is calculated. d ; and based on the average value Y of the reflection intensity d At least one of the aforementioned second candidate regions is determined from among the multiple second selection boxes. For example, the average value of the reflection intensity Y d If the intensity value is greater than the preset value, then the second selection box is determined as the second candidate region (e.g., Figure 9 (Second candidate region A and second candidate region B in search row 902).

[0114] Wherein, the aforementioned second preset distance is not less than the length of the aforementioned target area in the aforementioned second direction. When the aforementioned target area is the license plate of a vehicle, the aforementioned second preset distance is not less than the horizontal length of the license plate (i.e., the length of the vehicle).

[0115] In an exemplary embodiment, for the first candidate region and the second candidate region, step S250 is executed: based on the plurality of candidate regions and image data, the target region is located. As a specific implementation method, refer to... Figure 6 Execute S2502 and S2504.

[0116] In S2502, based on the transformation relationship between the first coordinate system corresponding to the lidar and the second coordinate system corresponding to the camera device, the first candidate region and / or the second candidate region are transformed into the second coordinate system to obtain the candidate region coordinates.

[0117] The aforementioned lidar is used to acquire the first point cloud, and the aforementioned camera device is used to acquire the aforementioned image data. By transforming the candidate region determined from the point cloud into the second coordinate system corresponding to the image according to the transformation relationship between the first coordinate system corresponding to the lidar and the second coordinate system corresponding to the camera device, the coordinates of the candidate region are obtained.

[0118] In an exemplary embodiment, if the ambient light is insufficient, the system controls a supplementary lighting component or other device to supplement the ambient light, thereby improving the clarity of the image acquired by the camera device.

[0119] In S2504, image recognition is performed on the region corresponding to the coordinates of the candidate region in the above image data to achieve the localization of the target region.

[0120] For the candidate region determined based on reflection intensity in S240, that is, the location with a high probability of being the target region has been obtained. Further, the image information of the corresponding region is used to further verify whether the candidate region is the target region. For example, the image information of the corresponding region is used to further verify whether the candidate region is a vehicle license plate. In this embodiment, by setting the coordinates of the candidate region in the above image coordinate system in S2502, it is easier to obtain the image region at the corresponding coordinates. After obtaining the image region corresponding to the candidate region, image recognition is performed, and the image recognition result is used to determine whether the current candidate region is a vehicle license plate (target region).

[0121] For example, if the image recognition result determines that the current candidate region is not a vehicle license plate, then the coordinates of the next candidate region are transformed, and the response image information is used to verify whether the next candidate region is a vehicle license plate. If the image recognition result determines that the current candidate region is a vehicle license plate, then the target region is located. Furthermore, vehicle license plate information can also be obtained as the recognition result for the target region.

[0122] In an exemplary embodiment, after identifying the target area, the system controls the corresponding hardware operation, such as raising the vehicle access barrier. For example, at this time, the LiDAR also identifies obstacles under the barrier to prevent injury to pedestrians or non-motorized personnel. It should be noted that the scanned point cloud of the barrier itself needs to be excluded first to improve the robustness of obstacle detection.

[0123] It should be noted that steps S240 and S250 can be executed in two ways. One method is to execute S240 to obtain at least one candidate region (e.g., obtain a candidate region corresponding to a search row / column) while simultaneously executing S250 to determine if the current candidate region corresponds to a license plate. If it is determined that the current candidate region does not correspond to a license plate, S240 is executed again to obtain the next candidate region, thus looping through S240 and S250 until the vehicle's license plate is determined. The other method is to obtain all candidate regions through S240, and then execute S250 in parallel for all candidate regions.

[0124] It should be noted that the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may, for example, be executed synchronously or asynchronously in multiple modules.

[0125] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein. For details not disclosed in the apparatus embodiments of this disclosure, please refer to the embodiments of the method disclosed herein.

[0126] in, Figure 10 A schematic diagram of a target area positioning device applicable to an embodiment of this disclosure is shown. Please refer to... Figure 10 The positioning device for the target area shown in the figure can be implemented as a whole or part of an electronic device through software, hardware, or a combination of both, or it can be integrated as an independent module into an electronic device or server.

[0127] The target area positioning device 1000 in this embodiment includes: a point cloud set determination module 1010, an attitude information determination module 1020, an initial search position determination module 1030, a candidate area determination module 1040, and a positioning module 1050.

[0128] The point cloud determination module 1010 is used to acquire a first point cloud corresponding to the target object; the pose information determination module 1020 is used to determine the pose information of the target object based on the first point cloud; the initial search position determination module 1030 is used to determine the initial search position of the target region in the target object based on the pose information; the candidate region determination module 1040 is used to project the first point cloud onto a first plane and determine multiple candidate regions based on the initial search position and the reflection intensity of the point cloud projected onto the first plane; and the positioning module 1050 is used to locate the target region based on the multiple candidate regions and image data.

[0129] In an exemplary embodiment, Figure 11 A schematic diagram illustrating a positioning device for a target area according to another exemplary embodiment of this disclosure is provided. See also... Figure 11 :

[0130] In an exemplary embodiment, based on the foregoing scheme, the positioning device 1000 for the target area further includes a target object determination module 1060.

[0131] The target object determination module 1060 is used to: before acquiring the point cloud data of the target object:

[0132] Obtain a second point cloud corresponding to the target environment area; project the second point cloud onto a second plane, and rasterize the projection of the second point cloud onto the second plane to obtain a first grid set; wherein the first plane is perpendicular to the second plane; obtain a target grid set in the first grid set whose height information change is greater than a first preset threshold; and, when the area of ​​the target grid set is greater than a second preset threshold, determine that the probability of the target object appearing in the target environment area is greater than a third preset threshold; wherein the second preset threshold is related to the size of the target object.

[0133] In an exemplary embodiment, based on the foregoing scheme, the target object determination module 1060 is further configured to, before obtaining the first point cloud corresponding to the target object:

[0134] Obtain the second point cloud of the target environment region, and divide the target environment region into N regions, where N is a positive integer; obtain the average depth information D of the i-th region in the t-th scan cycle. i(t) And obtain the average depth information D of the i-th region in the (t+1)-th scan cycle. i(t+1) i takes the value of a positive integer not greater than N, and t is a positive integer; according to the average depth information D i(t) and the average depth information D i(t+1) Determine the average depth change information of the i-th region, and, based on the average depth change information of the i-th region, determine whether the probability of the target object appearing in the target environment region is greater than a third preset threshold.

[0135] In an exemplary embodiment, based on the aforementioned scheme, the target object determination module 1060 is specifically used to: determine that the probability of the target object appearing in the target environment area is greater than the third preset threshold when the number of regions with average depth change information greater than the fourth preset threshold is greater than the second preset threshold.

[0136] In an exemplary embodiment, based on the aforementioned scheme, the target object determination module 1060 is specifically used to: determine that the probability of the target object appearing in the target environment area is greater than the third preset threshold when the average depth change information of the target area is greater than the fourth preset threshold.

[0137] In an exemplary embodiment, based on the aforementioned scheme, the target object determination module 1060 is specifically used to: determine that the probability of the target object appearing in the target environment area is greater than the third preset threshold when the average depth change information of the i-th region is greater than the fourth preset threshold.

[0138] In an exemplary embodiment, based on the foregoing solution, the target object determination module 1060 is further configured to, after determining that the probability of the target object appearing in the target environment area is greater than a third preset threshold:

[0139] The camera device is activated to acquire an image of the target environment area; the target environment area is identified as containing the target object based on the image; if the target environment area is not contained based on the image, a second point cloud corresponding to the target environment area is acquired again; and if the target environment area is contained based on the image, a first point cloud corresponding to the target object is acquired.

[0140] In an exemplary embodiment, based on the aforementioned scheme, the point cloud set determination module 1010 includes: a centroid determination unit 10101 and a filtering unit 10102.

[0141] The centroid determination unit 10101 is used to: determine the centroid of the target object based on the first point cloud; and the filtering unit 10102 is used to: filter the scan points in the first point cloud whose distance from the centroid is greater than a fifth preset threshold to obtain an updated first point cloud; wherein the updated first point cloud is the minimum envelope point cloud of the target object.

[0142] In an exemplary embodiment, based on the foregoing scheme, the centroid determination unit 10101 is specifically configured to: obtain the standard size of the target object from the standard sizes of multiple objects stored in the database; and determine the fifth preset threshold according to the standard size of the target object, wherein the fifth preset threshold is not greater than half of the standard size.

[0143] In an exemplary embodiment, based on the aforementioned scheme, the centroid determination unit 10101 is specifically used to: obtain M scanning points closest to the laser radar scanning center in the first point cloud, where M is a positive integer; and calculate the centroid of the M scanning points and determine the centroid of the M scanning points as the centroid corresponding to the target object.

[0144] In an exemplary embodiment, based on the aforementioned scheme, the posture information determination module 1020 includes: a first projection unit 10201, a minimum circumscribed rectangle determination unit 10202, and a posture information determination unit 10203.

[0145] The first projection unit 10201 is used to project the first points onto the second plane to obtain a first projection point set; the minimum bounding rectangle determination unit 10202 is used to determine the minimum bounding rectangle of the target object in the second plane based on the first projection point set, wherein the first plane is perpendicular to the second plane; and the attitude information determination unit 10203 is used to determine the attitude information of the target object based on the minimum bounding rectangle and the direction of motion of the target object.

[0146] In an exemplary embodiment, based on the aforementioned scheme, the minimum bounding rectangle determination unit 10202 is specifically used to: perform dilation and erosion processing on the first projection point set in the manner of a binarized image to obtain a second projection point set; and determine the minimum axially bounding rectangle of the second projection point set to obtain the minimum bounding rectangle of the target object in the second plane.

[0147] Wherein, the longer side of the aforementioned minimum bounding rectangle corresponds to the horizontal direction of the aforementioned target object, and the wider side of the aforementioned minimum bounding rectangle corresponds to the vertical direction of the aforementioned target object.

[0148] In an exemplary embodiment, based on the aforementioned scheme, the initial search position determination module 1030 is specifically used to: take the center position of the long side of the minimum bounding rectangle as the initial search position of the target area.

[0149] In an exemplary embodiment, based on the foregoing scheme, the candidate region determination module 1040 includes: a second projection unit 10401.

[0150] The second projection unit 10401 is used to: rasterize the first point cloud in the second plane to obtain a second raster set; and to project the second raster set in polar coordinates to obtain a third raster set, so as to project the first point cloud onto the first plane.

[0151] In this context, the scan point contained in the (i,j)th grid in the third grid set is the same as the scan point contained in the (i,j)th grid in the second grid set, where i and j are both positive integers.

[0152] In an exemplary embodiment, based on the foregoing scheme, the candidate region determination module 1040 further includes a noise reduction unit 10402.

[0153] The denoising unit 10402 is configured to: determine the centroid of the target object based on the point cloud corresponding to the third grid set; identify outliers among the scan points corresponding to the third grid set whose distance from the centroid is greater than a sixth preset threshold; and delete the outliers from the scan points corresponding to the third grid set.

[0154] In an exemplary embodiment, based on the foregoing scheme, the denoising unit 10402 is further configured to: determine the sixth preset threshold based on the resolution of the polar coordinate grid, the distance between the centroid and the center of the lidar, and / or the scanning resolution of the lidar.

[0155] In an exemplary embodiment, based on the foregoing scheme, the denoising unit 10402 is further configured to: perform mean filtering based on the reflection intensity of the point cloud corresponding to the third grid set.

[0156] In an exemplary embodiment, based on the aforementioned scheme, the candidate region includes a first candidate region and a second candidate region; the candidate region determination module 1040 further includes a first candidate region determination unit 10403 and a second candidate region determination unit 10404.

[0157] The first candidate region determination unit 10403 is configured to: determine multiple first selection boxes in a first direction starting from the initial search position; and calculate the average reflection intensity X of the point cloud projected onto the first plane corresponding to the s-th first selection box. s ; and based on the average value X of the reflection intensity s The system determines at least one first candidate region from among the plurality of first selection boxes, where s is a positive integer; and the second candidate region determination unit 10404 is configured to: determine a plurality of second selection boxes in a second direction starting from the initial search position; and calculate the average value Y of the reflection intensity of the point cloud projected onto the first plane corresponding to the d-th second selection box. d ; and based on the average value Y of the reflection intensity d At least one of the aforementioned second candidate regions is determined from among the aforementioned multiple second selection boxes; wherein d takes the value of a positive integer, and the aforementioned first direction is perpendicular to the aforementioned second direction.

[0158] In an exemplary embodiment, based on the foregoing scheme, the first candidate region determination unit 10403 is specifically used to: starting from the initial search position, determine a first selection box at intervals of a first preset distance in the first direction; wherein, the first preset distance is not less than the length of the target region in the first direction;

[0159] The second candidate region determination unit 10404 is specifically used to: starting from the initial search position, determine a second selection box at intervals of a second preset distance in the second direction; wherein the second preset distance is not less than the length of the target region in the second direction.

[0160] In an exemplary embodiment, based on the foregoing scheme, the positioning module 1050 is specifically configured to: transform the first candidate region and / or the second candidate region into the second coordinate system according to the transformation relationship between the first coordinate system corresponding to the lidar and the second coordinate system corresponding to the camera device, thereby obtaining candidate region coordinates; wherein the lidar is used to acquire the first point cloud, and the camera device is used to acquire the image data; and to perform image recognition on the region corresponding to the candidate region coordinates in the image data, so as to achieve the positioning of the target region.

[0161] In an exemplary embodiment, based on the foregoing scheme, the positioning module 1050 is further specifically used to: perform image recognition on the region corresponding to the coordinates of the candidate region in the image data, and determine the candidate region as the target region if license plate information exists in the region corresponding to the coordinates of the candidate region.

[0162] In an exemplary embodiment, based on the foregoing scheme, the positioning device 1000 for the target area further includes: an identification result determination module 1070.

[0163] The aforementioned recognition result determination module 1070 is used to: determine the aforementioned license plate information as the information recognition result for the aforementioned target area.

[0164] It should be noted that the target area positioning device provided in the above embodiments is only illustrated by the division of the above functional modules when executing the target area positioning method. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the target area positioning device and the target area positioning method embodiments provided in the above embodiments belong to the same concept. Therefore, for details not disclosed in the device embodiments of this disclosure, please refer to the above embodiments of the target area positioning method of this disclosure, which will not be repeated here.

[0165] The sequence numbers of the embodiments disclosed above are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0166] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described in the foregoing embodiments. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0167] This disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of any of the methods described above.

[0168] Figure 12 This schematically illustrates a structural diagram of an electronic device according to an exemplary embodiment of the present disclosure. Please see [link to relevant documentation]. Figure 12 As shown, the electronic device 1200 includes a processor 1201 and a memory 1202.

[0169] In this embodiment, the processor 1201 is the control center of the computer system, and can be a processor of a physical machine or a processor of a virtual machine. The processor 1201 may include one or more processing cores, such as a 4-core processor or an 8-core processor. The processor 1201 can be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 1201 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state.

[0170] In this embodiment of the disclosure, the processor 1201 is specifically used for:

[0171] Obtain a first point cloud corresponding to the target object; determine the pose information of the target object based on the first point cloud; determine the initial search position of the target region in the target object based on the pose information; project the first point cloud onto a first plane, and determine multiple candidate regions based on the initial search position and the reflection intensity of the point cloud projected onto the first plane; and locate the target region based on the multiple candidate regions and image data.

[0172] Furthermore, the processor 1201 is specifically used for:

[0173] Before acquiring the point cloud data of the target object, a second point cloud corresponding to the target environment region is acquired; the second point cloud is projected onto a second plane, and the projection of the second point cloud onto the second plane is rasterized to obtain a first raster set; wherein, the first plane is perpendicular to the second plane; a target raster set in the first raster set whose height information change is greater than a first preset threshold is acquired; when the area of ​​the target raster set is greater than a second preset threshold, it is determined that the probability of the target object appearing in the target environment region is greater than a third preset threshold; wherein the second preset threshold is related to the size of the target object.

[0174] Furthermore, the processor 1201 described above is specifically used for:

[0175] Before acquiring the first point cloud corresponding to the target object, the second point cloud of the target environment region is acquired, and the target environment region is divided into N regions, where N is a positive integer; the average depth information D of the i-th region in the t-th scan cycle is acquired. i(t) And obtain the average depth information D of the i-th region in the (t+1)-th scan cycle. i(t+1) i takes the value of a positive integer not greater than N, and t is a positive integer; according to the average depth information D i(t) and the average depth information D i(t+1) Determine the average depth change information of the i-th region, and, based on the average depth change information of the i-th region, determine whether the probability of the target object appearing in the target environment region is greater than a third preset threshold.

[0176] Further, determining whether the probability of the target object appearing in the target environment area is greater than a third preset threshold based on the average depth change information of the i-th region includes: determining that the probability of the target object appearing in the target environment area is greater than the third preset threshold when the number of regions with average depth change information greater than a fourth preset threshold is greater than a second preset threshold.

[0177] Further, determining whether the probability of the target object appearing in the target environment area is greater than the third preset threshold based on the average depth change information of the i-th region includes: determining that the probability of the target object appearing in the target environment area is greater than the third preset threshold when the average depth change information of the target area is greater than the fourth preset threshold.

[0178] Further, determining whether the probability of the target object appearing in the target environment area is greater than a third preset threshold based on the average depth change information of the i-th region includes: determining that the probability of the target object appearing in the target environment area is greater than the third preset threshold if the average depth change information of the i-th region is greater than a fourth preset threshold.

[0179] Furthermore, the processor 1201 described above is specifically used for:

[0180] After determining that the probability of the target object appearing in the target environment area is greater than a third preset threshold, the camera device is activated to acquire an image of the target environment area; the target object is identified in the target environment area based on the image; if the target object is not identified in the target environment area based on the image, a second point cloud corresponding to the target environment area is acquired again; and if the target object is identified in the target environment area based on the image, a first point cloud corresponding to the target object is acquired.

[0181] Furthermore, the above-mentioned acquisition of the first point cloud corresponding to the target object includes: determining the point cloud corresponding to the multiple target grids as the first point cloud.

[0182] Furthermore, the processor 1201 described above is specifically used for:

[0183] After obtaining the first point cloud corresponding to the target object, the centroid of the target object is determined based on the first point cloud; and the scan points in the first point cloud whose distance from the centroid is greater than a fifth preset threshold are filtered to obtain the updated first point cloud; wherein the updated first point cloud is the minimum envelope point cloud of the target object.

[0184] Furthermore, the processor 1201 described above is specifically used for:

[0185] Obtain the standard size of the target object from the standard sizes of multiple objects stored in the database; and determine the fifth preset threshold based on the standard size of the target object, wherein the fifth preset threshold is not greater than half of the standard size.

[0186] Furthermore, determining the centroid of the target object based on the first point set includes: acquiring M scanning points closest to the laser radar scanning center from the first point set, where M is a positive integer; and calculating the centroid of the M scanning points and determining the centroid of the M scanning points as the centroid of the target object.

[0187] Furthermore, determining the attitude information of the target object based on the first point set includes: projecting the first point set onto a second plane to obtain a first projection point set; determining the minimum bounding rectangle of the target object in the second plane based on the first projection point set, wherein the first plane is perpendicular to the second plane; and determining the attitude information of the target object based on the minimum bounding rectangle and the direction of motion of the target object.

[0188] Further, determining the minimum bounding rectangle of the target object in the second plane based on the first projection point set includes: performing dilation and erosion processing on the first projection point set in the manner of a binarized image to obtain a second projection point set; and determining the minimum axially bounding rectangle of the second projection point set to obtain the minimum bounding rectangle of the target object in the second plane.

[0189] Wherein, the longer side of the aforementioned minimum bounding rectangle corresponds to the horizontal direction of the aforementioned target object, and the wider side of the aforementioned minimum bounding rectangle corresponds to the vertical direction of the aforementioned target object.

[0190] Furthermore, determining the initial search position of the target region in the target object based on the above-mentioned attitude information includes: taking the center position of the long side of the minimum bounding rectangle as the initial search position of the target region.

[0191] Furthermore, the above-mentioned projection of the second point cloud onto the first plane includes: rasterizing the first point cloud in the second plane to obtain a second raster set; and projecting the second raster set in polar coordinates to obtain a third raster set, so as to project the first point cloud onto the first plane.

[0192] In this context, the scan point contained in the (i,j)th grid in the third grid set is the same as the scan point contained in the (i,j)th grid in the second grid set, where i and j are both positive integers.

[0193] Furthermore, the processor 1201 described above is specifically used for:

[0194] Before determining multiple candidate regions, the centroid of the target object is determined based on the point cloud corresponding to the third grid set; scan points in the third grid set whose distance from the centroid is greater than a sixth preset threshold are identified as outliers; and the outliers are deleted from the scan points corresponding to the third grid set.

[0195] Furthermore, the processor 1201 described above is specifically used for:

[0196] Before determining multiple candidate regions, the sixth preset threshold is determined based on the resolution of the polar coordinate grid, the distance between the centroid and the center of the lidar, and / or the scanning resolution of the lidar.

[0197] Furthermore, the processor 1201 described above is specifically used for:

[0198] Before determining multiple candidate regions, mean filtering is performed based on the reflection intensity of the point cloud corresponding to the third grid set.

[0199] Furthermore, the aforementioned candidate regions include a first candidate region and a second candidate region;

[0200] The above-mentioned determination of multiple candidate regions based on the initial search position and the reflection intensity of the point cloud projected onto the first plane includes: determining multiple first selection boxes in a first direction starting from the initial search position; and calculating the average value X of the reflection intensity X of the point cloud projected onto the first plane corresponding to the s-th first selection box. s ; and based on the average value X of the reflection intensity s Among the aforementioned plurality of first selection boxes, at least one of the aforementioned first candidate regions is determined, where s is a positive integer; and, starting from the aforementioned initial search position, a plurality of second selection boxes are determined in a second direction; the average reflection intensity Y of the point cloud corresponding to the d-th second selection box projected onto the aforementioned first plane is calculated. d ; and based on the average value Y of the reflection intensity d At least one of the aforementioned second candidate regions is determined from among the aforementioned multiple second selection boxes; wherein d takes the value of a positive integer, and the aforementioned first direction is perpendicular to the aforementioned second direction.

[0201] Furthermore, starting from the initial search position, multiple first selection boxes are determined in the first direction, including: starting from the initial search position, determining one of the first selection boxes at a first preset distance interval in the first direction; wherein the first preset distance is not less than the length of the target area in the first direction;

[0202] Starting from the initial search position, multiple second selection boxes are determined in the second direction, including: starting from the initial search position, determining one second selection box at a second preset distance in the second direction; wherein the second preset distance is not less than the length of the target area in the second direction.

[0203] Furthermore, the method for locating the target region based on the multiple candidate regions and image data includes: transforming the first candidate region and / or the second candidate region into the second coordinate system according to the transformation relationship between the first coordinate system corresponding to the lidar and the second coordinate system corresponding to the camera device, thereby obtaining candidate region coordinates; wherein the lidar is used to acquire the first point cloud, and the camera device is used to acquire the image data; and performing image recognition on the region corresponding to the candidate region coordinates in the image data to achieve the location of the target region.

[0204] Furthermore, the above-mentioned image recognition of the region corresponding to the coordinates of the candidate region in the above-mentioned image data to locate the target region includes: performing image recognition on the region corresponding to the coordinates of the candidate region in the above-mentioned image data, and determining the candidate region as the target region when license plate information exists in the region corresponding to the coordinates of the candidate region.

[0205] Furthermore, the aforementioned license plate information is identified as the information recognition result for the aforementioned target area.

[0206] The memory 1202 may include one or more computer-readable storage media, which may be non-transitory. The memory 1202 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments of this disclosure, the non-transitory computer-readable storage media in the memory 1202 are used to store at least one instruction, which is executed by the processor 1201 to implement the methods in the embodiments of this disclosure.

[0207] In some embodiments, the electronic device 1200 further includes a peripheral device interface 1203 and at least one peripheral device. The processor 1201, memory 1202, and peripheral device interface 1203 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 1203 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of a display screen 1204, a camera 1205, and an audio circuit 1206.

[0208] Peripheral device interface 1203 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 1201 and memory 1202. In some embodiments of this disclosure, processor 1201, memory 1202, and peripheral device interface 1203 are integrated on the same chip or circuit board; in other embodiments of this disclosure, any one or two of processor 1201, memory 1202, and peripheral device interface 1203 can be implemented on separate chips or circuit boards. This disclosure does not specifically limit the scope of the embodiments.

[0209] Display screen 1204 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 1204 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 1201 for processing. In this case, display screen 1204 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments of this disclosure, there may be one display screen 1204, which serves as the front panel of electronic device 1200; in other embodiments, there may be at least two display screens 1204, respectively disposed on different surfaces of electronic device 1200 or in a folded design; in still other embodiments, display screen 1204 may be a flexible display screen, disposed on a curved or folded surface of electronic device 1200. Furthermore, display screen 1204 may also be configured as a non-rectangular irregular shape, i.e., a non-rectangular screen. The display screen 1204 can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0210] Camera 1205 is used to capture images or videos. Optionally, camera 1205 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the electronic device, and the rear-facing camera is located on the back of the electronic device. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusing the main camera and the depth-sensing camera, panoramic shooting by fusing the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments of this disclosure, camera 1205 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cool light flash, which can be used for light compensation at different color temperatures.

[0211] The audio circuit 1206 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals that are input to the processor 1201 for processing. For stereo sound acquisition or noise reduction purposes, there may be multiple microphones, each located in a different part of the electronic device 1200. The microphone may also be an array microphone or an omnidirectional microphone.

[0212] Power supply 1207 is used to supply power to various components in electronic device 1200. Power supply 1207 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 1207 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0213] The block diagram of the electronic device shown in this embodiment does not constitute a limitation on the electronic device 1200. The electronic device 1200 may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0214] In the description of this disclosure, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this disclosure based on the specific circumstances. Furthermore, in the description of this disclosure, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0215] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, equivalent variations made in accordance with the claims of this disclosure are still within the scope of this disclosure.

Claims

1. A method for locating a target area, characterized in that, The method includes: Obtain the second point cluster corresponding to the target environment area; The probability of a target object appearing in the target environment area is obtained based on the second point cluster of the target environment area. When the probability that the target object appears in the target environment area is greater than a third preset threshold; Obtain the first point cloud corresponding to the target object; The attitude information of the target object is determined based on the first point set; The initial search position for the target region in the target object is determined based on the posture information; The first point cloud is projected onto the first plane, and multiple candidate regions are determined based on the initial search position and the reflection intensity of the point cloud projected onto the first plane. Based on the multiple candidate regions and image data, the target region is located; Determining the pose information of the target object based on the first point set includes: Projecting the first point onto the second plane yields the first projection point set; Based on the first set of projection points, determine the minimum bounding rectangle of the target object in the second plane; The pose information of the target object is determined based on the minimum bounding rectangle and the direction of motion of the target object; Determining the initial search position of the target region in the target object based on the posture information includes: The center position of the longer side of the smallest bounding rectangle is used as the initial search position of the target region.

2. The method according to claim 1, characterized in that, The step of obtaining the probability of the target object appearing in the target environment area based on the second point aggregation of the target environment area specifically includes: The second point cloud is projected onto the second plane, and the projection of the second point cloud onto the second plane is rasterized to obtain the first raster set. Obtain the target grid set whose height information change is greater than a first preset threshold from the first grid set; When the area of ​​the target grid set is greater than a second preset threshold, the probability that the target object appears in the target environment area is greater than a third preset threshold; wherein, the second preset threshold is related to the size of the target object.

3. The method according to claim 1, characterized in that, The step of obtaining the probability of the target object appearing in the target environment area based on the second point aggregation of the target environment area specifically includes: The target environment area is divided into N regions, where N is a positive integer; Obtain the average depth information D of the i-th region in the t-th scan cycle. i(t) And obtain the average depth information D of the i-th region in the (t+1)-th scan cycle. i(t+1) i takes the value of a positive integer not greater than N, and t is a positive integer; According to the average depth information D i(t) and the average depth information D i(t+1) Determine the average depth change information of the i-th region, and, based on the average depth change information of the i-th region, determine whether the probability of the target object appearing in the target environment region is greater than a third preset threshold.

4. The method according to claim 3, characterized in that, The step of determining whether the probability of the target object appearing in the target environment area is greater than a third preset threshold based on the average depth change information of the i-th region includes: If the number of regions with average depth change information greater than the fourth preset threshold is greater than the second preset threshold, then the probability that the target object appears in the target environment region is greater than the third preset threshold.

5. The method according to claim 1, characterized in that, After obtaining the first point cloud corresponding to the target object, the method further includes: The centroid of the target object is determined based on the first point set. The first point cloud is filtered out by the scan points whose distance from the centroid is greater than the fifth preset threshold, and the updated first point cloud is obtained. The updated first point cloud is the minimum envelope point cloud set of the target object.

6. The method according to claim 5, characterized in that, Determining the centroid of the target object based on the first point set includes: In the first point cloud set, obtain the M scanning points closest to the LiDAR scanning center, where M is a positive integer; Calculate the centroids of the M scan points and determine the centroids of the M scan points as the centroids corresponding to the target object.

7. The method according to any one of claims 1 to 6, characterized in that, The candidate regions include a first candidate region and a second candidate region; The step of determining multiple candidate regions based on the initial search position and the reflection intensity of the point cloud projected onto the first plane includes: Starting from the initial search position, multiple first selection boxes are determined in a first direction; the average reflection intensity X of the point cloud corresponding to the s-th first selection box projected onto the first plane is calculated. s ; and based on the average value X of the reflection intensity s At least one first candidate region is determined from the plurality of first selection boxes, where s takes the value of a positive integer; Starting from the initial search position, multiple second selection boxes are determined in the second direction; the average reflection intensity Y of the point cloud corresponding to the d-th second selection box projected onto the first plane is calculated. d ; and based on the average value Y of the reflection intensity d At least one second candidate region is determined from the plurality of second selection boxes, where d takes the value of a positive integer; Wherein, the first direction is perpendicular to the second direction.

8. A positioning device for a target area, characterized in that, The device includes: The point cloud determination module is used to obtain the first point cloud corresponding to the target object; The attitude information determination module is used to determine the attitude information of the target object based on the first point set; An initial search position determination module is used to determine the initial search position of the target region in the target object based on the attitude information. The candidate region determination module is used to project the first point cloud onto a first plane and determine multiple candidate regions based on the initial search position and the reflection intensity of the point cloud projected onto the first plane. The positioning module is used to locate the target region based on the multiple candidate regions and image data; The attitude information determination module is specifically used for: Projecting the first point onto the second plane yields the first projection point set; Based on the first set of projection points, determine the minimum bounding rectangle of the target object in the second plane; The pose information of the target object is determined based on the minimum bounding rectangle and the direction of motion of the target object; The initial search position determination module is specifically used for: The center position of the longer side of the smallest bounding rectangle is used as the initial search position of the target region.

Images