Target object identification method and device

Abstract

The invention discloses a target object identification method and device which are applied to the field of feature identification, and the method comprises the steps: obtaining point cloud data which are scanned by scanning equipment and comprise a target object; performing data preprocessing on the point cloud data, and positioning the position information of the target object in the processed point cloud data and the point cloud data related to the position information according to the feature information of the target object; filling the position information and the point cloud data related to the position information into the processed point cloud data; when it is detected that the target object moves, obtaining new point cloud data of the target object; and updating the filled point cloud data according to the new point cloud data to obtain identification information of the target object. The position of the target object in the point cloud data can be positioned through the feature information of the target object. Even if the scanning equipment cannot acquire complete point cloud data after the target object is displaced, the complete point cloud data can still be recovered according to the feature information, so that the information of the target object is accurately acquired.

Description

Technical Field

[0001] This application relates to the field of feature recognition technology, and particularly to a method and device for identifying a target object. Background Art

[0002] LiDAR is a technology that measures the distance between a target object and a sensor through laser pulses. It calculates the distance by emitting laser pulses and measuring the time it takes for the pulses to return, thereby generating high-precision three-dimensional data.

[0003] In a traditional production environment, the observation angle of LiDAR is usually limited. Especially when the installation position and height are restricted, it may not be able to cover the entire area of the target object, resulting in the loss of positioning information. In the absence of accurate positioning, it is difficult to accurately determine the position and orientation of the target object in space. If the relative position of the target object is unclear or deviated, it will lead to the inability to accurately obtain the target object information.

[0004] Therefore, how to accurately obtain the target object information has become an urgent problem to be solved in this field. Summary of the Invention

[0005] This application provides a method and device for identifying a target object, aiming to accurately obtain the target object information.

[0006] To achieve the above object, this application provides the following technical solutions:

[0007] A method for identifying a target object, comprising:

[0008] Obtaining point cloud data containing the target object scanned by a scanning device;

[0009] Performing data preprocessing on the point cloud data to obtain processed point cloud data;

[0010] Locating the position information of the target object in the processed point cloud data according to the feature information of the target object, and the point cloud data related to the position information;

[0011] Filling the position information and the point cloud data related to the position information into the processed point cloud data to obtain filled point cloud data;

[0012] When it is detected that the target object moves, obtaining new point cloud data of the target object through the scanning device;

[0013] Updating the filled point cloud data according to the new point cloud data to obtain the identification information of the target object.

[0014] Optionally, the data preprocessing of the point cloud data to obtain the processed point cloud data includes:

[0015] Performing normalization processing on the point cloud data to obtain the normalized point cloud data;

[0016] Performing rasterization processing on the normalized point cloud data to obtain the processed point cloud data.

[0017] Optionally, the updating the filled point cloud data according to the new point cloud data to obtain the recognition information of the target object includes:

[0018] When the feature information of the target object in the new point cloud data is lost, converting the target object in the new point cloud data into image data;

[0019] Performing feature extraction on the image data to obtain the feature information of the target object;

[0020] Obtaining the coordinate information of the feature information of the target object in the image data and converting the coordinate information into three-dimensional space coordinates;

[0021] Comparing the three-dimensional space coordinates with the position information to obtain displacement data;

[0022] Updating the filled point cloud data according to the displacement data to obtain the recognition information of the target object.

[0023] Optionally, the updating the filled point cloud data according to the displacement data to obtain the recognition information of the target object includes:

[0024] For each point in the target object, adding the point to the displacement data to obtain a displacement point;

[0025] Filling the displacement point into the filled point cloud data to obtain the recognition information of the target object.

[0026] Optionally, the updating the filled point cloud data according to the new point cloud data to obtain the recognition information of the target object includes:

[0027] When the feature information of the target object in the new point cloud data is not lost, positioning the target position information of the target object in the new point cloud data according to the feature information of the target object, and the point cloud data related to the target position information;

[0028] Filling the target position information and the point cloud data related to the target position information into the filled point cloud data to obtain the recognition information of the target object.

[0029] An identification device for a target object, comprising:

[0030] A first acquisition unit for acquiring point cloud data containing the target object scanned by a scanning device;

[0031] A preprocessing unit for performing data preprocessing on the point cloud data to obtain processed point cloud data;

[0032] A positioning unit for positioning the position information of the target object in the processed point cloud data and the point cloud data related to the position information according to the feature information of the target object;

[0033] A filling unit for filling the position information and the point cloud data related to the position information into the processed point cloud data to obtain filled point cloud data;

[0034] A second acquisition unit for acquiring new point cloud data of the target object through the scanning device when it is detected that the target object moves;

[0035] An updating unit for updating the filled point cloud data according to the new point cloud data to obtain the identification information of the target object.

[0036] Optionally, the preprocessing unit is specifically configured to:

[0037] Perform normalization processing on the point cloud data to obtain normalized point cloud data;

[0038] Perform rasterization processing on the normalized point cloud data to obtain processed point cloud data.

[0039] Optionally, the updating unit is specifically configured to:

[0040] When the feature information of the target object is lost in the new point cloud data, convert the target object in the new point cloud data into image data;

[0041] Extract features from the image data to obtain the feature information of the target object;

[0042] Obtain the coordinate information of the feature information of the target object in the image data and convert the coordinate information into three-dimensional space coordinates;

[0043] Compare the three-dimensional space coordinates with the position information to obtain displacement data;

[0044] Update the filled point cloud data according to the displacement data to obtain the identification information of the target object.

[0045] Optionally, the updating unit is specifically configured to:

[0046] For each point in the target object, add the point to the displacement data to obtain a displaced point.

[0047] Fill the displaced points into the filled point cloud data to obtain the identification information of the target object.

[0048] Optionally, the updating unit is specifically configured to:

[0049] When the feature information of the target object in the new point cloud data is not lost, locate the target position information of the target object in the new point cloud data according to the feature information of the target object, and the point cloud data related to the target position information.

[0050] Fill the target position information and the point cloud data related to the target position information into the filled point cloud data to obtain the identification information of the target object.

[0051] The technical solution provided by this application obtains the point cloud data including the target object scanned by the scanning device; performs data preprocessing on the point cloud data, locates the position information of the target object in the processed point cloud data according to the feature information of the target object, and the point cloud data related to the position information; fills the position information and the point cloud data related to the position information into the point cloud data; when it is detected that the target object moves, obtains the new point cloud data of the target object through the scanning device; updates the filled point cloud data according to the new point cloud data to obtain the identification information of the target object. Through the feature information of the target object, the position of the target object in the point cloud data can be located. Even after the target object has been displaced and the scanning device cannot obtain complete point cloud data, the complete point cloud data can still be restored according to the feature information of the object, so as to accurately obtain the information of the target object. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0053] Figure 1 It is a flowchart of a method for identifying a target object provided by an embodiment of the present application;

[0054] Figure 2 It is a flowchart of a method for obtaining identification information provided by an embodiment of the present application;

[0055] Figure 3Flow chart of another method for obtaining identification information provided by an embodiment of the present application;

[0056] Figure 4 Flow chart of an identification device for a target object provided by an embodiment of the present application. Detailed implementation manners

[0057] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0058] In the present application, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "including a..." does not exclude the presence of additional identical elements in the process, method, article or device including the said element.

[0059] As Figure 1 shown, it is a flow chart of an identification method for a target object provided by an embodiment of the present application, including the following steps:

[0060] S101: Obtain point cloud data scanned by a scanning device and containing a target object.

[0061] Among them, point cloud data refers to a discrete point set on the surface of an object in space obtained by a three-dimensional scanning device. Each point has spatial coordinates (usually X, Y, Z) and possibly other attributes, such as color, intensity, normal, etc. Specifically, point cloud data is a set of scattered and unprocessed points. These points may contain noise and redundant information, have irregular shapes, and are difficult to directly use for further analysis or modeling.

[0062] Optionally, the scanning device includes but is not limited to: lidar, structured light sensor. For example, scan point cloud data containing a train carriage through lidar.

[0063] S102: Perform data preprocessing on the point cloud data to obtain processed point cloud data.

[0064] Among them, performing data preprocessing on the point cloud data aims to improve data quality, remove noise, reduce redundancy, and provide more accurate and efficient data for subsequent analysis and applications.

[0065] Optionally, in another embodiment of the present application, the specific implementation manner of step S102 includes:

[0066] Perform normalization processing on the point cloud data to obtain the normalized point cloud data.

[0067] It can be understood that, specifically for performing normalization processing on the point cloud data, the x and y coordinates in the point cloud data are multiplied by a fixed multiple and then rounded to an integer to obtain the normalized point cloud data. For example, if the x coordinate in the point cloud data is 3.1415 and the y coordinate is 1.128, after multiplying the x and y coordinates in the point cloud data by a fixed multiple (e.g., 100 times), the obtained normalized data is: x coordinate is 314 and y coordinate is 112.

[0068] Perform rasterization processing on the normalized point cloud data to obtain the processed point cloud data.

[0069] Among them, the specific implementation process of performing rasterization processing on the normalized point cloud data is as follows: Obtain the data range of the normalized point cloud data; divide the data range according to a fixed length to obtain multiple grids; for each point cloud data, fill the point cloud data into the corresponding grid according to the coordinates of the point cloud data to obtain the processed point cloud data.

[0070] It should be noted that duplicate point cloud data in the grid are regarded as the same point. For example, if the fixed length is 5 cm, all duplicate point cloud data within the range of 5*5*5 are regularized into the same three-dimensional space coordinate point.

[0071] For example, if the fixed length is 100, the data range of the normalized point cloud data includes the x data range and the y data range, x: [0, 300], y: [0, 500]. The x-axis is divided into 3 grids according to the fixed length: [0, 100], [100, 200], [200, 300]. The y-axis is divided into 5 grids: [0, 100], [100, 200], [200, 300], [300, 400], [400, 500]. The normalized point cloud data 1 (x = 35, y = 113) belongs to the grid (0, 1); the normalized point cloud data 2 (x = 265, y = 378) belongs to the grid (2, 3); the normalized point cloud data 3 (x = 154, y = 88) belongs to the grid (1, 0).

[0072] S103: According to the feature information of the target object, locate the position information of the target object in the processed point cloud data and the point cloud data related to the position information.

[0073] Among them, the feature information of the target object is three-dimensional features, including but not limited to: geometric shape, texture, color. For example, a bulging bump.

[0074] Optionally, the point cloud data related to the position information indicates points that surround the surface of the target object and describe its shape.

[0075] It can be understood that, according to the feature information of the target object, the position information of the target object in the processed point cloud data and the point cloud data related to the position information are determined. Specifically, a matching algorithm is used to match the feature information of the target object with the feature information in the point cloud data to determine the position information of the target object; according to the position information of the target object, the point cloud data related to the target object is filtered out.

[0076] S104: Fill the position information and the point cloud data related to the position information into the processed point cloud data to obtain the filled point cloud data.

[0077] It can be understood that by filling the position information and the point cloud data related to the position information into the processed point cloud data, the point cloud data before the target object moves can be obtained, that is, the filled point cloud data.

[0078] S105: When it is detected that the target object moves, new point cloud data of the target object is obtained through a scanning device.

[0079] It can be understood that when it is detected that the target object moves, new point cloud data of the target object is obtained through a scanning device, that is, the point cloud data after the target object moves.

[0080] S106: Update the filled point cloud data according to the new point cloud data to obtain the recognition information of the target object.

[0081] Among them, the recognition information of the target object helps to understand the latest state of the target object in three-dimensional space, including its position, shape, posture, and movement conditions, etc.

[0082] It can be understood that the filled point cloud data is updated according to the new point cloud data. Specifically, the new position information of the target object and the point cloud data related to the new position information are determined according to the feature information in the new point cloud data; according to the new position information of the target object and the point cloud data related to the new position information, the filled point cloud data is updated to obtain the recognition information of the target object.

[0083] Optionally, in another embodiment of the present application, the specific implementation manner of step S106 is as Figure 2 shown, and includes the following steps:

[0084] S201: When the feature information of the target object in the new point cloud data is lost, convert the target object in the new point cloud data into image data.

[0085] Among them, when the feature information of the target object in the new point cloud data is lost, that is, the scanning device cannot obtain all the new point cloud data that completely represents the target object, in order to obtain the new point cloud data of the complete target object, it is necessary to convert the target object in the new point cloud data into image data.

[0086] It can be understood that converting the target object in the new point cloud data into image data specifically means mapping the x and y coordinates of the points to the pixel positions of the image, and the z value is the pixel value in the grayscale image; finally, an image data is generated to display the new point cloud data.

[0087] S202: Extract features from the image data to obtain the feature information of the target object.

[0088] Among them, the feature information of the target object includes but is not limited to: geometric shape, texture, color.

[0089] S203: Obtain the coordinate information of the feature information of the target object in the image data and convert the coordinate information into three-dimensional space coordinates.

[0090] It can be understood that converting the coordinate information into three-dimensional space coordinates specifically means converting the pixel position into the x coordinate and the y coordinate, and converting the pixel value in the coordinate information into the z value.

[0091] S204: Compare the three-dimensional space coordinates with the position information to obtain displacement data.

[0092] It can be understood that comparing the three-dimensional space coordinates with the position information is to compare the moved position information with the position data before movement. Through the change of position, the displacement of the target object, that is, the displacement data, can be obtained.

[0093] S205: Update the filled point cloud data according to the displacement data to obtain the recognition information of the target object.

[0094] Optionally, it is also possible to supplement the missing new point cloud data by performing a spatial displacement on the point cloud data so that the point cloud data overlaps on the new point cloud data, thereby updating the filled point cloud data to obtain the recognition information of the target object.

[0095] Optionally, in another embodiment of the present application, the specific implementation manner of step S205 includes:

[0096] For each point in the target object, add the point to the displacement data to obtain a displaced point.

[0097] It can be understood that for each point in the target object, adding the point to the displacement data to obtain a displaced point is equivalent to performing a spatial displacement on the point cloud data to obtain the displaced point cloud data, that is, the displaced point.

[0098] Fill the displacement points into the filled point cloud data to obtain the recognition information of the target object.

[0099] It can be understood that when the new point cloud data of the complete target object is obtained (i.e., all displacement points), all the displaced displacement points are supplemented into the filled point cloud data to obtain the recognition information of the target object.

[0100] Optionally, in another embodiment of the present application, the specific implementation manner of step S106 is as Figure 3 shown, and includes the following steps:

[0101] S301: When the feature information of the target object in the new point cloud data is not lost, according to the feature information of the target object, locate the target position information of the target object in the new point cloud data and the point cloud data related to the target position information.

[0102] It should be noted that the specific implementation manner of step S301 can be correspondingly referred to step S103, which will not be elaborated here.

[0103] S302: Fill the target position information and the point cloud data related to the target position information into the filled point cloud data to obtain the recognition information of the target object.

[0104] It should be noted that the specific implementation manner of step S302 can be correspondingly referred to step S104, which will not be elaborated here.

[0105] In summary, through the feature information of the target object, the position of the target object in the point cloud data can be located. Even after the target object is displaced and the scanning device cannot obtain the complete point cloud data, the complete point cloud data can still be restored according to the feature information of the object, so as to accurately obtain the target object information.

[0106] As Figure 4 shown, it is a schematic architecture diagram of an identification device for a target object provided by an embodiment of the present application. The identification device includes: a first acquisition unit 100, a preprocessing unit 200, a positioning unit 300, a filling unit 400, a second acquisition unit 500, and an update unit 600.

[0107] The first acquisition unit 100 is used to acquire the point cloud data including the target object scanned by the scanning device.

[0108] The preprocessing unit 200 is used to perform data preprocessing on the point cloud data to obtain the processed point cloud data.

[0109] The preprocessing unit 200 is specifically configured to: perform normalization processing on the point cloud data to obtain the normalized point cloud data; perform rasterization processing on the normalized point cloud data to obtain the processed point cloud data.

[0110] The positioning unit 300 is configured to locate the position information of the target object in the processed point cloud data and the point cloud data related to the position information according to the feature information of the target object.

[0111] The filling unit 400 is configured to fill the position information and the point cloud data related to the position information into the processed point cloud data to obtain the filled point cloud data.

[0112] The second acquisition unit 500 is configured to, when it is detected that the target object moves, acquire new point cloud data of the target object through a scanning device.

[0113] The updating unit 600 is configured to update the filled point cloud data according to the new point cloud data to obtain the recognition information of the target object.

[0114] The updating unit 600 is specifically configured to: when the feature information of the target object is lost in the new point cloud data, convert the target object in the new point cloud data into image data; perform feature extraction on the image data to obtain the feature information of the target object; acquire the coordinate information of the feature information of the target object in the image data and convert the coordinate information into three-dimensional space coordinates; compare the three-dimensional space coordinates with the position information to obtain displacement data; update the filled point cloud data according to the displacement data to obtain the recognition information of the target object.

[0115] The updating unit 600 is specifically configured to: for each point in the target object, add the point to the displacement data to obtain a displaced point; fill the displaced point into the filled point cloud data to obtain the recognition information of the target object.

[0116] The updating unit 600 is specifically configured to: when the feature information of the target object is not lost in the new point cloud data, locate the target position information of the target object in the new point cloud data and the point cloud data related to the target position information according to the feature information of the target object; fill the target position information and the point cloud data related to the target position information into the filled point cloud data to obtain the recognition information of the target object.

[0117] In summary, through the feature information of the target object, the position of the target object in the point cloud data can be located. Even after the target object has been displaced and the scanning device cannot acquire complete point cloud data, the complete point cloud data can still be restored according to the feature information of the object, so as to accurately acquire the target object information.

[0118] In this specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for a system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the method embodiment. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative work.

[0119] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0120] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying a target object, characterized in that: include: Acquire point cloud data containing the target object scanned by the scanning device; Performing data preprocessing on the point cloud data to obtain processed point cloud data; Locating the position information of the target object in the processed point cloud data and the point cloud data related to the position information according to the feature information of the target object; Filling the position information and the point cloud data related to the position information into the processed point cloud data to obtain filled point cloud data; When the target object is detected to move, new point cloud data of the target object is acquired by the scanning device; The filled point cloud data is updated according to the new point cloud data to obtain identification information of the target object.

2. The method according to claim 1, characterized in that: The step of preprocessing the point cloud data to obtain processed point cloud data includes: Performing standardization processing on the point cloud data to obtain standardized point cloud data; The standardized point cloud data is rasterized to obtain processed point cloud data.

3. The method according to claim 1, characterized in that The updating of the filled point cloud data according to the new point cloud data to obtain identification information of the target object includes: When feature information of the target object in the new point cloud data is lost, converting the target object in the new point cloud data into image data; Extracting features from the image data to obtain feature information of the target object; Acquire coordinate information of the feature information of the target object in the image data, and convert the coordinate information into three-dimensional space coordinates; Comparing the three-dimensional space coordinates with the position information to obtain displacement data; The filled point cloud data is updated according to the displacement data to obtain identification information of the target object.

4. The method according to claim 3, characterized in that The updating of the filled point cloud data according to the displacement data to obtain identification information of the target object includes: For each point in the target object, adding the point to the displacement data to obtain a displacement point; The displacement points are filled into the filled point cloud data to obtain identification information of the target object.

5. The method according to claim 1, characterized in that The updating of the filled point cloud data according to the new point cloud data to obtain identification information of the target object includes: When the feature information of the target object in the new point cloud data is not lost, locating the target position information of the target object in the new point cloud data and the point cloud data related to the target position information according to the feature information of the target object; The target position information and the point cloud data related to the target position information are filled into the filled point cloud data to obtain identification information of the target object.

6. A target object recognition device, characterized in that: include: A first acquisition unit, used to acquire point cloud data containing a target object scanned by a scanning device; A preprocessing unit, used for performing data preprocessing on the point cloud data to obtain processed point cloud data; a positioning unit, configured to locate the position information of the target object in the processed point cloud data and the point cloud data related to the position information according to the feature information of the target object; A filling unit, used for filling the position information and the point cloud data related to the position information into the processed point cloud data to obtain the filled point cloud data; a second acquisition unit, configured to acquire new point cloud data of the target object through the scanning device when movement of the target object is detected; An updating unit is used to update the filled point cloud data according to the new point cloud data to obtain identification information of the target object.

7. The device according to claim 6, characterized in that The pre-processing unit is specifically used for: Performing standardization processing on the point cloud data to obtain standardized point cloud data; The standardized point cloud data is rasterized to obtain processed point cloud data.

8. The device according to claim 6, characterized in that The updating unit is specifically used for: When feature information of the target object in the new point cloud data is lost, converting the target object in the new point cloud data into image data; Extracting features from the image data to obtain feature information of the target object; Acquire coordinate information of the feature information of the target object in the image data, and convert the coordinate information into three-dimensional space coordinates; Comparing the three-dimensional space coordinates with the position information to obtain displacement data; The filled point cloud data is updated according to the displacement data to obtain identification information of the target object.

9. The device according to claim 8, characterized in that The updating unit is specifically used for: For each point in the target object, adding the point to the displacement data to obtain a displacement point; The displacement points are filled into the filled point cloud data to obtain identification information of the target object.

10. The device according to claim 6, characterized in that The updating unit is specifically used for: When the feature information of the target object in the new point cloud data is not lost, locating the target position information of the target object in the new point cloud data and the point cloud data related to the target position information according to the feature information of the target object; The target position information and the point cloud data related to the target position information are filled into the filled point cloud data to obtain identification information of the target object.

Images