A lane line determination method, apparatus, device and storage medium

Abstract

A lane line determination method, device, equipment and storage medium are disclosed. The method comprises: acquiring an image to be recognized; performing filtering processing on the image to be recognized based on a preset convolution kernel set to obtain a filtered image; determining at least two target regions according to the filtered image; fitting the at least two target regions to obtain a lane line. Through the technical scheme of the present application, the lane line recognition can be completed without using a neural network, the cumbersome labeling and training process is saved, and the hardware cost of the algorithm is greatly reduced.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of image processing, and particularly relate to a lane line determination method, device, equipment and storage medium. BACKGROUND

[0002] In the work of highway traffic safety management, the situation of car compression caused by bad weather / traffic accidents occurs from time to time, especially in bad weather and at night. Car compression caused by car compression often leads to rear-end accidents, and even personnel casualties. Therefore, users need to place roadblocks in front of the accident point while handling the accident site, or use eye-catching means such as lights and alarms to remind the rear vehicles, which will consume additional labor costs, and the high-speed vehicles on the highway will also increase the personal risk of users.

[0003] In the application scenario of outdoor unmanned vehicles, the SLAM technology will fail due to the variability of the surrounding environment. Therefore, how to enable the outdoor early warning and inspection unmanned vehicle to freely navigate to the desired location becomes a major technical problem in this scenario. The outdoor early warning and inspection unmanned vehicle usually drives in the emergency lane, and the relatively fixed and continuous lane line becomes a good reference in this scenario. Correspondingly, the identification and tracking navigation algorithm of the lane line becomes a technical point to be solved. The conventional lane line identification algorithm is mostly based on neural network, which inevitably needs to collect a large number of samples for labeling and training, and the calculation based on neural network has a large hardware overhead. SUMMARY

[0004] Embodiments of the present application provide a lane line determination method, device, equipment and storage medium, which can avoid using neural network to complete lane line identification, save the cumbersome labeling and training process, and greatly reduce the hardware overhead of the algorithm.

[0005] According to an aspect of the present application, a lane line determination method is provided, comprising:

[0006] obtaining an image to be identified;

[0007] filtering the image to be identified based on a preset convolution kernel set to obtain a filtered image;

[0008] determining at least two target regions according to the filtered image;

[0009] fitting the at least two target regions to obtain a lane line.

[0010] According to another aspect of the present application, a lane line determination device is provided, which comprises:

[0011] an acquisition module configured to acquire an image to be recognized;

[0012] a filtering module configured to perform filtering processing on the image to be recognized based on a preset kernel set, to obtain a filtered image;

[0013] a determination module configured to determine at least two target regions according to the filtered image;

[0014] a fitting module configured to fit the at least two target regions, to obtain a lane line.

[0015] According to another aspect of the present application, an electronic device is provided, which comprises:

[0016] at least one processor; and

[0017] a memory connected to the at least one processor in communication; wherein

[0018] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the lane line determination method according to any one of the embodiments of the present application.

[0019] According to another aspect of the present application, a computer readable storage medium is provided, which stores computer instructions for enabling a processor to implement the lane line determination method according to any one of the embodiments of the present application when executed.

[0020] The embodiments of the present application acquire an image to be recognized, perform filtering processing on the image to be recognized based on a preset kernel set, to obtain a filtered image, determine at least two target regions according to the filtered image, fit the at least two target regions, and obtain a lane line. Through the technical solution of the present application, the lane line recognition can be completed without using a neural network, the cumbersome labeling and training process is saved, and the hardware cost of the algorithm is greatly reduced.

[0021] It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the following drawings only show some of the embodiments of the present application, and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can also be obtained without creative labor.

[0023] Figure 1 is a flow chart of a lane line determination method in the embodiments of the present application;

[0024] Figure 2 is a schematic diagram of an image to be identified in the embodiments of the present application;

[0025] Figure 3 is a schematic diagram of a gradient rising edge image in the embodiments of the present application;

[0026] Figure 4 is a schematic diagram of a gradient falling edge image in the embodiments of the present application;

[0027] Figure 5 is a schematic diagram of a merged image in the embodiments of the present application;

[0028] Figure 6 is a schematic diagram of a filtered image in the embodiments of the present application;

[0029] Figure 7 is a schematic diagram of a third image in the embodiments of the present application;

[0030] Figure 8 is a schematic diagram of a target region in the embodiments of the present application;

[0031] Figure 9 is a structural schematic diagram of a lane line determination device in the embodiments of the present application;

[0032] Figure 10 is a structural schematic diagram of an electronic device for implementing the lane line determination method in the embodiments of the present application. DETAILED DESCRIPTION

[0033] In order to make the person skilled in the art better understand the present application, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of the present application.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and in the above drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to such a process, method, product or device.

[0035] Embodiment one

[0036] Figure 1 is a flowchart of a lane line determination method in an embodiment of the present application. The embodiment can be applicable to the case of lane line determination. The method can be executed by a lane line determination device in an embodiment of the present application. The device can be realized in the form of software and / or hardware. As shown in the figure, the method specifically includes the following steps: Figure 1

[0037] S101, obtaining an image to be recognized.

[0038] In the embodiment, the image to be recognized can be an image to be subjected to lane line recognition.

[0039] Specifically, a surround-view camera is installed on an outdoor early-warning inspection unmanned vehicle, which can obtain a bird's-eye view containing lane lines, and the coordinates of the vehicle body in the bird's-eye view are with the vehicle head facing up. The obtained bird's-eye view is converted into an HLS (Hue, Lightness, Saturation) color space and the lightness color gamut (L) is extracted to obtain the image to be recognized.

[0040] S102, performing filtering processing on the image to be recognized based on a preset convolution kernel set to obtain a filtered image.

[0041] The preset convolution kernel set can be three custom operators (kernel).

[0042] It should be noted that the filtering processing can be a convolution operation on the image to be recognized by the preset convolution kernel set to complete the filtering of the image to be recognized.

[0043] The filtered image can be an image after filtering completion obtained by performing convolution operation on the image to be recognized by the preset convolution kernel set.

[0044] ​Specifically, the preset convolution kernel set is used for convolution operation on the to-be-recognized image, so as to obtain a filtered image.

[0045] S103, determining at least two target regions according to the filtered image.

[0046] It should be explained that the target region can be a region constituting a lane line.

[0047] Specifically, skeleton extraction and noise reduction filtering are performed on the filtered image, and at least two target regions are determined.

[0048] S104, fitting the at least two target regions to obtain a lane line.

[0049] It should be noted that the fitting operation can be fitting the at least two target regions to form a complete lane line, so as to complete lane line detection.

[0050] Specifically, the at least two target regions are fitted to obtain a lane line.

[0051] The embodiment of the application obtains a to-be-recognized image, performs filtering processing on the to-be-recognized image based on a preset convolution kernel set, obtains a filtered image, determines at least two target regions according to the filtered image, and fits the at least two target regions to obtain a lane line. Through the technical scheme of the application, the lane line recognition can be completed without using a neural network, the cumbersome labeling and training process is saved, and the hardware cost of the algorithm is greatly reduced.

[0052] Optionally, the preset convolution kernel set includes a preset rising edge convolution kernel, a preset falling edge convolution kernel, and a preset middle line convolution kernel.

[0053] Edge detection is essentially a filtering algorithm, the difference lies in the selection of the filter, and the filtering rule is completely consistent. In the embodiment of the application, the convolution kernel is used for convolution operation on the image, and the image can be subjected to arbitrary linear filtering processing according to the selection of the convolution kernel, and the calculation formula is as follows:

[0054]

[0055] Wherein, src: input image; kernel: convolution kernel; dst: result image; anchor: anchor point of convolution kernel, used for positioning the point of aligning the kernel matrix with the current processing pixel point, and the default anchor point is located at the center of the kernel, and the anchor point position has a great influence on the result of the convolution processing.

[0056] In the embodiment, the preset rising edge convolution kernel, the preset falling edge convolution kernel, and the preset middle line convolution kernel can be convolution kernels preset by a user according to actual conditions, and the embodiment does not limit this.

[0057] The preset rising edge kernel can be represented as: rising edge kernel. The preset falling edge kernel can be represented as: falling edge kernel. The preset middle line kernel can be represented as: middle line kernel, which can be a 15 row x 24 column matrix, for example, the first column of the matrix is all 1, the 24th column is all -1, and the rest of the numbers are 0.

[0058] Optionally, the filtering processing is performed on the to-be-recognized image based on the preset kernel to obtain a filtered image, including:

[0059] The gradient rising edge image and the gradient falling edge image corresponding to the to-be-recognized image are determined according to the to-be-recognized image, the preset rising edge kernel and the preset falling edge kernel.

[0060] The gradient rising edge image can be a gradient rising edge image in the horizontal direction of the to-be-recognized image, and the gradient falling edge image can be a gradient falling edge image in the horizontal direction of the to-be-recognized image.

[0061] Specifically, the gradient rising edge and the gradient falling edge in the horizontal direction of the to-be-recognized image are obtained by using the preset rising edge kernel and the preset falling edge kernel, that is, the gradient rising edge image and the gradient falling edge image corresponding to the to-be-recognized image are obtained by performing convolution operation on the to-be-recognized image using the preset rising edge kernel and the preset falling edge kernel.

[0062] The gradient rising edge image and the gradient falling edge image are merged to obtain a merged image.

[0063] The merged image can be an image obtained by merging the gradient rising edge image and the gradient falling edge image. Preferably, the merging operation can be, for example, subtracting the pixel value of the gradient falling edge image from the pixel value of the gradient rising edge image.

[0064] Specifically, the gradient rising edge image and the gradient falling edge image are merged to obtain a merged image.

[0065] The filtered image is determined according to the preset middle line kernel and the merged image.

[0066] Specifically, the convolution operation can be performed on the merged image using the preset middle line kernel to obtain the filtered image.

[0067] Optionally, at least two target regions are determined according to the filtered image, including:

[0068] The filtered image is binarized to obtain a first image.

[0069] It can be known that the binarization processing is to set the gray value of the pixel point on the image to 0 or 255, that is, the entire image presents a visual effect of only black and white.

[0070] The first image can be an image obtained by performing binarization processing on the filtered image.

[0071] Specifically, the gray value of the pixel point on the filtered image is set to 0 or 255, that is, the entire filtered image presents a visual effect of only black and white, and a first image after binarization processing is obtained.

[0072] If there is a pixel point with a preset value in the first image, an erosion operation is performed on the first image.

[0073] The preset value can be a value set by a user according to actual conditions, and preferably, the preset value can be 255.

[0074] It can be known that the erosion operation can be represented as detecting the image with a structure element, and finding the region in the image that can put down the structure element. The erosion is a process of eliminating boundary points and making the boundary shrink inward, and can be used to eliminate small and meaningless target objects. If there is a small connection between two target objects, a structure element with a large enough size can be selected to erode the small connection.

[0075] Specifically, if there is a pixel point with a value of 255 in the first image, an erosion operation is performed on the first image to obtain a first image after the erosion operation.

[0076] An opening operation is performed on the first image after the erosion operation to obtain a second image.

[0077] It can be known that the opening operation can be to erode the image first and then dilate, which can exclude small block objects (assuming that the object is brighter than the background). The result of the opening operation deletes the object region that cannot contain the structure element, smoothens the burr, and finally obtains the image after the de-noising processing of the original image while keeping the shape of the original image unchanged.

[0078] The second image can be a second image obtained by performing an opening operation on the first image after the erosion operation.

[0079] Specifically, an opening operation is performed on the first image after the erosion operation to obtain a second image.

[0080] The pixel value of the first image after the erosion operation is subtracted from the pixel value of the second image to obtain a third image.

[0081] The third image can be obtained by subtracting the pixel value of the second image after the opening operation from the pixel value of the first image after the erosion operation.

[0082] The third image can be obtained by subtracting the pixel value of the second image after the opening operation from the pixel value of the first image after the erosion operation.

[0083] In actual operation, the skeleton extraction is irreversible in the opening operation process; the skeleton is related to the shape of the target region. The skeleton can be calculated from the boundary of the region. A common method for extracting the skeleton is to reconstruct the opening operation to achieve continuous refinement of the target region while maintaining the connectivity of the endpoints and lines. The algorithm steps are as follows:

[0084] Input: img (the first image obtained after binarization processing)

[0085] Output: out (an image with the same size as img, with all pixel values initialized to 0)

[0086] While there is a pixel value of 255 in img (in this loop, the first image is continuously eroded until it is all black):

[0087] Erode the img image

[0088] Open operation on img

[0089] img2 = image before opening operation minus image after opening operation

[0090] out += img2

[0091] Output out

[0092] If the pixel values of the pixel points in the third image are all non-preset values, at least two initial regions composed of contour points with the same brightness value in the third image are obtained.

[0093] In actual operation, the contour can be simply understood as a curve connecting all continuous points (along the object boundary), and these points usually have the same color or brightness.

[0094] It should be noted that the initial region can be a lane line region composed of contour points with the same brightness value in the third image.

[0095] Specifically, if the pixel values of the pixel points in the third image are all non-preset values (such as 255), at least two initial regions composed of contour points with the same brightness value in the third image are obtained.

[0096] The at least two initial regions are screened to obtain at least two target regions.

[0097] In the embodiment, the screening can be performed according to the size of the contour area of the initial region.

[0098] Specifically, the filtering is performed according to the contour area of the detected at least two initial regions by contour finding, to obtain at least two target regions.

[0099] Optionally, the screening of the at least two initial regions to obtain the at least two target regions comprises:

[0100] The initial region with a region area less than a preset region area threshold in the at least two initial regions is deleted to obtain the at least two target regions.

[0101] The preset region area threshold can be a region area threshold preset by a user according to an actual situation, and the embodiment does not limit this.

[0102] Specifically, the initial region with a region area less than a preset region area threshold in the at least two initial regions is deleted to obtain the at least two target regions.

[0103] Optionally, the at least two target regions are fitted to obtain the lane line, comprising:

[0104] A center pixel point of the at least two target regions is obtained.

[0105] It should be noted that the center pixel point can be a pixel point of a center point of the contour of each target region.

[0106] Specifically, the center pixel point of the contour of each target region is obtained.

[0107] The center pixel points of the at least two target regions are fitted by a preset order of curve to obtain the lane line.

[0108] The preset order can be an order preset by a user according to an actual situation, and the embodiment does not limit this. Preferably, the preset order can be 1 order.

[0109] It should be noted that the curve fitting can be fitting the center pixel points of the at least two target regions to form a complete curve.

[0110] Specifically, the center pixel points of the at least two target regions are fitted by a preset order of curve, a polynomial equation can be generated according to a set polynomial power, and then a series of points are generated according to the equation to form a complete curve, so as to complete the lane line detection.

[0111] Optionally, the image to be recognized is obtained, comprising:

[0112] An initial image is obtained.

[0113] The initial image is an image in an RGB color space.

[0114] In actual operation, the initial image can be obtained by a panoramic camera installed on an outdoor early warning and inspection unmanned vehicle, i.e., a surround view camera.

[0115] The panoramic camera, also known as a 360-degree camera, is a camera that captures a full view of a scene. Unlike traditional cameras that capture a limited field of view, panoramic cameras can capture a complete 360-degree view, providing users with a full view of the scene. In recent years, this type of camera has become increasingly popular, especially in the automotive industry, where it is used to provide drivers with a bird's eye view of the vehicle.

[0116] The initial image in the RGB color space is converted into an initial image in the HLS color space.

[0117] H is hue, L is lightness, and S is saturation.

[0118] It can be known that the HLS color gamut has three components: hue, saturation, and lightness. The L component in HLS is lightness, and a lightness of 100 represents white, while a lightness of 0 represents black. Hue is the color appearance of a color, which simply means what color it is, such as blue, cyan, purple, etc. Saturation is the purity of a color, and the higher the saturation, the more intense the color, and the lower the saturation, the more subtle the color. Brightness is the brightness of a color, and the higher the brightness, the brighter the color, and the darker the brightness, the darker the color.

[0119] Specifically, the calculation process of converting the initial image in the RGB color space into the initial image in the HLS color space can be represented as follows:

[0120] RGB→HLS:

[0121] Let M = max(R, G, B), m = min(R, G, B), P = M + m, and D = M - m, then L = (M + m) / 2.

[0122] If M = m, then S = H = 0.

[0123] If M ≠ m,

[0124] If L < 0.5, then S = D / P.

[0125] If L ≥ 0.5, then S = D / (2 - P).

[0126] If R = M, then H = (G - B) / D.

[0127] If G = M, then H = 2 + (B - R) / D.

[0128] If B = M, then H = 4 + (R-G) / D;

[0129] If H >= 0, then H = H*60;

[0130] If H < 0, then H = H*60+360.

[0131] An initial image under the brightness L parameter is determined as the to-be-recognized image.

[0132] Specifically, an initial image under the brightness L parameter is determined as the to-be-recognized image.

[0133] As an exemplary description of an embodiment of the present application, Figure 2 is a schematic diagram of a to-be-recognized image in an embodiment of the present application. An original image containing a lane line is obtained through a surround-view camera installed on an outdoor early-warning patrol unmanned vehicle, an initial image under an RGB color space is converted into an initial image under an HLS color space, and an initial image under a brightness L parameter is determined as the to-be-recognized image, that is, a to-be-recognized image as shown in Figure 2 .

[0134] Figure 3 is a schematic diagram of a gradient rising edge image in an embodiment of the present application. After obtaining the to-be-recognized image, a gradient rising edge image corresponding to the to-be-recognized image is determined according to the to-be-recognized image and a preset rising edge convolution kernel, that is, a gradient rising edge image as shown in Figure 3 .

[0135] Figure 4 is a schematic diagram of a gradient falling edge image in an embodiment of the present application. After obtaining the to-be-recognized image, a gradient falling edge image corresponding to the to-be-recognized image is determined according to the to-be-recognized image and a preset falling edge convolution kernel, that is, a gradient falling edge image as shown in Figure 4 .

[0136] Figure 5 is a schematic diagram of a merged image in an embodiment of the present application. After determining the gradient rising edge image and the gradient falling edge image corresponding to the to-be-recognized image, the gradient rising edge image and the gradient falling edge image are merged to obtain a merged image, that is, a merged image as shown in Figure 5 .

[0137] Figure 6 is a schematic diagram of a filtered image in an embodiment of the present application. After obtaining the merged image, a filtered image is determined according to a preset center line convolution kernel and the merged image, that is, a filtered image as shown in Figure 6 .

[0138] Figure 7is a schematic view of a third image in an embodiment of the present application. After obtaining the filtered image, the filtered image is binarized to obtain a first image. If there is a pixel point with a preset value in the first image, an erosion operation is performed on the first image, an opening operation is performed on the first image after the erosion operation, a second image is obtained, the pixel value of the first image after the erosion operation is subtracted by the pixel value of the second image, a third image is obtained, and the above process is repeated until the pixel value of the pixel point in the third image is not the preset value, and a third image as shown in Figure 7 is obtained.

[0139] Figure 8 is a schematic view of a target region in an embodiment of the present application. After obtaining the third image in which the pixel value of the pixel point is not the preset value, at least two initial regions composed of contour points with the same brightness value in the third image are obtained, an initial region with an area smaller than a preset region area threshold in the at least two initial regions is deleted, and at least two target regions as shown in Figure 8 are obtained. Then, the center pixel points of the at least two target regions are obtained, curve fitting of a preset order is performed on the center pixel points of the at least two target regions, and a lane line is obtained.

[0140] The technical scheme of the embodiment of the present application obtains an initial image, converts the initial image in the RGB color space into an initial image in the HLS color space, determines the initial image under the brightness L parameter as a to-be-recognized image, determines the gradient rising edge image and the gradient falling edge image corresponding to the to-be-recognized image according to the to-be-recognized image, a preset rising edge convolution kernel and a preset falling edge convolution kernel, merges the gradient rising edge image and the gradient falling edge image to obtain a merged image, determines a filtered image according to a preset center line convolution kernel and the merged image, performs binarization processing on the filtered image to obtain a first image, performs an erosion operation on the first image if there is a pixel point with a preset value in the first image, performs an opening operation on the first image after the erosion operation to obtain a second image, subtracts the pixel value of the first image after the erosion operation by the pixel value of the second image to obtain a third image, and obtains at least two initial regions composed of contour points with the same brightness value in the third image if the pixel value of the pixel point in the third image is not the preset value, deletes an initial region with an area smaller than a preset region area threshold in the at least two initial regions, obtains at least two target regions, obtains the center pixel points of the at least two target regions, and performs curve fitting of a preset order on the center pixel points of the at least two target regions to obtain a lane line. Through the technical scheme of the present application, the lane line recognition can be completed without using a neural network, the cumbersome labeling and training process is saved, and the hardware cost of the algorithm is greatly reduced.

[0141] Embodiment two

[0142] Figure 9 is a structural schematic diagram of a lane line determination device in an embodiment of the present application. The embodiment can be applicable to the case of lane line determination. The device can be realized in the form of software and / or hardware. The device can be integrated in any device that provides the function of lane line determination, such as a vehicle, a mobile terminal, a server, etc. Figure 9 As shown in the figure, the lane line determination device specifically comprises an acquisition module 201, a filtering module 202, a determination module 203 and a fitting module 204.

[0143] The acquisition module 201 is configured to acquire an image to be recognized.

[0144] The filtering module 202 is configured to perform filtering processing on the image to be recognized based on a preset kernel set to obtain a filtered image.

[0145] The determination module 203 is configured to determine at least two target regions according to the filtered image.

[0146] The fitting module 204 is configured to fit the at least two target regions to obtain a lane line.

[0147] Optionally, the preset kernel set comprises a preset rising edge kernel, a preset falling edge kernel and a preset middle line kernel.

[0148] Optionally, the filtering module 202 comprises:

[0149] A first determination unit is configured to determine a gradient rising edge image and a gradient falling edge image corresponding to the image to be recognized according to the image to be recognized, the preset rising edge kernel and the preset falling edge kernel.

[0150] A merging unit is configured to merge the gradient rising edge image and the gradient falling edge image to obtain a merged image.

[0151] A second determination unit is configured to determine a filtered image according to the preset middle line kernel and the merged image.

[0152] Optionally, the determination module 203 comprises:

[0153] A binarization unit is configured to perform binarization processing on the filtered image to obtain a first image.

[0154] An erosion unit is configured to perform an erosion operation on the first image if there is a pixel point with a preset value in the first image.

[0155] An opening operation unit is configured to perform an opening operation on the first image after the erosion operation to obtain a second image.

[0156] A third determination unit is configured to subtract the pixel value of the second image from the pixel value of the first image after the corrosion operation to obtain a third image.

[0157] A first acquisition unit is configured to acquire at least two initial regions composed of contour points with the same brightness value in the third image if the pixel value of each pixel point in the third image is not a preset value.

[0158] A screening unit is configured to screen the at least two initial regions to obtain at least two target regions.

[0159] Optionally, the screening unit is specifically configured to:

[0160] The screening unit deletes an initial region with an area smaller than a preset area threshold from the at least two initial regions to obtain the at least two target regions.

[0161] Optionally, the fitting module 204 includes:

[0162] A second acquisition unit is configured to acquire the center pixel points of the at least two target regions.

[0163] A fitting unit is configured to perform curve fitting of a preset order on the center pixel points of the at least two target regions to obtain a lane line.

[0164] Optionally, the acquisition module 201 includes:

[0165] A third acquisition unit is configured to acquire an initial image, wherein the initial image is an image in an RGB color space.

[0166] A conversion unit is configured to convert the initial image in the RGB color space into the initial image in an HLS color space, wherein H is a hue, L is a brightness, and S is a saturation.

[0167] A fourth determination unit is configured to determine the initial image in the brightness L parameter as a to-be-recognized image.

[0168] The product can execute the lane line determination method provided by any embodiment of the application and has the corresponding functional modules and beneficial effects of the lane line determination method.

[0169] Embodiment three

[0170] Figure 10A structural diagram of an electronic device 30 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices such as personal digital assistants, cellular telephones, smartphones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present application described and / or claimed in this document.

[0171] As shown, Figure 10 The electronic device 30 includes at least one processor 31, and a memory, such as a read-only memory (ROM) 32, a random access memory (RAM) 33, etc., connected to the at least one processor 31 in communication, where the memory stores computer programs executable by the at least one processor. The processor 31 can perform various appropriate actions and processes according to the computer programs stored in the read-only memory (ROM) 32 or loaded into the random access memory (RAM) 33 from the storage unit 38. In the RAM 33, various programs and data required for the operation of the electronic device 30 can also be stored. The processor 31, the ROM 32, and the RAM 33 are connected to each other through a bus 34. An input / output (I / O) interface 35 is also connected to the bus 34.

[0172] A plurality of components in the electronic device 30 are connected to the I / O interface 35, including an input unit 36, such as a keyboard, a mouse, etc., an output unit 37, such as various types of displays, speakers, etc., a storage unit 38, such as a magnetic disk, an optical disk, etc., and a communication unit 39, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 39 allows the electronic device 30 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunications networks.

[0173] The processor 31 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 31 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 31 performs various methods and processes described above, such as the lane line determination method:

[0174] Obtaining an image to be recognized;

[0175] filter the to-be-identified image based on a preset convolution kernel set to obtain a filtered image;

[0176] determine at least two target regions according to the filtered image;

[0177] fit the at least two target regions to obtain a lane line.

[0178] In some embodiments, the lane line determination method can be implemented as a computer program tangibly embodied in a computer readable storage medium, such as storage unit 38. In some embodiments, portions or all of the computer program can be loaded and / or installed onto electronic device 30 via, for example, ROM 32 and / or communication unit 39. When the computer program is loaded onto RAM 33 and executed by processor 31, one or more steps of the lane line determination method described above can be performed. Alternatively, in other embodiments, processor 31 can be configured to perform the lane line determination method by way of other any suitable means (e.g., by way of firmware).

[0179] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, specially designed application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0180] Computer programs used to implement the methods of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0181] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0182] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; 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 acoustic, speech, or tactile input.

[0183] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0184] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0185] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, each step described in the present application can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which is not limited herein.

[0186] The above detailed description does not constitute a limitation on the protection scope of the present application. 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 replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A lane line determination method characterized by, The method comprises the following steps: obtaining a to-be-recognized image; wherein the to-be-recognized image is obtained by converting a bird's-eye view containing lane lines obtained by a surround-view camera into an HLS color space and extracting a luminance color domain; performing filtering processing on the to-be-recognized image based on a preset kernel set to obtain a filtered image; determining at least two target regions according to the filtered image; wherein the target region is a region constituting a lane line, and the at least two target regions are determined by performing skeleton extraction and noise filtering on the filtered image; fitting the at least two target regions to obtain a lane line; wherein the preset kernel set comprises a preset rising edge kernel, a preset falling edge kernel and a preset middle line kernel; wherein the filtering processing on the to-be-recognized image based on the preset kernel set to obtain the filtered image comprises: determining a gradient rising edge image and a gradient falling edge image corresponding to the to-be-recognized image according to the to-be-recognized image, the preset rising edge kernel and the preset falling edge kernel; merging the gradient rising edge image and the gradient falling edge image to obtain a merged image; determining the filtered image according to the preset middle line kernel and the merged image; wherein the determination of the at least two target regions according to the filtered image comprises: performing binaryzation processing on the filtered image to obtain a first image; if there is a pixel point with a preset value in the first image, performing an erosion operation on the first image; performing an opening operation on the first image after the erosion operation to obtain a second image; subtracting the pixel value of the second image from the pixel value of the first image after the erosion operation to obtain a third image; if the pixel value of the pixel point in the third image is not the preset value, obtaining at least two initial regions composed of contour points with the same luminance value in the third image; performing screening on the at least two initial regions to obtain at least two target regions; wherein the fitting of the at least two target regions to obtain a lane line comprises: obtaining the center pixel points of the at least two target regions; performing curve fitting of a preset order on the center pixel points of the at least two target regions to obtain a lane line.

2. The method of claim 1, wherein, The screening of the at least two initial regions to obtain at least two target regions comprises: deleting an initial region with an area smaller than a preset area threshold from the at least two initial regions to obtain at least two target regions.

3. The method of claim 1, wherein, The obtaining of the to-be-recognized image comprises: obtaining an initial image; wherein the initial image is an image in an RGB color space; converting the initial image in the RGB color space into an initial image in an HLS color space; wherein H is a hue parameter, L is a luminance parameter, and S is a saturation parameter; determining the initial image under the luminance parameter L as the to-be-recognized image.

4. A lane line determination device characterized by comprising: The method comprises the following steps: an obtaining module is configured to obtain a to-be-recognized image; wherein the to-be-recognized image is obtained by converting a bird's-eye view containing lane lines obtained by a surround-view camera into an HLS color space and extracting a luminance color domain; The filtering module is configured to perform filtering processing on the to-be-identified image based on a preset kernel set to obtain a filtered image. The determining module is configured to determine at least two target regions from the filtered image; the target regions are regions constituting lane lines, and the determining at least two target regions from the filtered image specifically includes skeleton extraction and noise filtering on the filtered image. The fitting module is configured to fit the at least two target regions to obtain lane lines. The preset kernel set includes a preset rising edge kernel, a preset falling edge kernel, and a preset middle line kernel. The filtering module includes: The first determining unit is configured to determine a gradient rising edge image and a gradient falling edge image corresponding to the to-be-identified image based on the to-be-identified image, the preset rising edge kernel, and the preset falling edge kernel. The merging unit is configured to merge the gradient rising edge image and the gradient falling edge image to obtain a merged image. The second determining unit is configured to determine a filtered image based on the preset middle line kernel and the merged image. The determining module includes: The binarization unit is configured to perform binarization processing on the filtered image to obtain a first image. The erosion unit is configured to perform an erosion operation on the first image if there is a pixel point with a preset value in the first image. The opening operation unit is configured to perform an opening operation on the first image after the erosion operation to obtain a second image. The third determining unit is configured to subtract the pixel value of the second image from the pixel value of the first image after the erosion operation to obtain a third image. The first obtaining unit is configured to obtain at least two initial regions composed of contour points with the same brightness value in the third image if the pixel value of each pixel point in the third image is not the preset value. The screening unit is configured to screen the at least two initial regions to obtain at least two target regions. The fitting module includes: The second obtaining unit is configured to obtain center pixel points of the at least two target regions. The fitting unit is configured to perform curve fitting of a preset order on the center pixel points of the at least two target regions to obtain lane lines.

5. An electronic device, comprising: The electronic device includes: at least one processor; and a memory connected with the at least one processor in communication; wherein The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the lane line determination method in any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for enabling the processor to execute the lane line determination method in any one of claims 1-3 when executed.

Images