Vehicle wiper water spraying state detection method, device, equipment and medium

Abstract

The application discloses a vehicle wiper water spraying state detection method, device, equipment and medium. The method obtains a wiper water spraying image of a target vehicle to be detected; detects the distribution of water columns in the wiper water spraying image through a segmentation detection algorithm to obtain initial water column image data; filters target water column image data from the initial water column image data based on a straight line fitting algorithm; extracts start and end point information of each target water column in the target water column image data, and judges whether the start and end point information conforms to a start and end point distribution rule obtained through clustering analysis of a plurality of sample water columns; counts the number of water columns conforming to the start and end point distribution rule, and compares whether the number of water columns is consistent with a designed expected number; if not, it is determined that the wiper water spraying state of the vehicle is abnormal. The method can automatically detect whether the wiper water spraying state of the vehicle is abnormal, has high detection accuracy, and is beneficial to reducing the work burden of inspectors. The application can be widely applied in the field of artificial intelligence technology.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a vehicle wiper water spraying state detection method, device, equipment and medium. BACKGROUND

[0002] Automobile is an important means of transportation at present, which brings great convenience to people's life and work. On the automobile production line, it is generally necessary to detect whether various parts and components in the vehicle production process are normal. For example, for vehicle wipers, it is necessary to detect whether the water holes can normally spray water columns. If there is an abnormality, the defective products need to be repaired or the problem needs to be traced.

[0003] In related technologies, inspectors generally visually observe the water spraying function of the wiper for corresponding inspection. However, on the one hand, after some vehicle models are designed and installed with integrated wipers, the number of water holes of the wipers increases and the water holes become thinner, which leads to high observation difficulty. On the other hand, the detection means through manual inspection has many uncertainties (such as external shielding factors or fatigue of the inspectors, etc. non-controllable human factors), and there are many misjudgment cases, which leads to low detection accuracy and the risk of defective product outflow, affecting the normal use of customers. SUMMARY

[0004] The present application aims to at least partly solve one of the technical problems existing in the related art.

[0005] To this end, an object of the present application is to provide a vehicle wiper water spraying state detection method, device, equipment and medium.

[0006] In order to achieve the above technical purpose, the technical scheme adopted by the embodiments of the present application comprises:

[0007] On the one hand, the present application provides a vehicle wiper water spraying state detection method, which comprises:

[0008] obtaining a wiper water spraying image of a target vehicle to be detected;

[0009] detecting the distribution of water columns in the wiper water spraying image by a segmentation detection algorithm to obtain initial water column image data;

[0010] Based on a straight line fitting algorithm, target water column image data is obtained from the initial water column image data;

[0011] extracting the start and end point information of each target water column in the target water column image data, and judging whether the start and end point information of each target water column conforms to the start and end point distribution rule obtained by clustering analysis of a plurality of sample water columns in advance;

[0012] Count the number of water columns that meet the start and end point distribution rule according to the start and end point information, and compare whether the number of water columns is consistent with the designed expected number;

[0013] If the number of water columns is inconsistent with the designed expected number, it is determined that the rain wiping and water spraying state of the vehicle is abnormal.

[0014] In addition, the vehicle rain wiping and water spraying state detection method according to the above-mentioned embodiments of the application can further have the following additional technical features:

[0015] Further, in an embodiment of the application, the distribution of water columns in the rain wiping and water spraying image is detected by a segmentation detection algorithm to obtain initial water column image data, which includes:

[0016] Determine the pixel mean value of the rain wiping and water spraying image;

[0017] According to the pixel mean value, a plurality of first target pixel points are determined from the rain wiping and water spraying image; wherein the absolute value of the difference between the pixel value of the first target pixel point and the pixel mean value is greater than a first preset threshold value;

[0018] In the rain wiping and water spraying image, a plurality of connected regions formed by the first target pixel points are determined;

[0019] Each connected region is cropped from the rain wiping and water spraying image to obtain the initial water column image data.

[0020] Further, in an embodiment of the application, the start and end point distribution rule is obtained by analyzing the following:

[0021] Obtain sample start and end point information of a batch of sample water columns; the sample start and end point information includes first start and end point information of real water columns and second start and end point information of false water columns;

[0022] Extract first feature data of the first start and end point information and second feature data of the second start and end point information;

[0023] According to the first feature data and the second feature data, cluster analysis is performed on each sample water column to obtain a first cluster center coordinate corresponding to a real water column category and a second cluster center coordinate corresponding to a false water column category;

[0024] According to the first cluster center coordinate and the second cluster center coordinate, the start and end point distribution rule is determined.

[0025] Further, in an embodiment of the application, according to the first feature data and the second feature data, cluster analysis is performed on each sample water column:

[0026] Selecting feature data corresponding to an arbitrary sample water column as an initial cluster center to establish an initial clustering cluster;

[0027] Calculating the similarity between each to-be-clustered feature data and the initial cluster center; the to-be-clustered feature data is feature data corresponding to a sample water column that has not been added to a clustering cluster;

[0028] When the similarity between the to-be-clustered feature data and the initial cluster center is less than or equal to a second preset threshold, adding the sample water column corresponding to the to-be-clustered feature data to the initial clustering cluster corresponding to the initial cluster center, and updating the cluster center coordinates of the initial clustering cluster;

[0029] Or, when the similarity between the to-be-clustered feature data and any initial cluster center is greater than the second preset threshold, taking the to-be-clustered feature data as an initial cluster center to additionally establish a new initial clustering cluster.

[0030] Further, in an embodiment of the present application, judging whether the start and end point information of the target water column conforms to the start and end point distribution rule comprises:

[0031] Extracting third feature data of the start and end point information of the target water column;

[0032] According to the third feature data, determining the clustering cluster category corresponding to the target water column;

[0033] If the clustering cluster category corresponding to the target water column is a real water column category, it is determined that the start and end point information of the target water column conforms to the start and end point distribution rule;

[0034] Or, if the clustering cluster category corresponding to the target water column is a false water column category, it is determined that the start and end point information of the target water column does not conform to the start and end point distribution rule.

[0035] Further, in an embodiment of the present application, after the step of screening the target water column image data from the initial water column image data, the method further comprises:

[0036] Excluding water column image data that is overlapped and branched through an NMS algorithm, and excluding water column image data with an area less than a third preset threshold and point-shaped water column image data through an outlier exclusion algorithm.

[0037] Further, in an embodiment of the present application, the expected number of designs is obtained through the following steps:

[0038] Scanning frame number information of the to-be-detected target vehicle;

[0039] According to the frame number information, obtaining the expected number of water injection holes of the to-be-detected target vehicle.

[0040] In another aspect, the embodiments of the present application provide a vehicle wiper water spray state detection device, which comprises:

[0041] An acquisition unit is configured to acquire a wiper water spray image of a target vehicle to be detected;

[0042] A segmentation unit is configured to detect the distribution of water columns in the wiper water spray image by using a segmentation detection algorithm to obtain initial water column image data;

[0043] A fitting unit is configured to filter target water column image data from the initial water column image data based on a straight line fitting algorithm;

[0044] An extraction unit is configured to extract start and end point information of each target water column in the target water column image data, and determine whether the start and end point information of each target water column conforms to a start and end point distribution rule obtained by clustering analysis of a plurality of sample water columns;

[0045] A comparison unit is configured to count the number of water columns whose start and end point information conforms to the start and end point distribution rule, and compare whether the number of water columns is consistent with a designed expected number;

[0046] A processing unit is configured to determine that the wiper water spray state of the vehicle is abnormal if the number of water columns is not consistent with the designed expected number.

[0047] In another aspect, the embodiments of the present application provide a computer device, which comprises:

[0048] At least one processor;

[0049] At least one memory configured to store at least one program;

[0050] When the at least one program is executed by the at least one processor, the at least one processor is caused to implement the vehicle wiper water spray state detection method described above.

[0051] In another aspect, the embodiments of the present application further provide a computer readable storage medium, which stores a processor executable program, and the processor executable program is used to implement the vehicle wiper water spray state detection method described above when executed by a processor.

[0052] The advantages and beneficial effects of the present application will be partially given in the following description, partially will become obvious from the following description, or will be understood by the practice of the present application:

[0053] The vehicle wiper water spraying state detection method disclosed by the embodiment of the application comprises: acquiring a wiper water spraying image of a target vehicle to be detected; detecting the distribution of water columns in the wiper water spraying image by a segmentation detection algorithm to obtain initial water column image data; screening target water column image data from the initial water column image data based on a straight line fitting algorithm; extracting start and end point information of each target water column in the target water column image data, and judging whether the start and end point information of each target water column conforms to a start and end point distribution rule obtained by clustering analysis of a plurality of sample water columns; counting the number of water columns whose start and end point information conforms to the start and end point distribution rule, and comparing whether the number of water columns is consistent with a designed expected number; and if the number of water columns is not consistent with the designed expected number, determining that the wiper water spraying state of the vehicle is abnormal. With the aid of artificial intelligence and other technical means, the method can realize automatic detection of whether the wiper water spraying state of the vehicle is abnormal, has high detection accuracy, and is beneficial to reducing the work burden of inspectors. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following introduces the drawings of the related technical solutions in the embodiments of the application or the prior art. It should be understood that the drawings in the following introduction are only for the convenience of clearly describing some of the embodiments of the technical solutions of the application, and other drawings can be obtained by those skilled in the art without creative labor on the premise that there is no creative labor.

[0055] Figure 1 A flowchart of a vehicle wiper water spraying state detection method provided in the embodiments of the application;

[0056] Figure 2 A structural diagram of a computer device provided in the embodiments of the application. DETAILED DESCRIPTION

[0057] The application will be further described below in conjunction with the drawings of the specification and specific embodiments. The described embodiments should not be regarded as limiting the application, and all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0058] In the following description, "some embodiments" are described, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to be limiting of this application.

[0060] Artificial Intelligence (AI) is the use of digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science, which attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence is the design principle and implementation method of various intelligent machines, so that the machine has the functions of perception, reasoning and decision-making. Artificial intelligence technology is a comprehensive discipline, involving a wide range of fields, both hardware and software technologies. Artificial intelligence basic technologies generally include sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0061] Machine Learning (ML) is a multi-disciplinary subject, involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is dedicated to studying how computers simulate or implement human learning behavior to acquire new knowledge or skills, and reorganize existing knowledge structure to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental approach to making computers intelligent. Its application is widespread in various fields of artificial intelligence. Machine learning (deep learning) generally includes artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and rule-based learning.

[0062] Automobile is an important means of transportation at present, which brings great convenience to people's life and work. On the automobile production line, it is generally necessary to detect whether various parts and components in the vehicle production process are normal. For example, for the vehicle wiper, it is necessary to detect whether the water hole can normally spray water column, if there is an abnormality, the defective product needs to be repaired or the problem needs to be traced.

[0063] In the related art, the inspector generally visually observes the water spraying function of the wiper to perform corresponding inspection. However, on the one hand, after the integrated wiper is designed and installed on part of the vehicle models, the number of water spraying holes of the wiper increases and the water holes become thinner, which leads to high observation difficulty. On the other hand, the detection means through manual inspection has many uncertainties (such as uncontrollable human factors such as external shielding factors or detection fatigue), and there are many misjudgment cases, which leads to low detection accuracy and risks of outflow of defective products, and affects the normal use of customers.

[0064] Therefore, in the embodiments of the present application, a vehicle wiper water spraying state detection method is provided. A wiper water spraying image of a target vehicle to be detected is obtained. The distribution of water columns in the wiper water spraying image is detected by a segmentation detection algorithm to obtain initial water column image data. Target water column image data is selected from the initial water column image data based on a straight line fitting algorithm. The start and end point information of each target water column in the target water column image data is extracted. It is judged whether the start and end point information of each target water column conforms to the start and end point distribution rule obtained by clustering analysis of a plurality of sample water columns in advance. The number of water columns whose start and end point information conforms to the start and end point distribution rule is counted, and it is compared whether the number of water columns is consistent with the designed expected number. If the number of water columns is not consistent with the designed expected number, it is determined that the wiper water spraying state of the vehicle is abnormal. This method can realize automatic detection of whether the wiper water spraying state of the vehicle is abnormal by means of artificial intelligence and other technical means, has high detection accuracy, and is beneficial to reduce the work burden of the inspector. Moreover, the function of the vehicle parts is detected by using an automatic intelligent detection system, which can improve the detection accuracy in the production process and realize traceable management, which is not only beneficial to reduce the cost of enterprises, but also conforms to the concept of intelligent manufacturing and is the development trend of the automobile industry.

[0065] Please refer to Figure 1 , Figure 1 is a flowchart of a vehicle wiper water spraying state detection method provided by the embodiments of the present application. Referring to Figure 1 , the vehicle wiper water spraying state detection method includes but is not limited to:

[0066] Step 110, obtaining a wiper water spraying image of a target vehicle to be detected;

[0067] In this step, when detecting the water spraying state of the vehicle wiper, the wiper water spraying image of the target vehicle to be detected is obtained. Specifically, the relevant staff can drive the car to the corresponding detection position, trigger the wiper to work, and then synchronously collect the wiper water spraying image through the camera device arranged at the specified position. In some embodiments, when the vehicle is tested, the frame number information of the vehicle can be obtained through the code scanner, so as to record the model information of the car. The detection result can be stored in the server at the same time, which is convenient for real-time query and traceability, and improves the availability of information.

[0068] Step 120, detecting the distribution of water column in the wiper water spraying image through a segmentation detection algorithm to obtain initial water column image data;

[0069] In this step, after obtaining the wiper water spraying image of the target vehicle to be detected, the wiper water spraying image can be input into the state detection model. In the embodiment of the present application, the state detection model is used to detect the number of water columns of the vehicle wiper water spraying, so as to judge whether each water outlet of the wiper can normally water. It can integrate various artificial intelligence algorithms inside, output the detected wiper water spraying state result by executing each artificial intelligence algorithm. Specifically, the working principle of the state detection model will be described in the form of algorithm logic below, and the model architecture of the present application is not limited.

[0070] In the embodiment of the present application, it can be understood that in the wiper water spraying image, the background is mainly the normal vehicle surface color (such as black), and the position of the water column will be different from the color of the background. Therefore, there will be an obvious boundary between the image content of the water column and the image content of the background in the wiper water spraying image. In the embodiment of the present application, when the wiper water spraying image is processed by image recognition, an image segmentation detection algorithm can be used to segment each region of interest, and the region of interest is the image region where the water column may exist, so as to obtain the initial water column image data.

[0071] Specifically, in some embodiments, the segmentation detection algorithm adopted can be an edge detection-based method, i.e., detecting edges containing different regions to solve the segmentation problem, by identifying points in the image data where the gray scale changes significantly. In the embodiments of the present application, the edges of the background environment image data and the region of interest are detected, which often have very different gray value characteristics. A commonly used method for detecting edges is to calculate the derivative of the gray scale change at the point where the gray scale changes. The specific method can be divided into two categories: a lookup-based category and a zero-crossing-based category. The lookup-based method detects the boundary by finding the maximum and minimum values of the first derivative of the gray scale change in the image to be identified, which is usually located in the direction of the maximum gradient. The zero-crossing-based method finds the boundary by finding the zero-crossing of the second derivative of the gray scale change in the image data, which is usually the zero-crossing of the Laplacian or the zero-crossing of the nonlinear difference representation. The specific means adopted in the embodiments of the present application are not limited.

[0072] In other embodiments, the task mode of target detection combined with semantic segmentation can also be used to implement the segmentation of the initial water column image data. For example, a network model such as Faster RCNN can be used to first detect the rectangular frame region where the target (i.e., the water column) in the rain wiper water spray image is located, i.e., to complete the target detection task, and then the obtained rectangular frame region is further input into a semantic segmentation model to obtain the class information of each position at the pixel level. Integrating these information, the corresponding initial water column image data can be obtained. Specifically, the semantic segmentation model that can be used in the present application includes but is not limited to FCN, U-net, SegNet, etc., and the present application does not limit this.

[0073] Step 130, filtering target water column image data from the initial water column image data based on a straight line fitting algorithm;

[0074] In this step, after obtaining the initial water column image data, a straight line fitting algorithm can be used to filter the initial water column image data to obtain the target water column image data. Here, it can be understood that generally, the water column sprayed by the rain wiper is a straight line or can be fitted as a straight line. Therefore, in the embodiments of the present application, the straight line fitting algorithm is used to remove the image data whose trajectory does not obviously conform to the water column. In this way, on the one hand, it can remove some interference data and improve the accuracy of the detection result, and on the other hand, it can also reduce the data processing amount and improve the use efficiency of hardware resources.

[0075] Step 140, extracting start and end point information of each target water column in the target water column image data, and judging whether the start and end point information of each target water column conforms to the start and end point distribution rule obtained by clustering analysis of a plurality of sample water columns;

[0076] In this step, after obtaining the target water column image data, the start and end point information of each target water column can be extracted, i.e., the positions of the start and end points of the water column in each target water column, which can be represented by the coordinate positions of the pixels in the image. When the position of the camera device for collecting the rain wiper water spray image is fixed, the deviation of the start and end points of the target water column of each vehicle is generally not too large. If some target water columns deviate abnormally, it is likely that the reflection of the water column is misidentified, so in the embodiments of the present application, the water column images of a plurality of sample water columns can be collected in advance at the same camera position, and the real water columns and false water columns are clustered and analyzed to obtain the start and end point distribution rule of the sample water columns. When detecting the target vehicle to be detected, it can be judged whether the start and end point information of each target water column conforms to the start and end point distribution rule, so as to determine which target water columns are real water columns and which target water columns are false reflection water columns.

[0077] Step 150, count the number of water columns whose start and end point information conforms to the start and end point distribution rule, and compare whether the number of water columns is consistent with the designed expected number;

[0078] Step 160, if the number of water columns is not consistent with the designed expected number, it is determined that the rain wiper water spray state of the vehicle is abnormal.

[0079] In the embodiments of the present application, when it is determined that the start and end point information of the target water column conforms to the start and end point distribution rule, it means that it is a real water column. Therefore, the number of water columns conforming to the start and end point distribution rule can be counted, and then the number of water columns is compared with the designed expected number corresponding to the target vehicle to be detected, to determine whether they are consistent. If they are consistent, it means that the rain wiper of the target vehicle to be detected can normally spray water, and the state is normal. On the contrary, if they are not consistent, it means that the rain wiper of the target vehicle to be detected cannot normally spray water, and the state is abnormal; generally, the reason for the abnormal state is that there is a problem with the holes on the rain wiper, which causes the rain wiper to be unable to normally spray water, so the target vehicle to be detected can be repaired at this time.

[0080] In some embodiments, the distribution of the water column in the rain wiper water spray image is detected by a segmentation detection algorithm to obtain initial water column image data, including:

[0081] Determine the pixel mean value of the rain wiper water spray image;

[0082] According to the pixel mean value, a plurality of first target pixel points are determined from the rain wiper water spray image; wherein the absolute value of the difference between the pixel value of the first target pixel point and the pixel mean value is greater than a first preset threshold value;

[0083] In the rain wiper water spray image, a connected region composed of a plurality of first target pixel points is determined;

[0084] Crop each of the connected regions from the rain screen water spraying image to obtain the initial water column image data.

[0085] In the embodiments of the present application, when the initial water column image data is segmented from the rain screen water spraying image, the pixel mean value of the rain screen water spraying image can be determined first. As described above, generally, the pixel value of the pixel point in the water column part of the rain screen water spraying image will be different from the pixel value of the pixel point in the other background part. Therefore, the region where the water column is located can be determined based on the pixel point with a large deviation in the rain screen water spraying image.

[0086] Specifically, in the embodiments of the present application, a deviation comparison threshold can be preset, the difference between the pixel value of each pixel point in the rain screen water spraying image and the pixel mean value is calculated, and the deviation between the pixel value of each pixel point and the pixel mean value is determined by comparing the difference with the comparison threshold. When the absolute value of the difference between the pixel value of a certain pixel point and the pixel mean value is greater than a first preset threshold, it is indicated that the pixel value of the pixel point deviates greatly, and it can be determined that the pixel point has a large probability of being in the water column region. In this way, a plurality of first target pixel points can be determined from the rain screen water spraying image, and the position of the first target pixel point is likely to be the position of the water column. Then, according to the determined first target pixel points, the connected regions formed by these pixel points can be determined. Then, according to the position of the determined connected regions, each connected region can be cropped from the rain screen water spraying image, so as to obtain the initial water column image data.

[0087] In some embodiments, the start and end point distribution rule is analyzed by the following way:

[0088] Obtaining sample start and end point information of a batch of sample water columns; the sample start and end point information includes first start and end point information of a real water column and second start and end point information of a false water column;

[0089] Extracting first feature data of the first start and end point information and second feature data of the second start and end point information;

[0090] According to the first feature data and the second feature data, performing cluster analysis on each sample water column to obtain a first cluster center coordinate corresponding to a real water column category and a second cluster center coordinate corresponding to a false water column category;

[0091] According to the first cluster center coordinate and the second cluster center coordinate, determining a start and end point distribution rule.

[0092] In the embodiments of the present application, the starting and ending point distribution rule can be obtained through clustering analysis. Specifically, sample starting and ending point information corresponding to a batch of sample water columns can be obtained, the sample starting and ending point information including first starting and ending point information of a real water column and second starting and ending point information of a false water column. The false water column here can be a water column caused by a reflection of collected images and the like. Then, first feature data and second feature data can be extracted from the first starting and ending point information and the second starting and ending point information. Clustering analysis can be performed according to the feature data, and two categories, i.e., a real water column category and a false water column category, can be obtained. The cluster center coordinates of the two categories are different, the real water column category corresponds to a first cluster center coordinate, and the false water column category corresponds to a second cluster center coordinate. Here, the information of the first cluster center coordinate and the second cluster center coordinate can be used as the starting and ending point distribution rule.

[0093] Specifically, in the embodiments of the present application, when clustering analysis is performed, first, feature data corresponding to an arbitrary sample water column can be selected as an initial cluster center to establish an initial clustering cluster. Then, for each sample water column whose clustering category is not determined, the feature data corresponding to the sample water column is recorded as to-be-clustered feature data. According to the to-be-clustered feature data, the similarity between the to-be-clustered feature data and the initial cluster center can be determined. It can be understood that the similarity between the to-be-clustered feature data and the initial cluster center can reflect the similarity between two sample water columns. The greater the similarity, the closer the sample water column is to the category corresponding to the initial clustering cluster. Therefore, in the embodiments of the present application, a similarity threshold value, recorded as a second preset threshold value, can be set in advance. If the similarity between the to-be-clustered feature data and a certain initial cluster center is greater than or equal to the second preset threshold value, it indicates that the sample water column corresponding to the to-be-clustered feature data is sufficiently similar to the sample water column in the initial clustering cluster corresponding to the initial cluster center. Therefore, at this time, the sample water column corresponding to the to-be-clustered feature data can be added to the initial clustering cluster corresponding to the initial cluster center, and the cluster center coordinates of the initial clustering cluster are updated. On the contrary, if the similarity between the to-be-clustered feature data and all the established initial cluster centers is less than the second preset threshold value, it indicates that the sample water column corresponding to the to-be-clustered feature data is not similar to the sample water columns in the initial clustering clusters corresponding to the initial cluster centers. At this time, the sample water column corresponding to the to-be-clustered feature data needs to be additionally classified into a category, i.e., a new initial clustering cluster is additionally established with the to-be-clustered feature data as the initial cluster center.

[0094] In some embodiments, determining whether the starting and ending point information of each target water column conforms to the starting and ending point distribution rule comprises:

[0095] extracting third feature data of the starting and ending point information of the target water column;

[0096] determining a clustering cluster category corresponding to the target water column according to the third feature data;

[0097] If the cluster category corresponding to the target water column is a real water column category, it is determined that the start and end point information of the target water column conforms to the start and end point distribution rule.

[0098] Alternatively, if the cluster category corresponding to the target water column is a false water column category, it is determined that the start and end point information of the target water column does not conform to the start and end point distribution rule.

[0099] In the embodiments of the present application, when judging whether the start and end point information of the target water column conforms to the start and end point distribution rule, clustering analysis can be performed thereon. Specifically, the third feature data corresponding thereto is extracted, and clustering is performed thereon according to the foregoing process to determine the cluster category corresponding to the target water column. If the cluster category corresponding to the target water column is a real water column category, it is indicated that the start and end point information of the target water column conforms to the start and end point distribution rule; otherwise, if the cluster category corresponding to the target water column is a false water column category, it is indicated that the start and end point information of the target water column does not conform to the start and end point distribution rule.

[0100] In some embodiments, after the step of screening the target water column image data from the initial water column image data, the NMS algorithm can be used to screen out the water column image data with overlapping branches, and the outlier screening algorithm can be used to screen out the water column image data with an area less than a third preset threshold value and the point-like water column image data, so as to improve the accuracy of the target water column image data.

[0101] In some embodiments, the expected number of designs is obtained by the following steps:

[0102] Scanning the frame number information of the target vehicle to be detected;

[0103] According to the frame number information, the expected number of water spray holes of the target vehicle to be detected is obtained.

[0104] In the embodiments of the present application, the expected number of designs corresponding to the target vehicle to be detected can be the number of water spray holes on the wiper. Specifically, this number can be related to the vehicle model of each vehicle, and therefore it can be obtained by scanning the frame number information of the target vehicle to be detected, and the present application does not limit the specific value thereof.

[0105] The embodiments of the present application also provide a vehicle wiper water spray state detection device, which comprises:

[0106] An acquisition unit is configured to acquire a wiper water spray image of a target vehicle to be detected;

[0107] A segmentation unit is configured to detect the distribution of water columns in the wiper water spray image by using a segmentation detection algorithm to obtain initial water column image data;

[0108] a fitting unit, configured to filter target water column image data from the initial water column image data based on a straight line fitting algorithm;

[0109] an extraction unit, configured to extract start and end point information of each target water column in the target water column image data, and determine whether the start and end point information of the each target water column conforms to a start and end point distribution rule obtained through clustering analysis of a plurality of sample water columns;

[0110] a comparison unit, configured to count a number of water columns whose start and end point information conforms to the start and end point distribution rule, and compare whether the number of water columns is consistent with a designed expected number;

[0111] a processing unit, configured to determine that a rain and water spraying state of the vehicle is abnormal if the number of water columns is not consistent with the designed expected number.

[0112] It can be understood that, Figure 1 the content in the vehicle rain and water spraying state detection method embodiment shown in the vehicle rain and water spraying state detection device embodiment, the vehicle rain and water spraying state detection device embodiment specifically realizes the same functions as the vehicle rain and water spraying state detection method embodiment shown in Figure 1 and achieves the same beneficial effects as the vehicle rain and water spraying state detection method embodiment shown in Figure 1 and achieves the same beneficial effects as the vehicle rain and water spraying state detection method embodiment shown in

[0113] Referring to Figure 2 , the embodiment of the application further discloses a computer device, which comprises:

[0114] at least one processor 201;

[0115] at least one memory 202 for storing at least one program;

[0116] When the at least one program is executed by the at least one processor 201, the at least one processor 201 realizes the vehicle rain and water spraying state detection method embodiment shown in Figure 1 .

[0117] It can be understood that, the content in the vehicle rain and water spraying state detection method embodiment shown in the computer device embodiment, the computer device embodiment specifically realizes the same functions as the vehicle rain and water spraying state detection method embodiment shown in Figure 1 and achieves the same beneficial effects as the vehicle rain and water spraying state detection method embodiment shown in Figure 1 and achieves the same beneficial effects as the vehicle rain and water spraying state detection method embodiment shown in Figure 1 and achieves the same beneficial effects as the vehicle rain and water spraying state detection method embodiment shown in

[0118] The embodiment of the present application further discloses a computer readable storage medium, wherein a processor executable program is stored, and the processor executable program is used for realizing the vehicle wiper water spraying state detection method as shown in the embodiment of the present application when executed by the processor. Figure 1 The embodiment of the present application further discloses a computer readable storage medium, wherein a processor executable program is stored, and the processor executable program is used for realizing the vehicle wiper water spraying state detection method as shown in the embodiment of the present application when executed by the processor.

[0119] It can be understood that the contents in the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application are all applicable to the computer readable storage medium embodiment, the function realized by the computer readable storage medium embodiment is the same as the function realized by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application, and the beneficial effects achieved are the same as the beneficial effects achieved by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application. Figure 1 It can be understood that the contents in the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application are all applicable to the computer readable storage medium embodiment, the function realized by the computer readable storage medium embodiment is the same as the function realized by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application, and the beneficial effects achieved are the same as the beneficial effects achieved by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application. Figure 1 It can be understood that the contents in the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application are all applicable to the computer readable storage medium embodiment, the function realized by the computer readable storage medium embodiment is the same as the function realized by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application, and the beneficial effects achieved are the same as the beneficial effects achieved by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application. Figure 1 It can be understood that the contents in the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application are all applicable to the computer readable storage medium embodiment, the function realized by the computer readable storage medium embodiment is the same as the function realized by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application, and the beneficial effects achieved are the same as the beneficial effects achieved by the embodiment of the vehicle wiper water spraying state detection method as shown in the embodiment of the present application.

[0120] In some alternative embodiments, the functions / operations mentioned in the block diagram can not occur in the order mentioned in the operation diagram. For example, depending on the functions / operations involved, two blocks shown in succession can actually be executed substantially simultaneously or the blocks can sometimes be executed in reverse order. In addition, the embodiments presented and described in the flowcharts of the present application are provided by way of example, and the purpose is to provide a more comprehensive understanding of the technology. The disclosed method is not limited to the operations and logical flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of larger operations are independently executed.

[0121] In addition, although the present application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features can be integrated in a single physical system and / or software module, or one or more functions and / or features can be implemented in separate physical systems or software modules. It can also be understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present application. Rather, given the properties, functions and internal relationships of the various functional modules in the system disclosed herein, the actual implementation of the module will be understood within the routine skill of an engineer. Therefore, those skilled in the art can implement the present application as set forth in the claims without undue experimentation using ordinary skill. It can also be understood that the disclosed specific concepts are merely illustrative and are not intended to limit the scope of the present application, the scope of the present application being determined by the full scope of the appended claims and their equivalents.

[0122] If the functions are implemented in software, the functions can be stored in or implemented as one or more instructions or code on a computer-readable medium. Computer-readable media include both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage medium can be any available medium that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, or twisted pair, then the coaxial cable, fiber optic cable, or twisted pair are included in the definition of medium. Disk and disc, as used herein, include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), and Blu-Ray® disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0123] In other words, like a human driver of a vehicle, an autonomous vehicle can be programmed to follow traffic laws and to make decisions based on its environment. For example, an autonomous vehicle can be programmed to follow a speed limit, to stop at a stop sign, to yield to a pedestrian, to merge onto a highway, to change lanes, to park, and so on. In some embodiments, an autonomous vehicle can be programmed to follow traffic laws and to make decisions based on its environment using a machine learning algorithm. For example, an autonomous vehicle can be programmed to follow a speed limit, to stop at a stop sign, to yield to a pedestrian, to merge onto a highway, to change lanes, to park, and so on using a machine learning algorithm.

[0124] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber (optical), and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can also be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example, via optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

[0125] It should be understood that aspects of the application can be implemented in hardware, software, firmware or a combination of them. In the above embodiments, various steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies, or combinations thereof, can be used: a discrete logic circuit having logic gates for implementing logic functions upon data signals, an application specific integrated circuit having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), and / or the like.

[0126] In the above description of the present specification, the description referring to the terms "one embodiment", "another embodiment" or "certain embodiments" or the like means that a specific feature, structure, material or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. The illustrative expressions of the above terms do not necessarily refer to the same embodiment or example in the present specification. Also, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in an appropriate manner.

[0127] Although the embodiments of the present application have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, alternatives and variations to these embodiments can be made without departing from the principles and spirit of the application, and the scope of the application is defined by the claims and their equivalents.

[0128] The above is a specific description of the preferred embodiments of the present application, but the present application is not limited to the embodiments, and those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present application, and these equivalent modifications or substitutions are included in the scope defined by the claims of the present application

[0129] In the above description of the present specification, the description referring to the terms "one embodiment", "another embodiment" or "certain embodiments" or the like means that a specific feature, structure, material or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. The illustrative expressions of the above terms do not necessarily refer to the same embodiment or example in the present specification. Also, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in an appropriate manner.

[0130] Although the embodiments of the present application have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, alternatives and variations to these embodiments can be made without departing from the principles and spirit of the application, and the scope of the application is defined by the claims and their equivalents.

Claims

1. A vehicle wiper water spray state detection method characterized by, The method comprises: acquiring a wiper water spraying image of a target vehicle to be detected; detecting the distribution of water columns in the wiper water spraying image by a segmentation detection algorithm to obtain initial water column image data; screening target water column image data from the initial water column image data based on a straight line fitting algorithm; extracting start and end point information of each target water column in the target water column image data, and judging whether the start and end point information of each target water column conforms to a start and end point distribution rule obtained by clustering analysis of a plurality of sample water columns in advance; counting the number of water columns whose start and end point information conforms to the start and end point distribution rule, and comparing whether the number of water columns is consistent with a designed expected number; if the number of water columns is not consistent with the designed expected number, determining that the wiper water spraying state of the vehicle is abnormal.

2. The method of claim 1, wherein The method comprises: determining the pixel mean value of the wiper water spraying image; determining a plurality of first target pixel points from the wiper water spraying image according to the pixel mean value, wherein the absolute value of the difference between the pixel value of the first target pixel point and the pixel mean value is greater than a first preset threshold value; determining a plurality of connected regions formed by the first target pixel points in the wiper water spraying image; cropping each connected region from the wiper water spraying image to obtain the initial water column image data.

3. The method of claim 1, wherein The start and end point distribution rule is obtained by the following method: acquiring sample start and end point information of a batch of sample water columns; the sample start and end point information comprises first start and end point information of real water columns and second start and end point information of false water columns; extracting first feature data of the first start and end point information and second feature data of the second start and end point information; performing clustering analysis on each sample water column according to the first feature data and the second feature data to obtain first cluster center coordinates corresponding to the real water column category and second cluster center coordinates corresponding to the false water column category; determining the start and end point distribution rule according to the first cluster center coordinates and the second cluster center coordinates.

4. The method of claim 3, wherein The clustering analysis on each sample water column according to the first feature data and the second feature data comprises: selecting feature data corresponding to an arbitrary sample water column as an initial cluster center to establish an initial clustering cluster; calculating the similarity between each clustering feature data and the initial cluster center; the clustering feature data is feature data corresponding to a sample water column not added to the clustering cluster; when the similarity between the clustering feature data and the initial cluster center is less than or equal to a second preset threshold value, adding the sample water column corresponding to the clustering feature data to the initial clustering cluster corresponding to the initial cluster center, and updating the cluster center coordinates of the initial clustering cluster; or, when the similarity between the clustering feature data and any initial cluster center is greater than the second preset threshold value, taking the clustering feature data as an initial cluster center to additionally establish a new initial clustering cluster.

5. The method of claim 4, wherein The method comprises: extracting third feature data of the start and end point information of the target water column; According to the third feature data, a cluster cluster category corresponding to the target water column is determined; If the cluster cluster category corresponding to the target water column is a real water column category, it is determined that the start and end point information of the target water column conforms to the start and end point distribution rule; Or, if the cluster cluster category corresponding to the target water column is a false water column category, it is determined that the start and end point information of the target water column does not conform to the start and end point distribution rule.

6. The method of claim 1, wherein After the step of screening the target water column image data from the initial water column image data, the method further comprises: The NMS algorithm is used to filter out the overlapping and branched water column image data, and the outlier filtering algorithm is used to filter out the water column image data with an area less than a third preset threshold and the point-like water column image data.

7. The method of claim 1-6, wherein The design expected number is obtained by the following steps: The frame number information of the target vehicle to be detected is scanned; According to the frame number information, the design expected number of water injection holes of the target vehicle to be detected is obtained.

8. A vehicle wiper water spray state detecting device characterized by comprising: The device comprises: An acquisition unit is configured to acquire a rain wiper water injection image of a target vehicle to be detected; A segmentation unit is configured to detect the distribution of water columns in the rain wiper water injection image by using a segmentation detection algorithm to obtain initial water column image data; A fitting unit is configured to screen target water column image data from the initial water column image data based on a straight line fitting algorithm; An extraction unit is configured to extract start and end point information of each target water column in the target water column image data, and determine whether the start and end point information of each target water column conforms to a start and end point distribution rule obtained by clustering analysis of a plurality of sample water columns in advance; A comparison unit is configured to count the number of water columns whose start and end point information conforms to the start and end point distribution rule, and compare whether the number of water columns is consistent with a design expected number; A processing unit is configured to determine that the rain wiper water injection state of the vehicle is abnormal if the number of water columns is not consistent with the design expected number.

9. A computer device, comprising: It comprises: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a vehicle rain wiper water injection state detection method according to any one of claims 1-7.

10. A computer readable storage medium having stored therein a program which is executable by a processor, characterized in that: The program executable by the processor when executed by the processor is used to implement a vehicle rain wiper water injection state detection method according to any one of claims 1-7.

Images