Laser line center point extraction method, device, computer equipment and storage medium

Abstract

The present invention belongs to the field of artificial intelligence, and relates to a method, device, computer device and storage medium for extracting the center point of a laser line. The method includes: obtaining a laser line picture; analyzing the frequency domain property and edge property of the laser picture; extracting high-frequency components from the frequency domain image as high-frequency features, extracting edge points from the edge detection result as edge features, using the extracted high-frequency features and edge features as inputs to construct a segmentation model, applying the trained segmentation model to segment the original laser line picture, binarizing the segmentation result to obtain a mask picture containing only the laser line; obtaining laser line width information according to the mask picture; using KMEANS to analyze the region division of the laser line width information and performing threshold segmentation to obtain the optimal demarcation point of the laser line width. It enhances the fineness and robustness of extracting the center point of the laser line; improves the accuracy and adaptability.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and particularly to a method, device, computer device and storage medium for extracting the center point of a laser line. Background Art

[0002] In the three-dimensional reconstruction of structured light, the accurate extraction of the center of the laser line is one of the key steps. However, when detecting objects in a complex environment, due to factors such as underwater refraction, background interference, and object reflection, the noise is greatly increased. Existing laser line extraction methods cannot achieve ideal accuracy and robustness, which is significantly reflected in noises similar to lasers. Summary of the Invention

[0003] To solve the above technical problems, the present invention provides a method for extracting the center point of a laser line, which adopts the following technical solutions, including:

[0004] Obtain a laser line picture;

[0005] Analyze the frequency domain properties and edge properties of the laser picture;

[0006] Extract high-frequency components from the frequency domain image as high-frequency features, extract edge points from the edge detection result as edge features, use the extracted high-frequency features and edge features as inputs to construct a segmentation model, apply the trained segmentation model to segment the original laser line picture, and binarize the segmentation result to obtain a mask picture containing only the laser line;

[0007] Obtain laser line width information according to the mask picture;

[0008] Use KMEANS to analyze the region division of the laser line width information and perform threshold segmentation to obtain the optimal demarcation point of the laser line width;

[0009] According to the optimal demarcation point and the relationship between the filtering kernel used in the Steger algorithm for Gaussian filtering and the laser line width, in the laser line picture, use different convolution kernel sizes for different width regions, and then extract the laser line center according to the constraint of the mask picture.

[0010] Preferably, the step of obtaining the laser line picture specifically includes:

[0011] Collect a scene picture irradiated by a laser line;

[0012] Preprocess the collected scene image.

[0013] Preferably, the step of analyzing the frequency domain properties and edge properties of the laser picture specifically includes:

[0014] Use a two-dimensional Fourier transform to transform the laser image from the spatial domain to the frequency domain, obtain a frequency-domain image, perform frequency-domain analysis, and identify high-frequency and low-frequency components;

[0015] Apply an edge detection operator to the laser image for edge detection to identify edge points in the laser image.

[0016] Preferably, the steps of extracting high-frequency components from the frequency-domain image as high-frequency features, extracting edge points from the edge detection result as edge features, using the extracted high-frequency features and edge features as inputs to construct a segmentation model, applying the trained segmentation model to segment the original laser line image, and binarizing the segmentation result to obtain a mask image containing only the laser line specifically include:

[0017] Extract high-frequency components from the frequency-domain image as high-frequency features, and extract edge points from the edge detection result as edge features;

[0018] Construct a segmentation model according to the extracted high-frequency features and edge features;

[0019] Apply the trained segmentation model to segment the original laser line image, binarize the segmentation result, and obtain a mask image containing only the laser line.

[0020] Preferably, the steps of obtaining laser line width information according to the mask image specifically include:

[0021] Scan the mask image line by line, and record the starting point and ending point of the laser line area in each line;

[0022] Calculate the width of the laser line on each line according to the positions of the starting point and the ending point;

[0023] Store the calculated width information in the form of an array or a matrix.

[0024] Preferably, the steps of using KMEANS to analyze and partition the laser line width information into regions and perform threshold segmentation to obtain the optimal demarcation point of the laser line width specifically include:

[0025] Use the laser line width information as input data and apply the KMEANS algorithm for clustering;

[0026] Analyze the clustering result, and determine the demarcation point between different clusters as the threshold;

[0027] Determine the optimal demarcation point by evaluating the stability and accuracy of the clustering result.

[0028] Preferably, the step of extracting the center of the laser line in the laser line picture according to the optimal demarcation point and according to the relationship between the filter kernel and the laser line width when performing Gaussian filtering using the Steger algorithm specifically includes:

[0029] Dividing the laser line into different width regions according to the optimal demarcation point;

[0030] Applying a convolution kernel of a corresponding size to the laser line picture of each width region for Gaussian filtering to remove noise;

[0031] Constructing a Hessian matrix for the filtered image and calculating its eigenvalues and eigenvectors;

[0032] Performing Taylor polynomial expansion on the pixel gray level in the normal direction to obtain a gray level distribution function;

[0033] According to the mask picture, only the center points within the laser line region are retained to extract the center of the laser line.

[0034] To solve the above technical problems, the present invention also provides a device for extracting the center point of a laser line, which adopts the following technical solutions, including:

[0035] A picture acquisition module for acquiring a laser line picture;

[0036] An analysis module for analyzing the frequency domain properties and edge properties of the laser picture;

[0037] A processing module for extracting high-frequency components from the frequency domain image as high-frequency features, extracting edge points from the edge detection result as edge features, using the extracted high-frequency features and edge features as inputs to construct a segmentation model, applying the trained segmentation model to segment the original laser line picture, and binarizing the segmentation result to obtain a mask picture containing only the laser line;

[0038] A width information acquisition module for obtaining laser line width information according to the mask picture;

[0039] A threshold segmentation module for using KMEANS to analyze the region division of the laser line width information and performing threshold segmentation to obtain the optimal demarcation point of the laser line width;

[0040] An extraction module for, according to the optimal demarcation point and according to the relationship between the filter kernel and the laser line width when performing Gaussian filtering using the Steger algorithm, using different convolution kernel sizes for different width regions in the laser line picture, and then extracting the center of the laser line according to the constraint of the mask picture.

[0041] To solve the above technical problems, the present invention also provides a computer device, which adopts the following technical solutions, including a memory and a processor. Computer-readable instructions are stored in the memory, and when the processor executes the computer-readable instructions, the steps of the above laser line center point extraction method are implemented.

[0042] To solve the above technical problems, the present invention also provides a computer-readable storage medium, which adopts the following technical solutions. Computer-readable instructions are stored on the computer-readable storage medium, and when the computer-readable instructions are executed by a processor, the steps of the above laser line center point extraction method are implemented.

[0043] Compared with the prior art, the present invention mainly has the following beneficial effects: By obtaining the laser line image and deeply analyzing its frequency domain properties and edge properties, a solid foundation is laid for subsequent processing; By constructing a segmentation model for laser frequency domain and edge information, the laser line image can be accurately processed to generate a high-quality mask image, effectively improving the accuracy and efficiency of subsequent processing; Using the mask image to obtain the laser line width information and adopting the KMEANS algorithm for region division and threshold segmentation to determine the optimal demarcation point of the laser line width, significantly enhancing the fineness and robustness of the processing result; According to the relationship between the optimal demarcation point and the filtering kernel and the laser line width in the Steger algorithm, this method uses different convolution kernel sizes for Gaussian filtering in different width regions and combines the mask image constraint to extract the laser line center. This flexible and fine processing method greatly improves the accuracy and adaptability of laser line center point extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] To more clearly illustrate the solutions in the present invention, the following will briefly introduce the drawings required for the description of the embodiments of the present invention. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0045] Figure 1 It is an exemplary system architecture diagram to which the present invention can be applied;

[0046] Figure 2 It is a schematic structural diagram of an embodiment of the laser line center point extraction device of the present invention;

[0047] Figure 3 It is a schematic structural diagram of an embodiment of the computer device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs; the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the terms "including" and "having" and any variations thereof in the specification and claims of this invention and the above drawings are intended to cover non-exclusive inclusion. The terms "first", "second", etc. in the specification and claims of this invention or the above drawings are used to distinguish different objects and not to describe a specific order.

[0049] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The phrase occurring in various places in the specification is not necessarily referring to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive of other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0050] To enable those skilled in the art to better understand the solution of this invention, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings.

[0051] It should be noted that the laser line center point extraction method provided by the embodiments of this invention is generally executed by a server / terminal device, and correspondingly, the laser line center point extraction device is generally disposed in the server / terminal device.

[0052] It should be understood that the numbers of terminal devices, networks, and servers are merely illustrative. According to the requirements of implementation, there can be any number of terminal devices, networks, and servers.

[0053] Example 1: Please refer to Figure 1 , which shows a flowchart of an embodiment of the laser line center point extraction method of this invention. The laser line center point extraction method includes the following steps:

[0054] Step S1, obtain a laser line picture.

[0055] In this embodiment, the electronic device (such as a server / terminal device) on which the laser line center point extraction method runs can receive a laser line center point extraction request through a wired connection method or a wireless connection method. It should be noted that the above wireless connection methods can include but are not limited to 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other currently known or future-developed wireless connection methods.

[0056] In this embodiment, step S1 of obtaining a laser line image may specifically include the following steps:

[0057] S11. Collect a scene image irradiated by a laser line.

[0058] Ensure that the photographic equipment (such as a camera) has high resolution and low noise performance to clearly capture the details of the laser line and the scene background. Secondly, according to the color and brightness of the laser line, reasonably adjust the white balance and exposure settings of the camera to avoid information loss caused by overexposure or underexposure. Fix the camera with a tripod to reduce blurring caused by vibration and ensure image stability. When collecting, consider the light conditions, such as the angle and intensity of natural light and the laser line, and use a lens hood to reduce stray light interference if necessary. Finally, use a remote shutter or the timer shooting function to further reduce the degradation of image quality caused by human factors.

[0059] S12. Preprocess the collected scene image.

[0060] The main objectives of preprocessing are to eliminate image noise, enhance target features, and simplify data.

[0061] Perform grayscale processing to convert a color image into a grayscale image to reduce the computational amount and improve the processing speed. This step can be achieved by the component method, maximum value method, average value method, or weighted average method, and the weighted average method is often used because it takes into account the sensitivity of the human eye to colors.

[0062] Apply geometric transformation to correct systematic errors and random errors in the image and ensure the spatial accuracy of the image.

[0063] Adopt image enhancement techniques, such as smoothing and sharpening algorithms in the spatial domain method, or low-pass and high-pass filters in the frequency domain method, to highlight the features of the laser line and suppress background noise.

[0064] Step S2. Analyze the frequency domain properties and edge properties of the laser image.

[0065] In this embodiment, step S2 of analyzing the frequency domain properties and edge properties of the laser image may specifically include the following steps:

[0066] S21. Use two-dimensional Fourier transform to convert the laser image from the spatial domain to the frequency domain to obtain a frequency domain image, perform frequency domain analysis, and identify high-frequency and low-frequency components.

[0067] High-frequency components usually correspond to the edges and details of the image, while low-frequency components correspond to the smooth regions of the image.

[0068] It is transformed into the frequency domain through two-dimensional Fourier transform, which can be realized by using the fast Fourier transform (FFT) algorithm. The gray distribution function of the image is converted into a frequency distribution function. The obtained frequency-domain image after conversion, that is, the spectrogram, reflects the distribution of various frequency components in the image.

[0069] On the spectrogram, high-frequency components usually appear as bright spots, corresponding to regions with drastic changes in the image, such as edges and details; while low-frequency components appear as darker or uniform regions, corresponding to smooth or slowly changing parts of the image. By visually inspecting the spectrogram, the specific distribution of high-frequency and low-frequency components can be identified.

[0070] Furthermore, a high-pass or low-pass filter can be used to process the frequency-domain image to highlight or suppress specific frequency components, thereby achieving purposes such as image enhancement, denoising, or feature extraction.

[0071] In specific implementation, three separately trained network channels (left, middle, and right) can also be used for network combination, respectively obtaining and fusing information from the edges, frequency domain, and original laser images of the laser pictures. The lateral connection between every two layers in the convolutional architecture uses a double-convolution structure. In the vertical connection between layers, stacking at the channels and using convolution for fusion while ensuring that the width and height information remain unchanged.

[0072] The processing method of the left channel in the network is to process the laser edge information. Different from Gated-SCNN, here the edge information is directly input into the edge auxiliary channel of the network as the input at the edge layer. It is equivalent to artificially designing a network layer with a fixed function before the input layer to identify the edge information and inserting it into the main channel for information exchange. It plays an auxiliary role in enhancing the recognition of laser features in the overall network. Due to the uniqueness of the laser edge information and the high light intensity of the laser line, in most scenarios, after edge extraction of the laser line picture, the laser information can be significantly displayed. There are great differences between the edge information of the laser stripe center, the edge information of the reflection noise affected by the strong laser around the laser, and the strong light noise in the general environment. The edge information of the laser can be effectively utilized to strengthen the network's integration and judgment of laser features.

[0073] The right channel of the network is a frequency-domain processing channel, and the input layer of the network is designed as a high-pass filter with multiple filtering intervals. Because when analyzing the laser line in the frequency domain, it is found that the laser line mainly shows high-frequency properties in the frequency domain, while the ambient light or the noise under reflection mainly shows low-frequency properties.

[0074] Therefore, in the input layer, the input picture is first subjected to Fourier transform to convert the original picture information from the spatial domain to the frequency domain. For a size of After Fourier transformation of the grayscale image, four different filter intervals are randomly selected for high-pass filtering. Each filter interval consists of five random filter centers and filter radii. Compared with the traditional single filter interval, this method can use the network to extract the characteristics of the laser in each filter interval, which is more generalized and accurate, and solves the problem of narrow applicability of a single filter radius and inaccurate target setting. During training, the size of the filter radius in each round is random, and during evaluation, the filter radius size is fixed to 20 for evaluation to determine the stability during use.

[0075] The processed image is now a spectrum diagram at the noise-removed frequency. By performing an inverse Fourier transform on the spectrum diagram at this time, an image after a certain amount of noise is removed in the spatial domain can be obtained.

[0076] In the frequency domain input layer of the network, we use The matrix expands the dimensions of the four images, superimposes them on the channels, inputs them to the next layer, and performs the same convolution operation with the network channel of the edge information to ensure that the dimensions are the same when they are integrated into the main channel.

[0077] The middle channel is the main channel of the network. The output dimension of each layer is twice the dimension of the left and right layers. The edge layer network channel and the frequency domain layer network channel are integrated into the main channel through the middle confluence layer. After separate and unified convolution, the result is output.

[0078] Taking into account the small amount of laser data and the difficulty of data collection diversity, the main effects of transfer learning and data enhancement were actually used during training. The laser images were randomly rotated, cut, and reversed to increase the diversity of data. At the same time, random spots were artificially generated into the laser image to increase the diversity of noise. The dice function and the weighted cross entropy function were set up as loss functions. Considering that the proportion of lasers is extremely small, in order to prevent the foreground segmentation effect from being poor due to the large background, the annealing algorithm and the weighted cross entropy loss were combined to initially give the foreground a large weight for training. While using transfer learning, the data set was gradually updated while the weight of the foreground was reduced. Finally, the model parameters at the end of training were obtained and laser line segmentation was performed.

[0079] In the specific implementation, most laser line images have high precision and require timely extraction. Therefore, the image is compressed to half of the original image at the input of the network, which effectively improves the running speed of the model.

[0080] S22, applying an edge detection operator to perform edge detection on the laser image, and identifying edge points in the laser image.

[0081] Edge detection operators include first-order differential operators such as Sobel, Prewitt, Robert, etc., and second-order differential operators such as Laplacian and Kirsch, etc. These operators detect edges by calculating the first derivative or second derivative of the grayscale values of the image.

[0082] Regarding the characteristics of laser images, the Canny operator can be preferentially considered because it can seek the best balance between anti-noise interference and precise positioning. The Canny operator removes noise through Gaussian filtering, calculates the gradient using the Sobel operator, and then performs non-maximum suppression and double-threshold screening to obtain an accurate edge image.

[0083] During the implementation process, the laser image needs to be first converted into a grayscale image, and then the Canny operator is applied for edge detection. By adjusting the parameters of the operator, such as the standard deviation of the Gaussian filter, the step size of gradient calculation, and the size of the double thresholds, the effect of edge detection can be optimized, thereby accurately identifying the edge points in the laser image.

[0084] Step S3, extract high-frequency components from the frequency-domain image as high-frequency features, extract edge points from the edge detection result as edge features, use the extracted high-frequency features and edge features as inputs to construct a segmentation model, apply the trained segmentation model to segment the original laser line image, and binarize the segmentation result to obtain a mask image containing only the laser line.

[0085] In this embodiment, step S3, extract high-frequency components from the frequency-domain image as high-frequency features, extract edge points from the edge detection result as edge features, use the extracted high-frequency features and edge features as inputs to construct a segmentation model, apply the trained segmentation model to segment the original laser line image, and binarize the segmentation result to obtain a mask image containing only the laser line, which can specifically include the steps:

[0086] S31. Extract high-frequency components from the frequency-domain image as high-frequency features, and extract edge points from the edge detection result as edge features.

[0087] In the frequency-domain image, high-frequency components correspond to the details and edges in the image, and they are located in the region outside the center of the spectrum. By designing a filter to selectively retain these high-frequency components and performing inverse Fourier transform on them, the high-frequency feature image can be reconstructed in the spatial domain.

[0088] The extraction of edge features relies on edge detection algorithms, such as the Canny operator or the Sobel operator. These algorithms are based on the calculation of image gradients and can identify positions in the image where the brightness changes sharply, i.e., edge points. By setting appropriate thresholds, significant edge points can be filtered out and used as edge features. These edge features are of great significance for tasks such as image understanding and object recognition because they can concisely describe the main structure and shape information in the image.

[0089] S32. Construct a segmentation model based on the extracted high-frequency features and edge features.

[0090] Use the extracted high-frequency features and edge features as inputs to construct a segmentation model. The segmentation model can adopt machine learning-based classifiers, such as support vector machines (SVMs) or random forests. These classifiers can effectively distinguish between laser line regions and non-laser line regions by learning the distribution laws of different classes in the feature space.

[0091] Deep learning models, especially convolutional neural networks (CNNs), are also powerful tools for constructing segmentation models. Through multiple layers of convolution and pooling operations, CNNs can automatically extract hierarchical features in images and obtain highly accurate classification capabilities through training on a large amount of data. In the task of extracting the center point of the laser line, a specific CNN architecture can be designed, taking high-frequency features and edge features as inputs, and finally outputting a binary image of the laser line region through layer-by-layer convolution and pooling.

[0092] Train the segmentation model to distinguish between laser line regions and non-laser line regions.

[0093] By combining high-frequency features and edge features and selecting machine learning or deep learning models for training, an efficient and accurate segmentation model for extracting the center point of the laser line can be constructed.

[0094] S33. Apply the trained segmentation model to segment the original laser line image, binarize the segmentation result, and obtain a mask image containing only the laser line.

[0095] During the segmentation process, the original laser line image is first input into the segmentation model. Specifically, in implementation, the preprocessed original laser line image can also be compressed to half of the original image and divided into two parts. One part is input as the original image, and the other part is input as edge information into the segmentation model.

[0096] The segmentation model will classify each pixel in the image to determine whether it belongs to a part of the laser line. This process depends on the feature representation and classification capabilities learned by the model.

[0097] To improve the accuracy and efficiency of segmentation, the segmentation model can also adopt methods such as dilated convolution and multi-scale feature fusion to capture richer image features, thereby better identifying laser lines.

[0098] After completing pixel-level classification, the model generates a segmentation mask, where the part of the laser line is marked with a specific value, while other parts are marked as background or other categories. Through this segmentation mask, the precise position and shape of the laser line can be easily extracted.

[0099] Perform binary processing on the segmentation result. Binary processing is the process of converting an image into an image that only contains two pixel values (usually 0 and 255), that is, converting the image into a black-and-white image. In this process, the global threshold method or the adaptive threshold method can be selected to determine the threshold for pixel value conversion. The global threshold method is suitable for cases where the background is relatively uniform, while the adaptive threshold method can better handle images with complex backgrounds or uneven lighting.

[0100] Through binary processing, a mask image containing only the laser line is obtained. This mask image can be used as a key input for extracting the center point of the laser line in subsequent processing, thereby further improving the accuracy and efficiency of extraction.

[0101] Specifically in implementation, the Lanczos interpolation algorithm can also be used to restore the output mask image.

[0102] To meet the timeliness requirements of code operation, the network model was compressed while ensuring the accuracy requirements. However, there are still problems with the running speed when facing high-precision input images. Therefore, before inputting the image into the segmentation model, the image was compressed once, reducing the width and height of the image to half of the original image. , 。Then, in order to make the dimension of the mask image the same as that of the original image, the output image needs to be enlarged. Here, Lanczos interpolation is used to complete the restoration of the image. First, establish the kernel function of Lanczos and use 's area as the kernel size.

[0103] (1)

[0104] For each point in the new image, with this point as the center of the area, construct a weight matrix according to the kernel function.

[0105] (2)

[0106] Among them, and The value-taking method is the relative offset position from the center of the area, For the ideal point of the original image corresponding to the new image, when the image is magnified by two times, the mapping relationship between the actual point on the magnified image and the point on the original image is:

[0107] (3)

[0108] is the original point on the magnified image. It can be seen from formula (3) that may not be an integer. Therefore, for non-integer the pixel values of the points are calculated by the method of weighted average.

[0109] The picture effect after interpolation processing can be obtained accordingly. Comparing the original picture with the image after interpolation restoration, the visual effects of the images are the same, and the information of the laser line can be represented without distortion.

[0110] Step S4: Obtain the laser line width information according to the mask picture.

[0111] In this embodiment, step S4: Obtaining the laser line width information according to the mask picture may specifically include the steps:

[0112] S41: Scan the mask picture line by line, and record the starting point and ending point of the laser line area in each line.

[0113] Most of the redundant information has been removed from the mask picture, highlighting the laser stripes. Start the line-by-line scanning program, and start scanning line by line from the first line of the mask picture.

[0114] During the scanning process, by comparing the pixel gray values or color differences, use the image analysis algorithm to detect the starting point and ending point of the laser line area in each line. When scanning to a certain pixel point in a certain line, whose gray value or color feature conforms to the feature of the laser stripe, it is recorded as the starting point. Continue to scan this line, and when encountering a pixel point that does not conform to the feature of the laser stripe again, it is recorded as the ending point.

[0115] To ensure the accuracy of extraction, a sub-pixel level image processing method, such as Gaussian fitting, etc., can be used to further determine the specific positions of the starting point and ending point. At the same time, to avoid the influence of noise and interference, the mask picture can be filtered and smoothed before scanning.

[0116] By scanning line by line and recording the starting point and ending point of the laser line area in each line, a reliable data basis can be provided for the subsequent center point extraction and laser stripe analysis.

[0117] S42: Calculate the width of the laser line on each line according to the positions of the starting point and ending point.

[0118] By setting a threshold, the image is binarized or threshold-segmented to distinguish the laser line from the background. This step helps to clearly define the starting and ending points of the laser line.

[0119] For each row of pixels, traverse and identify continuous pixel sequences whose gray values exceed a preset threshold. The starting and ending positions of these sequences correspond to the left and right boundaries of the laser line in that row.

[0120] Calculate the pixel distance between the left and right boundaries, which is the width of the laser line in that row. To improve the accuracy of the calculation, interpolation or sub-pixel level analysis methods can be used to finely adjust the boundary positions.

[0121] In addition, considering that the laser line may be distorted or discontinuous due to various factors (such as object surface characteristics, lighting conditions, etc.), filtering and smoothing can also be performed when calculating the width to reduce noise interference and ensure the stability and reliability of the results.

[0122] S43. Store the calculated width information in the form of an array or matrix.

[0123] To store this width information, a two-dimensional array or matrix can be created, whose dimensions match the size of the original image. Then, traverse the processed image data and assign the laser line width value corresponding to each pixel or pixel block to the element at the corresponding position in the two-dimensional array or matrix.

[0124] In specific implementation, lists (List) in programming languages such as Python or arrays (Array) in the NumPy library can be used to conveniently create and manipulate this two-dimensional data structure. For example, in Python, an empty list can be created first, and then nested loops are used to traverse the image data and add the width values to the corresponding positions in the list one by one. Or directly create an array of a specified size using the NumPy library and assign the width values to the array elements through index operations.

[0125] Step S5. Use KMEANS to perform regional analysis on the laser line width information and perform threshold segmentation to obtain the optimal demarcation point of the laser line width.

[0126] In this embodiment, step S5. Using KMEANS to perform regional analysis on the laser line width information and perform threshold segmentation to obtain the optimal demarcation point of the laser line width can specifically include the steps:

[0127] S51. Use the laser line width information as input data and apply the KMEANS algorithm for clustering.

[0128] Taking the laser line width information as input, apply the KMEANS algorithm for clustering. The KMEANS algorithm is an unsupervised learning clustering algorithm that can divide the data in the dataset into different classes or clusters. In this step, the algorithm randomly selects k initial cluster centers, then calculates the distance from each pixel point to these cluster centers, and assigns it to the closest cluster. Subsequently, the algorithm recalculates the cluster center of each cluster, that is, takes the average value of all data points in the cluster as the new cluster center. This process is repeated continuously until the cluster centers no longer change or reach the preset number of iterations.

[0129] Through KMEANS clustering, the pixel points on the laser line can be divided into different regions, and each region corresponds to a cluster center. These cluster centers can be regarded as representative points of different width regions on the laser line. Finally, by further processing these cluster centers, such as calculating their average value or median, etc., the position of the center point of the laser line can be determined. This method can fully consider the width information of the laser line and improve the accuracy and robustness of the center point extraction. Select an appropriate number of clusters (such as 2 or 3) to obtain the clustering result.

[0130] S52. Analyze the clustering result and determine the boundary points between different clusters as thresholds.

[0131] Evaluate the clustering quality through indicators such as the silhouette coefficient and Calinski-Harabasz Index to ensure good clustering results. On this basis, analyze the characteristic differences between different clusters and determine the boundary points between them. These boundary points usually correspond to the transition points between the laser line and the non-laser line regions in the image, so they can be used as thresholds for extracting the center point of the laser line.

[0132] According to the determined threshold, perform binary processing or threshold segmentation on the image to accurately extract the center point of the laser line. This method combines the accuracy of clustering analysis and the uniqueness of laser line features, and can significantly improve the accuracy and efficiency of laser line center point extraction.

[0133] Divide the laser line into different width regions according to the threshold.

[0134] S53. Determine the optimal boundary point by evaluating the stability and accuracy of the clustering result.

[0135] The optimal boundary point refers to the boundary value or boundary condition that can divide the data points most accurately and stably into different classes during the clustering process.

[0136] During the clustering process, it is necessary to continuously evaluate the stability and accuracy of the clustering results. Stability can be measured by running the clustering algorithm multiple times and comparing the consistency of the results; accuracy can be evaluated by comparing with the actual labels (if any) or using internal evaluation metrics such as the silhouette coefficient, Calinski-Harabasz index, etc.

[0137] According to the evaluation results of stability and accuracy, determine the optimal cut-off point. This cut-off point may be a specific value or a certain boundary condition, which can most effectively divide the data points into different categories, thereby accurately extracting the center point of the laser line. This process needs to comprehensively consider various factors, including the characteristics of the data, the choice of clustering algorithm, and the applicability of the evaluation metrics, etc.

[0138] Step S6: According to the optimal cut-off point and the relationship between the filter kernel and the laser line width when using the Steger algorithm for Gaussian filtering, in the laser line image, use different convolution kernel sizes for different width regions, and then extract the laser line center according to the constraints of the mask image.

[0139] In this embodiment, step S6: According to the optimal cut-off point and the relationship between the filter kernel and the laser line width when using the Steger algorithm for Gaussian filtering, in the laser line image, use different convolution kernel sizes for different width regions, and then extract the laser line center specifically may include the steps:

[0140] S61: Divide the laser line into different width regions according to the optimal cut-off point.

[0141] For each width region, select an appropriate convolution kernel size according to the relationship between the laser line width and the Gaussian filter kernel size. For a narrower laser line, choosing a smaller filter kernel can avoid edge blurring caused by over-smoothing; for a wider laser line, a larger filter kernel can be selected to better suppress noise and retain the overall shape of the light strip. The width of the laser line can be set according to the actual situation. When selecting the filter kernel, the image noise situation also needs to be considered, and the size and standard deviation of the filter kernel are appropriately adjusted to achieve the best filtering effect.

[0142] To achieve more accurate extraction, an adaptive filtering method can also be adopted, that is, dynamically adjust the size and parameters of the filter kernel according to the local features of each pixel point. This method can further improve the extraction accuracy of the laser line center point.

[0143] S62: Apply a Gaussian filter with a corresponding size convolution kernel to the laser line image of each width region to remove noise.

[0144] Gaussian filtering is a linear smoothing filter suitable for removing Gaussian noise. Its principle is to smooth the image by calculating the weighted average of surrounding pixels. According to the width and area characteristics of the laser line image, select the appropriate convolution kernel size. The size of the convolution kernel is usually proportional to the width of the laser line to ensure that the filtering effect can effectively remove noise and keep the detailed characteristics of the laser line as much as possible.

[0145] The laser line image is convolved using a convolution kernel of a selected size. The convolution operation involves sliding the convolution kernel over the image and calculating the sum of the products of the pixel values ​​in the area covered by the convolution kernel and the convolution kernel weight to obtain a new pixel value. In Gaussian filtering, the weight distribution of the convolution kernel follows a Gaussian function, that is, the center weight is the largest and the edge weight gradually decreases, which helps to preserve edge information while smoothing the image.

[0146] Through this step, the noise in the laser line image is effectively suppressed, providing a clearer and more accurate image basis for subsequent center point extraction.

[0147] S63. Construct a Hessian matrix for the filtered image and calculate its eigenvalues ​​and eigenvectors.

[0148] Construct the Hessian matrix. In a two-dimensional image, each pixel has a 2x2 Hessian matrix whose elements consist of the second-order partial derivatives of the image at that point in the x and y directions. These partial derivatives reflect the rate of change of the image grayscale value near that point.

[0149] After obtaining the Hessian matrix, its eigenvalues ​​and eigenvectors are calculated. The eigenvalue represents the rate of change of the image grayscale value in the direction of the eigenvector. In the context of light strip center extraction, the eigenvector corresponding to the maximum eigenvalue usually represents the normal direction of the light strip. This is because the grayscale value at the center of the light strip is the largest, and when leaving the center along the normal direction, the grayscale value gradually decreases, forming a shape similar to a Gaussian distribution.

[0150] By calculating the eigenvalues ​​and eigenvectors of the Hessian matrix, the normal direction of the light strip can be determined, and this provides a basis for subsequently applying Taylor polynomial expansion in the normal direction to locate the sub-pixel position of the center of the light strip.

[0151] S64. Apply Taylor polynomial expansion to the pixel grayscale in the normal direction to obtain a grayscale distribution function.

[0152] The eigenvalues and eigenvectors of each pixel in the image are calculated through the Hessian matrix, and the eigenvector corresponding to the largest eigenvalue is regarded as the normal direction of that point. Along the determined normal direction, the pixel gray values are expanded by Taylor polynomials. This expansion process is actually to approximate the gray distribution function as a polynomial function in this direction for subsequent calculation and analysis.

[0153] In specific implementation, the Taylor polynomial can be expanded to the second order or higher order to improve the approximation accuracy. After expansion, by solving the extreme points of the polynomial function, the center point coordinates of the light stripe in this normal direction can be determined. The realization of this step depends on the accurate calculation of the Taylor polynomial expansion and the effective solution of the extreme points.

[0154] S65: According to the mask image, only the center points within the laser line region are retained to extract the laser line center.

[0155] The original image is cropped or masked using the mask image, and only the pixel information within the laser line region is retained. This can be achieved by comparing each pixel of the mask image and the original image pixel by pixel. Only when the corresponding pixel value in the mask image is true (i.e., indicating that the pixel belongs to the laser line region), the information of that pixel in the original image is retained.

[0156] Within the cropped image region, a center extraction algorithm is applied. This can be based on gray gradient, extreme value method, gray centroid method, or more advanced algorithms such as the Steger algorithm, which determines the normal direction of the light stripe by constructing the Hessian matrix and finds the extreme points in the normal direction to obtain the sub-pixel position of the light stripe center.

[0157] Through verification and screening steps, it is ensured that the extracted center points are within the laser line region and meet certain accuracy and continuity requirements.

[0158] Through the above steps, the goal of retaining only the center points within the laser line region according to the mask image can be achieved.

[0159] Implementing this embodiment has the following beneficial effects: By obtaining the laser line image and deeply analyzing its frequency domain properties and edge properties, a solid foundation is laid for subsequent processing; By constructing a segmentation model for laser frequency domain and edge information, the laser line image can be accurately processed to generate a high-quality mask image, effectively improving the accuracy and efficiency of subsequent processing; Using the mask image to obtain the laser line width information and adopting the KMEANS algorithm for region division and threshold segmentation to determine the optimal demarcation point of the laser line width, significantly enhancing the fineness and robustness of the processing result; According to the relationship between the optimal demarcation point and the filtering kernel and the laser line width in the Steger algorithm, this method uses different convolution kernel sizes for Gaussian filtering in different width regions and combines the mask image constraint to extract the laser line center. This flexible and fine processing method greatly improves the accuracy and adaptability of laser line center point extraction.

[0160] The present invention can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. The present invention can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0161] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, read-only memory (ROM), or a random access memory (RAM), etc.

[0162] It should be understood that although the steps in the flowchart of the accompanying drawings are shown sequentially in the direction of the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise clearly stated in this article, there is no strict order restriction for the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0163] Example 2: Further reference Figure 2 is made to Figure 1 , as an implementation of the method shown above Figure 1 , the present invention provides an embodiment of a laser line center point extraction device. This device embodiment corresponds to

[0164] the method embodiment shown Figure 2 and can be specifically applied to various electronic devices.

[0165] As shown in

[0166] FIG. [FIGURE NUMBER NOT PROVIDED], the laser line center point extraction device 70 in this embodiment includes: an image acquisition module 71, an analysis module 72, a processing module 73, a width information acquisition module 74, a threshold segmentation module 75, and an extraction module 76. Among them:

[0165] The image acquisition module 71 is configured to acquire a laser line image;

[0166] The analysis module 72 is configured to analyze the frequency domain properties and edge properties of the laser image;

[0167] The processing module 73 is configured to extract high-frequency components from the frequency domain image as high-frequency features, extract edge points from the edge detection result as edge features, use the extracted high-frequency features and edge features as inputs to construct a segmentation model, apply the trained segmentation model to segment the original laser line image, and binarize the segmentation result to obtain a mask image containing only the laser line;

[0168] The width information acquisition module 74 is configured to obtain the laser line width information according to the mask image;

[0169] The threshold segmentation module 75 is configured to use KMEANS to analyze and divide regions of the laser line width information, and perform threshold segmentation to obtain the optimal demarcation point of the laser line width;

[0170] An extraction module 76 is configured to use different convolution kernel sizes for different width regions in a laser line image according to the optimal demarcation point and the relationship between the convolution kernel and the laser line width when performing Gaussian filtering using the Steger algorithm, and then extract the laser line center according to the constraints of the mask image.

[0171] Implementing this embodiment has the beneficial effects that: by obtaining the laser line image and deeply analyzing its frequency domain properties and edge properties, a solid foundation is laid for subsequent processing; by constructing a segmentation model for laser frequency domain and edge information, the laser line image can be precisely processed to generate a high-quality mask image, effectively improving the accuracy and efficiency of subsequent processing; by using the mask image to obtain the laser line width information and adopting the KMEANS algorithm for region division and threshold segmentation, the optimal demarcation point of the laser line width is determined, significantly enhancing the fineness and robustness of the processing result; according to the optimal demarcation point and the relationship between the convolution kernel and the laser line width in the Steger algorithm, this method uses different convolution kernel sizes for different width regions for Gaussian filtering and combines the mask image constraints to extract the laser line center. This flexible and fine processing method greatly improves the accuracy and adaptability of laser line center point extraction.

[0172] Example 3: To solve the above technical problems, an embodiment of the present invention also provides a computer device. Specifically, please refer to Figure 3 , Figure 3 which is the basic structural block diagram of the computer device in this embodiment.

[0173] The above computer device 8 includes a memory 81, a processor 82, and a network interface 83 that are communicatively connected to each other through a system bus. It should be noted that only the computer device 8 with components such as the memory 81, the processor 82, and the network interface 83 is shown in the figure, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented. Among them, those skilled in the art of the present technology can understand that a computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0174] The above computer device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The above computer device can perform human-computer interaction with the user through a keyboard, a mouse, a remote control, a touchpad, a voice control device, etc.

[0175] The above memory 81 includes at least one type of readable storage medium, and the above readable storage medium includes flash memory, a hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the above memory 81 may be an internal storage unit of the above computer device 8, such as the hard disk or memory of the computer device 8. In other embodiments, the above memory 81 may also be an external storage device of the above computer device 8, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a FlashCard, etc. equipped on the computer device 8. Of course, the above memory 81 may also include both the internal storage unit and the external storage device of the above computer device 8. In this embodiment, the above memory 81 is generally used to store the operating system installed on the above computer device 8 and various application software, such as computer-readable instructions of the laser line center point extraction method. In addition, the above memory 81 may also be used to temporarily store various data that have been output or will be output.

[0176] In some embodiments, the above processor 82 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 82 is generally used to control the overall operation of the above computer device 8. In this embodiment, the above processor 82 is used to run the computer-readable instructions stored in the above memory 81 or process data, such as running the computer-readable instructions of the above laser line center point extraction method.

[0177] The above network interface 83 may include a wireless network interface or a wired network interface, and the network interface 83 is generally used to establish a communication connection between the above computer device 8 and other electronic devices.

[0178] Implementing this embodiment has the following beneficial effects: By acquiring laser line images and deeply analyzing their frequency domain properties and edge properties, a solid foundation is laid for subsequent processing; By constructing a segmentation model for laser frequency domain and edge information, the laser line images can be precisely processed to generate high-quality mask images, effectively improving the accuracy and efficiency of subsequent processing; By using the mask images to obtain the laser line width information and adopting the KMEANS algorithm for region division and threshold segmentation, the optimal demarcation point of the laser line width can be determined, significantly enhancing the fineness and robustness of the processing results; According to the relationship between the optimal demarcation point and the filtering kernel and the laser line width in the Steger algorithm, this method performs Gaussian filtering with different convolution kernel sizes for different width regions and combines the mask image constraints to extract the laser line center. This flexible and precise processing method greatly improves the accuracy and adaptability of laser line center point extraction.

[0179] Example 4: The present invention also provides another implementation manner, that is, to provide a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor, so that the at least one processor executes the steps of the laser line center point extraction method as described above.

[0180] Implementing this embodiment has the following beneficial effects: By acquiring laser line images and deeply analyzing their frequency domain properties and edge properties, a solid foundation is laid for subsequent processing; By constructing a segmentation model for laser frequency domain and edge information, the laser line images can be precisely processed to generate high-quality mask images, effectively improving the accuracy and efficiency of subsequent processing; By using the mask images to obtain the laser line width information and adopting the KMEANS algorithm for region division and threshold segmentation, the optimal demarcation point of the laser line width can be determined, significantly enhancing the fineness and robustness of the processing results; According to the relationship between the optimal demarcation point and the filtering kernel and the laser line width in the Steger algorithm, this method performs Gaussian filtering with different convolution kernel sizes for different width regions and combines the mask image constraints to extract the laser line center. This flexible and precise processing method greatly improves the accuracy and adaptability of laser line center point extraction.

[0181] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described example methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0182] Obviously, the above-described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. The accompanying drawings show preferred embodiments of the present invention, but do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosed content of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing specific embodiments, or perform equivalent replacements on some of the technical features. Any equivalent structure directly or indirectly using the content of the specification and drawings of the present invention in other related technical fields is similarly within the scope of the patent protection of the present invention.

Claims

1. A method for extracting the center point of a laser line, characterized in that: The steps include: Get the laser line image; Analyzing the frequency domain properties and edge properties of the laser line image; Extract high-frequency components from the frequency domain image as high-frequency features, extract edge points from the edge detection results as edge features, use the extracted high-frequency features and edge features as input, build a segmentation model, apply the trained segmentation model to segment the original laser line image, binarize the segmentation results, and obtain a mask image containing only the laser line; According to the mask image, obtaining laser line width information; KMEANS is used to analyze the laser line width information by dividing the area, and threshold segmentation is performed to obtain the optimal dividing point of the laser line width; According to the optimal dividing point and the relationship between the filter kernel and the laser line width when Gaussian filtering is performed using the Steger algorithm, different convolution kernel sizes are used for different width areas in the laser line image, and then the laser line center is extracted according to the constraints of the mask image.

2. The method for extracting the center point of a laser line according to claim 1, characterized in that: The step of obtaining the laser line image specifically includes: Collect pictures of the scene illuminated by the laser line; The collected scene pictures are preprocessed.

3. The method for extracting the center point of a laser line according to claim 1, characterized in that: The step of analyzing the frequency domain properties and edge properties of the laser line image specifically includes: The laser line image is converted from the spatial domain to the frequency domain using a two-dimensional Fourier transform to obtain a frequency domain image, and a frequency domain analysis is performed to identify high-frequency and low-frequency components; An edge detection operator is applied to perform edge detection on the laser line image to identify edge points in the laser line image.

4. The method for extracting the center point of a laser line according to claim 3, characterized in that: The steps of extracting high-frequency components from the frequency domain image as high-frequency features, extracting edge points from the edge detection results as edge features, using the extracted high-frequency features and edge features as inputs, constructing a segmentation model, applying the trained segmentation model to segment the original laser line image, binarizing the segmentation results, and obtaining a mask image containing only the laser line specifically include: Extracting high-frequency components from the frequency domain image as high-frequency features, and extracting edge points from edge detection results as edge features; Constructing a segmentation model according to the extracted high-frequency features and edge features; The trained segmentation model is used to segment the original laser line image, and the segmentation result is binarized to obtain a mask image containing only the laser line.

5. The method for extracting the center point of a laser line according to claim 4, characterized in that: The step of obtaining the laser line width information according to the mask image specifically includes: Scan the mask image line by line, and record the starting point and ending point of the laser line area in each line; According to the positions of the starting point and the ending point, the width of the laser line on each row is calculated; Store the calculated width information in array or matrix form.

6. The method for extracting the center point of a laser line according to claim 5, characterized in that: The step of using KMEANS to analyze the laser line width information by dividing the area and performing threshold segmentation to obtain the optimal dividing point of the laser line width specifically includes: The laser line width information is used as input data and the KMEANS algorithm is applied for clustering; Analyze the clustering results and determine the dividing point between different clusters as the threshold; The optimal cut-off point was determined by evaluating the stability and accuracy of the clustering results.

7. The method for extracting the center point of a laser line according to claim 6, characterized in that: According to the optimal demarcation point and the relationship between the filter kernel and the laser line width when the Steger algorithm is used for Gaussian filtering, in the laser line image, different convolution kernel sizes are used for different width areas, and then according to the constraints of the mask image, the step of extracting the laser line center specifically includes: Divide the laser line into areas of different widths according to the optimal dividing point; Apply a convolution kernel of corresponding size to the laser line image of each width area to perform Gaussian filtering to remove noise; Construct the Hessian matrix for the filtered image and calculate its eigenvalues ​​and eigenvectors; Apply Taylor polynomial expansion to the pixel grayscale in the normal direction to obtain the grayscale distribution function; According to the mask image, only the center point in the laser line area is retained to extract the center of the laser line.

8. A laser line center point extraction device, characterized in that: include: Image acquisition module, used to acquire laser line images; An analysis module, used for analyzing the frequency domain properties and edge properties of the laser line image; A processing module is used to extract high-frequency components from the frequency domain image as high-frequency features, extract edge points from the edge detection results as edge features, use the extracted high-frequency features and edge features as input, build a segmentation model, apply the trained segmentation model to segment the original laser line image, binarize the segmentation results, and obtain a mask image containing only the laser line; A width information acquisition module, used to obtain laser line width information according to the mask image; A threshold segmentation module is used to analyze the laser line width information by dividing the area using KMEANS, and perform threshold segmentation to obtain the optimal dividing point of the laser line width; The extraction module is used to use different convolution kernel sizes for different width areas in the laser line image according to the optimal demarcation point and the relationship between the filter kernel and the laser line width when Gaussian filtering is performed using the Steger algorithm, and then extract the laser line center according to the constraints of the mask image.

9. A computer device, comprising a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the laser line center point extraction method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the laser line center point extraction method according to any one of claims 1 to 7 are implemented.

Images