A laser radar denoising method and device, vehicle and storage medium

Abstract

The application discloses a laser radar denoising method and device, a vehicle and a storage medium. The method comprises the following steps: acquiring point cloud data of a current frame; determining a current image moment of an outline corresponding to the point cloud data; filtering point cloud data of a target object in the point cloud data corresponding to each outline based on the current image moment and an adjacent image moment of an adjacent frame to obtain target point cloud data, wherein the adjacent frame comprises a previous frame of the current frame and a next frame of the current frame, and deformation of the target object between different frames is greater than a set deformation threshold. By filtering the point cloud data corresponding to the target object with deformation greater than the set deformation threshold in the current frame, the method can prevent the problem of laser radar false detection caused by dust and water mist.

Description

Technical Field

[0001] The present invention relates to the field of lidar technology, and in particular to a lidar noise reduction method, device, vehicle, and storage medium. Background Technology

[0002] Interference from dust and water mist on LiDAR point cloud data is a common challenge in the industry. The noise generated by dust and water mist can significantly affect the detection performance of LiDAR on obstacles, often resulting in false detections. Current technology cannot yet solve the impact of dust and water mist on obstacle detection in general scenarios. Therefore, a solution is urgently needed to address the problem of false detections caused by dust or water mist kicked up by vehicle wheels or operating modules during vehicle operation. Summary of the Invention

[0003] This invention provides a method, apparatus, vehicle, and storage medium for denoising lidar, in order to solve the problem of false detection of lidar due to the presence of dust and water mist in the prior art.

[0004] According to one aspect of the present invention, a lidar noise reduction method is provided, the method comprising:

[0005] Obtain the point cloud data for the current frame;

[0006] Determine the current image moments of the contour corresponding to the point cloud data;

[0007] Based on the current image moments and the adjacent image moments of adjacent frames, the point cloud data of the target object in the point cloud data corresponding to each contour is filtered to obtain the target point cloud data. The adjacent frames include the frame before the current frame and the frame after the current frame. The deformation of the target object between different frames is greater than a set deformation threshold.

[0008] According to another aspect of the present invention, a lidar noise reduction device is provided, the device comprising:

[0009] The acquisition module is used to acquire point cloud data for the current frame;

[0010] The determination module is used to determine the current image moments of the contour corresponding to the point cloud data;

[0011] The filtering module is used to filter the point cloud data of the target object in the point cloud data corresponding to each contour based on the current image moments and the adjacent image moments of adjacent frames to obtain the target point cloud data. The adjacent frames include the frame before the current frame and the frame after the current frame. The deformation of the target object between different frames is greater than a set deformation threshold.

[0012] According to another aspect of the present invention, a vehicle is provided, the vehicle comprising:

[0013] At least one lidar;

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; the at least one lidar communicatively connected to the at least one processor and the memory; wherein

[0016] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the lidar denoising method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the lidar denoising method according to any embodiment of the present invention.

[0018] This invention discloses a lidar noise reduction method, apparatus, vehicle, and storage medium. The method includes: acquiring point cloud data of a current frame; determining the current image moments of the contours corresponding to the point cloud data; and filtering the point cloud data of target objects in the point cloud data corresponding to each contour based on the current image moments and adjacent image moments of adjacent frames to obtain target point cloud data. The adjacent frames include the frame preceding the current frame and the frame following the current frame, and the deformation of the target object between different frames exceeds a set deformation threshold. This method, by filtering the point cloud data corresponding to target objects with deformation exceeding the set deformation threshold in the current frame, can prevent lidar false detections due to the presence of dust and water mist.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating a lidar noise reduction method according to Embodiment 1 of the present invention.

[0022] Figure 2This is a flowchart illustrating another lidar noise reduction method provided in Embodiment 1 of the present invention;

[0023] Figure 3 This is a flowchart illustrating a lidar noise reduction method provided in Embodiment 2 of the present invention;

[0024] Figure 4 This is a schematic diagram of the structure of a lidar noise reduction device provided in Embodiment 3 of the present invention;

[0025] Figure 5 This is a structural schematic diagram of a vehicle provided in Embodiment 4 of the present invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention. It should be understood that the various steps described in the method embodiments of the present invention can be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0027] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0030] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0031] Example 1

[0032] Figure 1 This is a flowchart illustrating a lidar denoising method according to Embodiment 1 of the present invention. This method is applicable to the detection of obstacles and can be executed by a lidar denoising device. The device can be implemented by software and / or hardware and is generally integrated into a vehicle. In this embodiment, the vehicle includes, but is not limited to, ordinary transport vehicles, special-purpose vehicles, and special-purpose vehicles.

[0033] like Figure 1 As shown, the lidar noise reduction method provided in Embodiment 1 of the present invention includes the following steps:

[0034] S110. Obtain the point cloud data of the current frame.

[0035] Here, the current frame can be the frame that needs to be processed. Point cloud data can refer to a set of vectors in a three-dimensional coordinate system.

[0036] In this embodiment, point cloud data in the current frame can be obtained using a lidar.

[0037] S120. Determine the current image moments of the contour corresponding to the point cloud data.

[0038] The outline can be the shape of an object in the current frame. Depending on the object's state, the outline of the same object can differ in different frames. The object in the current frame can be an obstacle, dust, or water vapor. The current image moment can be the image moment of the outline in the current frame. An image moment is the weighted average of the gray levels of certain specific pixels in an image (moment), or an attribute of the image that has a similar function or meaning. Image moments include zero-order moments, first-order moments, second-order moments, third-order moments, and Hu moments, etc.

[0039] In this embodiment, the contours of all objects in the current frame can be extracted, and the image moments of all contours in the current frame can be determined.

[0040] S130. Based on the current image moments and the adjacent image moments of adjacent frames, filter the point cloud data of the target object in the point cloud data corresponding to each contour to obtain the target point cloud data.

[0041] The adjacent frames include the frame before the current frame and the frame after the current frame, and the deformation of the target object between different frames is greater than a set deformation threshold.

[0042] In this context, adjacent frames can be frames adjacent to the current frame, such as the frame preceding and following the current frame. Adjacent image moments can be the image moments of contours in adjacent frames, which can be obtained from a database or calculated; this embodiment does not limit this. The target object can be an object whose deformation between different frames exceeds a set deformation threshold; for example, the target object can be dust or water mist. The set deformation threshold can be a pre-set deformation value, which can be set according to actual conditions. The target point cloud data can be data from which the point cloud data corresponding to the target object has been filtered out.

[0043] In this embodiment, since dust, water mist, and other particles in the air can affect the accuracy of LiDAR obstacle detection, the image moments of the contour in the current frame are compared with those in adjacent frames to determine whether the contour deformation in the current frame exceeds a set deformation threshold. If the contour deformation exceeds the threshold, it indicates that the contour has significant deformation across different frames, and the contour is likely dust or water mist. Therefore, the point cloud data corresponding to this contour can be deleted to reduce the impact of dust and water mist on obstacle detection.

[0044] This invention provides a lidar denoising method, comprising: acquiring point cloud data of the current frame; determining the current image moments of the contours corresponding to the point cloud data; and filtering the point cloud data of target objects in the point cloud data corresponding to each contour based on the current image moments and adjacent image moments of adjacent frames to obtain target point cloud data. The adjacent frames include the frame preceding the current frame and the frame following the current frame, and the deformation of the target object between different frames exceeds a set deformation threshold. This method, by filtering the point cloud data corresponding to target objects whose deformation exceeds the set deformation threshold in the current frame, can prevent false detections by lidar due to the presence of dust and water mist.

[0045] Based on the above embodiments, modified embodiments of the above embodiments are proposed. It should be noted that, in order to keep the description brief, only the differences from the above embodiments are described in the modified embodiments.

[0046] In one embodiment, determining the image moments of the contour corresponding to the point cloud data includes:

[0047] The point cloud data is mapped to a two-dimensional raster map, and the two-dimensional raster map is transformed to obtain a binary image; morphological operations are performed on the binary image; the contours included in the binary image after morphological operations are extracted; and the image moments of the contours are determined.

[0048] Two-dimensional raster maps can be based on two-dimensional grids. A raster map divides the environment into a series of grids, where each grid cell is given a possible value representing the probability of that cell being occupied. Binary images refer to images where each pixel has only two possible values ​​or grayscale levels. Binary images are often represented using terms like black and white, B&W, and monochrome images. In binary images, there are only two grayscale levels; that is, any pixel in the image has a grayscale value of either 0 or 255, representing black and white respectively. Morphological operations can be simple operations based on the shape of an image. Morphological operations are typically performed on binary images. Morphological operations can include erosion, dilation, opening, closing, etc.

[0049] In this embodiment, point cloud data can be rasterized. The core idea of ​​laser point cloud rasterization is to process the area scanned by the lidar using a grid. Each grid point cloud represents a small area in space, containing a portion of point cloud data. Point cloud rasterization processing is divided into two-dimensional rasterization and three-dimensional rasterization. Since three-dimensional point cloud data is time-consuming to process in real time, this embodiment can use two-dimensional rasterization to perform data dimensionality reduction, thereby improving the processing speed of the algorithm.

[0050] In this embodiment, point cloud data can be mapped to a two-dimensional raster map, and the pixel values ​​of the point cloud data portion in the two-dimensional raster map can be set to 255, while the pixel values ​​of the remaining portion can be set to 0, thereby converting it into a binary image. Morphological operations are then performed on the binary image to extract the contours included in the morphologically operated binary image, and the image moments of each contour are calculated.

[0051] This embodiment first maps point cloud data to a two-dimensional raster map, then converts the two-dimensional raster map into a binary image, and performs morphological operations on the binary image, which makes it easier to extract information from the image.

[0052] In one embodiment, performing morphological operations on the binary image includes:

[0053] The binary image is dilated to obtain connected components; the connected components are then eroded to obtain a morphologically operated binary image.

[0054] Dilation and erosion are two fundamental operations in digital morphology, typically used for binary images. Dilation and erosion apply to white areas (highlights), not black areas. Dilation expands the bright areas in a binary image, making them larger. Erosion expands the dark areas in a binary image, making them larger. A connected component generally refers to an image region composed of foreground pixels with the same pixel value and adjacent positions.

[0055] In this embodiment, a larger connected region can be obtained by dilating the binary image, and then the connected region can be eroded to obtain a morphologically operated binary image to remove discrete points on the contour edges, thereby enabling the extraction of a more accurate contour.

[0056] In one embodiment, after acquiring the point cloud data of the current frame, the method further includes one or more of the following:

[0057] Delete point cloud data within a first preset range of the lidar; perform upsampling operations on point cloud data within a second preset range of the lidar with a preset radius and a preset number of iterations, the target point cloud data including the upsampled point cloud data; use point cloud data located within a third preset range as the updated point cloud data for the current frame.

[0058] The distance value included in the second preset range is greater than the distance value included in the third preset range, and the distance value included in the third preset range is greater than the distance value included in the first preset range.

[0059] The lidar system is a radar system that uses laser beams to detect the position, velocity, and other characteristics of a target. The lidar can be installed at any suitable location on the vehicle. The first preset range can be a distance range relatively close to the lidar, the second preset range can be a distance range relatively far from the lidar, and the third preset range can be a distance between the first and second preset ranges. The values ​​of the different preset ranges can be set according to actual conditions, and this embodiment does not limit this.

[0060] The distance value can be the distance to the LiDAR. The preset radius can be a pre-set radius value, and the preset number of iterations can be a pre-set number of iterations required. Upsampling refers to any technique that can increase the image resolution. Upsampling methods can include interpolation, deconvolution, and unpooling. The preset radius and preset number of iterations for upsampling can be set according to actual conditions, and this embodiment does not limit them.

[0061] In this embodiment, the point cloud data to be processed can be divided into three parts, specifically based on the distance range from the lidar. For example, Figure 2This is a flowchart illustrating another lidar noise reduction method provided in Embodiment 1 of the present invention. The distance range can be divided into a first preset range, a second preset range, and a third preset range. The first preset range can be set to 0-20 cm, the second preset range can be set to greater than 200 cm, and the third preset range can be set to 20-200 cm.

[0062] For point cloud data within the first preset range, the point cloud data can be deleted because this part of the point cloud data is generally the point cloud of the vehicle itself and needs to be filtered out. For point cloud data within the second preset range, upsampling operations can be performed with a preset radius and a preset number of iterations to obtain point cloud data with richer density. This can ensure that the point cloud on the actual obstacle will not become thin due to the influence of dust and water mist. The preset radius and preset number of iterations can be set according to the computing power. For example, the preset radius can be set to 3 cm and the number of iterations can be set to 5. For point cloud data within the third preset range, dust and water mist can be filtered.

[0063] This embodiment, by dividing the point cloud data, enables more targeted processing of the point cloud data and improves computational efficiency. It is understood that, without considering computational efficiency savings, dust and water mist filtration can also be performed on point cloud data within the first and second preset ranges.

[0064] Example 2

[0065] Figure 3 This is a flowchart illustrating a lidar noise reduction method according to Embodiment 2 of the present invention. Embodiment 2 is an optimization based on the above embodiments. For details not covered in this embodiment, please refer to Embodiment 1.

[0066] like Figure 2 As shown in Embodiment 2 of the present invention, a lidar denoising method is provided, wherein the current image moment includes the current centroid position, and the adjacent image moments include the adjacent centroid positions, specifically including the following steps:

[0067] S210. Obtain the point cloud data of the current frame.

[0068] S220. Determine the current image moments of the contour corresponding to the point cloud data.

[0069] S230. For each contour, based on the current centroid position of the contour in the current frame and the adjacent centroid positions of the contour in the adjacent frames, determine whether the objects corresponding to the contour are the same objects.

[0070] The current centroid position can be the centroid position of the contour in the current frame, and the adjacent centroid positions can be the centroid positions of the contour in adjacent frames. The centroid position can be the location of the contour's centroid, and can be represented by coordinates or other forms. The centroid position can be obtained from the contour's image moments. The centroid positions of the same contour may differ in different frames.

[0071] In this embodiment, for each contour in the current frame, the distance between the centroid position of the contour in the current frame and the centroid position in the adjacent frame can be determined. If the distance between the centroid position of the contour in the current frame and the centroid position in the adjacent frame is large, it indicates that the objects corresponding to the contour in the current frame and the contour in the adjacent frame are not the same object, and step S250 can be continued.

[0072] S240. If yes, then based on the shape change information of the contour, determine whether to use the point cloud data corresponding to the contour as the point cloud data of the target object, and delete the point cloud data of the target object.

[0073] The morphological change information can be the morphological change status information of the contour, which may include changes in the size, position, pixels, etc. of the contour. This implementation does not limit this.

[0074] In this embodiment, the shape change information of the contour can be used to determine whether the change of the contour between two adjacent frames is too large. If the shape change of the contour between two adjacent frames exceeds a certain threshold, it indicates that the object corresponding to the contour is likely to be dust or water mist. The point cloud data corresponding to the contour can be used as the point cloud data of the target object, and the point cloud data of the target object can be deleted.

[0075] S250. Continue traversing the remaining contours of the current frame until all contours have been traversed.

[0076] In this embodiment, after processing one of the contours in the current frame, the remaining contours in the current frame can be traversed until all contours in the current frame have been traversed.

[0077] This invention provides a lidar denoising method in Embodiment 2. For each contour, based on the current centroid position of the contour in the current frame and the adjacent centroid positions of the contour in adjacent frames, it determines whether the object corresponding to the contour is the same object. If so, based on the morphological change information of the contour, it determines whether to use the point cloud data corresponding to the contour as the point cloud data of the target object, and deletes the point cloud data of the target object. The remaining contours in the current frame are then traversed until all contours have been traversed. This method further processes the contours in the current frame, comparing the centroid position and morphological change information of the contours in the current frame with the contours in adjacent frames. This allows for the identification of contours corresponding to the same object in different frames and the determination of whether the object corresponding to the contour is dust or water mist. This embodiment can prevent dust and water mist from obscuring the laser beam emitted by the lidar, thus preventing the lidar point cloud on real obstacles from becoming sparse and causing missed detection of real obstacles.

[0078] In one embodiment, determining whether the objects corresponding to the contour are the same object based on the current centroid position of the contour in the current frame and the adjacent centroid positions of the contour in adjacent frames includes:

[0079] Obtain adjacent centroid positions, including the previous centroid position of the previous frame and the next centroid position of the next frame.

[0080] If the distance between the current centroid position and the previous centroid position is less than a first preset threshold, and the distance between the current centroid position and the next centroid position is less than a second preset threshold, then the contours corresponding to the current frame, the previous frame, and the next frame are determined to belong to the same object.

[0081] Here, the previous centroid position can be the centroid position of the previous frame, and the next centroid position can be the centroid position of the next frame. Distance can be the distance between the two centroid positions in different frames. The first preset threshold and the second preset threshold can be pre-set distance thresholds. The first preset threshold and the second preset threshold can be the same or different.

[0082] In this embodiment, it can be determined whether the objects corresponding to the contour are the same object by judging the distance between the current centroid position and the adjacent centroid positions. It can be judged whether the distance between the current centroid position and the previous centroid position is less than a first preset threshold, and whether the distance between the current centroid position and the next centroid position is less than a second preset threshold.

[0083] For example, if the distance between the current centroid position and the previous centroid position is less than a first preset threshold, but the distance between the current centroid position and the next centroid position is greater than a second preset threshold, then the objects corresponding to the contours in the current frame and the previous frame are considered to be the same object, while the objects corresponding to the contours in the current frame and the next frame are not the same object. In this case, the centroid positions of other contours in the next frame can be compared with the centroid positions of contours in the current frame until a contour in the next frame that corresponds to the same object as the contour in the current frame is found, and vice versa. If no contour that corresponds to the same object as the contour in the current frame is found in the next frame, then the object corresponding to that contour can be considered to be dust or water mist, and the point cloud data corresponding to that contour in the current frame can be deleted. Only when the distance between the current centroid position and the previous centroid position is less than the first preset threshold, and the distance between the current centroid position and the next centroid position is less than the second preset threshold, are the objects corresponding to the contours in the current frame, the previous frame, and the next frame considered to be the same object.

[0084] This embodiment compares the centroid position of the contour in the current frame with the centroid positions of the contour in the previous and next frames to determine whether the objects corresponding to the contours in the current frame and the contours in adjacent frames are the same objects.

[0085] In one embodiment, the image moments include Hu moments, and the adjacent image moments include the Hu moments of the previous frame and the Hu moments of the next frame; correspondingly, determining whether to use the point cloud data corresponding to the contour as the point cloud data of the target object based on the shape change information of the contour includes:

[0086] Obtain the Hu moment of the previous frame and the Hu moment of the next frame;

[0087] If the difference between the Hu moment of the current frame and the Hu moment of the previous frame is greater than a third preset threshold, and the difference between the Hu moment of the current frame and the Hu moment of the next frame is greater than a fourth preset threshold, then the point cloud data corresponding to the contour is used as the point cloud data of the target object; otherwise, the point cloud data of the contour is retained.

[0088] Here, the Hu moment is a set of seven variables calculated using the central moment, which is invariant to image transformations. The first six moment invariants are invariant to translation, scaling, rotation, and mapping, while the seventh moment changes due to image mapping. The third and fourth preset thresholds can be pre-set values; they can be the same or different.

[0089] In this embodiment, Hu moments can be used to determine the morphological changes of a contour in adjacent frames. If the difference between the Hu moments of the contour in the current frame and the Hu moments of the contour in the previous frame is greater than a third preset threshold, and the difference between the Hu moments of the contour in the current frame and the Hu moments of the contour in the next frame is greater than a fourth preset threshold, it can be determined that the contour has a large deformation in different frames, indicating that the object corresponding to the contour is likely dust or water mist. In this case, the point cloud data corresponding to the contour can be used as the point cloud data of the target object; otherwise, the point cloud data of the contour is retained. It is understood that, in addition to Hu moments, other image moments can also be used to determine the morphological changes of a contour in adjacent frames.

[0090] This embodiment can determine whether the object corresponding to the contour is the target object (i.e., dust and water mist) by judging and comparing the changes in Hu moments of the contour of the current frame between different frames.

[0091] Example 3

[0092] Figure 4 This is a schematic diagram of a lidar noise reduction device provided in Embodiment 3 of the present invention. The device is applicable to the detection of obstacles. The device can be implemented by software and / or hardware and is generally integrated into a vehicle.

[0093] like Figure 4 As shown, the device includes:

[0094] The acquisition module 310 is used to acquire the point cloud data of the current frame;

[0095] The determining module 320 is used to determine the current image moments of the contour corresponding to the point cloud data;

[0096] The filtering module 330 is used to filter the point cloud data of the target object in the point cloud data corresponding to each contour based on the current image moments and the adjacent image moments of adjacent frames to obtain target point cloud data. The adjacent frames include the frame before the current frame and the frame after the current frame. The deformation of the target object between different frames is greater than a set deformation threshold.

[0097] This embodiment provides a lidar denoising device. The device includes: acquiring point cloud data of the current frame; determining the current image moments of the contours corresponding to the point cloud data; and filtering the point cloud data of target objects in the point cloud data corresponding to each contour based on the current image moments and adjacent image moments of adjacent frames to obtain target point cloud data. The adjacent frames include the frame before the current frame and the frame after the current frame, and the deformation of the target object between different frames is greater than a set deformation threshold. By filtering the point cloud data corresponding to target objects with deformation greater than the set deformation threshold in the current frame, this device can prevent lidar false detections caused by the presence of dust and water mist.

[0098] Furthermore, module 320 is defined as including:

[0099] The point cloud data is mapped to a two-dimensional raster map, and the two-dimensional raster map is transformed to obtain a binary image;

[0100] Perform morphological operations on the binary image;

[0101] Extract the contours included in the binary image after morphological operations;

[0102] Determine the image moments of the contour.

[0103] Furthermore, the morphological operation on the binary image includes:

[0104] Dilatation of the binary image yields connected components;

[0105] Erosion is performed on the connected components to obtain a binary image after morphological operations.

[0106] Furthermore, the current image moment includes the current centroid position, and the adjacent image moments include adjacent centroid positions. Correspondingly, the filtering module 330 includes:

[0107] The first judgment unit is used to determine, for each contour, whether the object corresponding to the contour is the same object based on the current centroid position of the contour in the current frame and the adjacent centroid positions of the contour in the adjacent frames.

[0108] The second judgment unit is used to determine, if yes, whether to use the point cloud data corresponding to the contour as the point cloud data of the target object based on the shape change information of the contour, and to delete the point cloud data of the target object.

[0109] The traversal unit is used to continue traversing the remaining contours of the current frame until all contours have been traversed.

[0110] Furthermore, the first judgment unit includes:

[0111] Obtain adjacent centroid positions, including the previous centroid position of the previous frame and the next centroid position of the next frame.

[0112] If the distance between the current centroid position and the previous centroid position is less than a first preset threshold, and the distance between the current centroid position and the next centroid position is less than a second preset threshold, then the contours corresponding to the current frame, the previous frame, and the next frame are determined to belong to the same object.

[0113] Furthermore, the image moments include Hu moments, and the adjacent image moments include the Hu moments of the previous frame and the Hu moments of the next frame; correspondingly, the second judgment unit includes:

[0114] Obtain the Hu moment of the previous frame and the Hu moment of the next frame;

[0115] If the difference between the Hu moment of the current frame and the Hu moment of the previous frame is greater than a third preset threshold, and the difference between the Hu moment of the current frame and the Hu moment of the next frame is greater than a fourth preset threshold, then the point cloud data corresponding to the contour is used as the point cloud data of the target object; otherwise, the point cloud data of the contour is retained.

[0116] Furthermore, the device also includes:

[0117] The deletion module is used to delete point cloud data within a first preset range of the lidar.

[0118] The upsampling module is used to perform upsampling operations on point cloud data within a second preset range of the lidar with a preset radius and a preset number of iterations. The target point cloud data includes the upsampled point cloud data.

[0119] The update module is used to use point cloud data located within a third preset range as the updated point cloud data of the current frame;

[0120] The distance value included in the second preset range is greater than the distance value included in the third preset range, and the distance value included in the third preset range is greater than the distance value included in the first preset range.

[0121] The aforementioned lidar denoising device can execute the lidar denoising method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0122] Example 4

[0123] Figure 5 This is a schematic diagram of the structure of a vehicle provided in an embodiment of the present invention, such as... Figure 5 As shown, the vehicle includes: at least one lidar 41, at least one processor 42, a memory 43 communicatively connected to the at least one processor, an input device 44, and an output device 45. Figure 5 Taking a lidar 41 and a processor 42 as an example; the lidar 41, processor 42, memory 43, input device 44, and output device 45 in the vehicle can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0124] The memory 43, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the lidar denoising method in this embodiment of the invention. The processor 42 executes various vehicle functions and data processing by running the software programs, instructions, and modules stored in the memory 43, thereby implementing the aforementioned lidar denoising method.

[0125] The memory 43 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, the memory 43 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 43 may further include memory remotely located relative to the processor 42, which can be connected to the vehicle via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0126] Input device 44 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the vehicle. Output device 45 may include display devices such as a display screen.

[0127] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0128] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0129] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0130] To provide interaction with the user, the systems and technologies described herein can be implemented in a vehicle having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the vehicle. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0131] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0132] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0133] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0134] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for denoising a lidar system, characterized in that, The method includes: Obtain the point cloud data for the current frame; Determine the current image moments of the contour corresponding to the point cloud data; Based on the current image moments and the adjacent image moments of adjacent frames, the point cloud data of the target object in the point cloud data corresponding to each contour is filtered to obtain the target point cloud data. The adjacent frames include the frame before the current frame and the frame after the current frame. The deformation of the target object between different frames is greater than a set deformation threshold. The current image moment includes the current centroid position, and the adjacent image moments include adjacent centroid positions. Correspondingly, filtering the point cloud data of the target object in the point cloud data corresponding to each contour based on the current image moment and the adjacent image moments of adjacent frames includes: For each contour, based on the current centroid position of the contour in the current frame and the adjacent centroid positions of the contour in the adjacent frames, it is determined whether the objects corresponding to the contour are the same objects. If so, based on the shape change information of the contour, determine whether to use the point cloud data corresponding to the contour as the point cloud data of the target object, and delete the point cloud data of the target object. Continue traversing the remaining contours of the current frame until all contours have been traversed; Specifically, when it is determined, based on the shape change information of the contour, that the deformation of the contour between adjacent frames exceeds a set deformation threshold, the point cloud data corresponding to the contour is determined to be the point cloud data of the target object.

2. The method according to claim 1, characterized in that, The step of determining the image moments corresponding to the contour of the point cloud data includes: The point cloud data is mapped to a two-dimensional raster map, and the two-dimensional raster map is transformed to obtain a binary image; Perform morphological operations on the binary image; Extract the contours included in the binary image after morphological operations; Determine the image moments of the contour.

3. The method according to claim 2, characterized in that, The morphological operations performed on the binary image include: Dilatation of the binary image yields connected components; Erosion is performed on the connected components to obtain a binary image after morphological operations.

4. The method according to claim 1, characterized in that, The step of determining whether the objects corresponding to the contour are the same object based on the current centroid position of the contour in the current frame and the adjacent centroid positions of the contour in adjacent frames includes: Obtain adjacent centroid positions, including the previous centroid position of the previous frame and the next centroid position of the next frame. If the distance between the current centroid position and the previous centroid position is less than a first preset threshold, and the distance between the current centroid position and the next centroid position is less than a second preset threshold, then the contours corresponding to the current frame, the previous frame, and the next frame are determined to belong to the same object.

5. The method according to claim 1, characterized in that, The image moments include Hu moments, and the adjacent image moments include the Hu moments of the previous frame and the Hu moments of the next frame; correspondingly, determining whether to use the point cloud data corresponding to the contour as the point cloud data of the target object based on the shape change information of the contour includes: Obtain the Hu moment of the previous frame and the Hu moment of the next frame; If the difference between the Hu moment of the current frame and the Hu moment of the previous frame is greater than a third preset threshold, and the difference between the Hu moment of the current frame and the Hu moment of the next frame is greater than a fourth preset threshold, then the point cloud data corresponding to the contour is used as the point cloud data of the target object; otherwise, the point cloud data of the contour is retained.

6. The method according to claim 1, characterized in that, After acquiring the point cloud data of the current frame, the process also includes one or more of the following: Delete point cloud data within the first preset range of the lidar; The point cloud data within a second preset range of the lidar is upsampled for a preset radius and a preset number of iterations, and the target point cloud data includes the upsampled point cloud data. Point cloud data located within the third preset range are used as the updated point cloud data for the current frame; The distance value included in the second preset range is greater than the distance value included in the third preset range, and the distance value included in the third preset range is greater than the distance value included in the first preset range.

7. A lidar noise reduction device, characterized in that, The device includes: The acquisition module is used to acquire point cloud data for the current frame; The determination module is used to determine the current image moments of the contour corresponding to the point cloud data; The filtering module is used to filter the point cloud data of the target object in the point cloud data corresponding to each contour based on the current image moments and the adjacent image moments of adjacent frames to obtain the target point cloud data. The adjacent frames include the frame before the current frame and the frame after the current frame. The deformation of the target object between different frames is greater than a set deformation threshold. The current image moment includes the current centroid position, and the adjacent image moments include adjacent centroid positions. Correspondingly, the filtering module includes: The first judgment unit is used to determine, for each contour, whether the object corresponding to the contour is the same object based on the current centroid position of the contour in the current frame and the adjacent centroid positions of the contour in the adjacent frames. The second judgment unit is used to determine, if yes, whether to use the point cloud data corresponding to the contour as the point cloud data of the target object based on the shape change information of the contour, and to delete the point cloud data of the target object. The traversal unit is used to continue traversing the remaining contours of the current frame until all contours have been traversed. Specifically, when it is determined, based on the shape change information of the contour, that the deformation of the contour between adjacent frames exceeds a set deformation threshold, the point cloud data corresponding to the contour is determined to be the point cloud data of the target object.

8. A vehicle, characterized in that, The vehicles include: At least one lidar; At least one processor; and A memory communicatively connected to the at least one processor; the at least one lidar communicatively connected to the at least one processor and the memory; wherein... The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the lidar denoising method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the lidar denoising method according to any one of claims 1-6.

Images