New energy vehicle cylinder cover detection system and method

Abstract

The invention belongs to the technical field of new energy vehicle cylinder cover detection, and particularly relates to a new energy vehicle cylinder cover detection system and method. According to the invention, through fine detection and management of the new energy vehicle cylinder cover, accurate identification and positioning of cylinder cover defects are realized, the efficiency and accuracy of cylinder cover quality detection are effectively improved, and by using an image processing and comparison technology, tiny defects on the surface of the cylinder cover can be accurately identified, and the detection precision is improved. The method avoids the possible leak detection or false detection condition in a traditional detection method, can also quantitatively evaluate the difference between the cylinder cover and the standard state, provides powerful support for continuous optimization of the cylinder cover quality, can further deeply analyze the reason that the cylinder cover deviates from the standard, helps a producer to accurately position the problem in the production link, and improves the production efficiency. And finally, a corresponding repairing scheme can be automatically matched according to the severity degree of the defect, so that the repairing efficiency and quality of the cylinder cover are improved, and the reliability and safety of cylinder cover production are ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of cylinder head detection for new energy vehicles, and in particular relates to a cylinder head detection system and method for new energy vehicles. Background Art

[0002] With the rapid development of new energy vehicles, the cylinder head, as a key component of the engine, its quality and performance are crucial to the operating efficiency and safety of the entire vehicle. However, various defects such as cracks, pores, inclusions, etc. are inevitable in the production process of the cylinder head. These defects will not only affect the service life of the cylinder head, but may also cause serious safety accidents. Therefore, how to efficiently and accurately detect defects on the surface of the cylinder head has become a technical problem that needs to be solved urgently in the field of new energy vehicle manufacturing.

[0003] Although there are some cylinder head inspection methods in the prior art, most of these methods rely on manual visual inspection or the use of traditional inspection equipment, and have problems such as low inspection efficiency, poor accuracy, and susceptibility to human factors. At the same time, there is a lack of corresponding verification process in the inspection process, which will lead to a decrease in the credibility of the cylinder head inspection results. As the cylinder head production process proceeds, its production accuracy will inevitably gradually show uncontrollable deviations. The inspection method in the prior art cannot locate the corresponding production link based on the inspection results, which will undoubtedly lead to the potential risks in the cylinder head production link cannot be discovered and improved in time. Based on this, the present invention proposes a new energy vehicle cylinder head inspection method to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide a cylinder head detection system and method for new energy vehicles, which can realize efficient and accurate detection of cylinder heads, improve the credibility of detection results through a verification process, and locate the corresponding production links based on the detection results, so as to timely discover and improve potential risks in the cylinder head production links.

[0005] The technical solution adopted by the present invention is as follows: A new energy vehicle cylinder head detection method, comprising: Collecting cylinder head image data of a new energy vehicle, wherein the cylinder head image data includes cylinder head surface feature information and edge feature information; Performing defect recognition on the cylinder head according to the cylinder head image data and preset standard image data, and determining the detection state of the cylinder head according to the result of the defect recognition, wherein the detection state includes a normal state and a defect state; Extracting the image features of all cylinder heads in the normal state, establishing an image feature database of the cylinder heads in the normal state, and outputting the allowable deviation between each cylinder head and the standard image data according to the image feature database; Determining the deviation type of the cylinder head according to the allowable deviation, and determining the improvement position of the cylinder head production process according to the deviation type; The defect area and defect type of the cylinder head in the defect state are extracted, and the repair level of the cylinder head is determined according to the defect area and defect type. Then, corresponding repair plans are matched one by one according to the repair level of the cylinder head, and uploaded to the management end synchronously.

[0006] In a preferred embodiment, the acquisition of cylinder head image data of new energy vehicles includes global acquisition and local acquisition, wherein the global acquisition is used to obtain overall image information of the cylinder head, and the local acquisition is used to obtain detailed image information of key parts of the cylinder head.

[0007] In a preferred embodiment, the defect recognition includes edge recognition and surface recognition, wherein the surface recognition is used to detect surface defects of the cylinder head, and the edge recognition is used to detect edge defects of the cylinder head; The priority of edge recognition is higher than that of surface recognition.

[0008] In a preferred embodiment, the step of identifying defects of the cylinder head based on the cylinder head image data and preset standard image data includes: Acquire cylinder head image data and standard image data, and perform edge curve extraction to obtain a curve to be calibrated corresponding to the cylinder head image data and a standard curve corresponding to the standard image data; Comparing the curve to be calibrated with the standard curve to determine whether the curve to be calibrated coincides with the standard curve; If the curve to be calibrated coincides with the standard curve, it indicates that the corresponding cylinder head is qualified, and the surface recognition of the cylinder head continues to be performed, and after the cylinder head passes the surface recognition, its corresponding detection state is recorded as a normal state; If the curve to be calibrated does not coincide with the standard curve, it indicates that the corresponding cylinder head is unqualified, and the detection status of the corresponding cylinder head is directly recorded as a defective status.

[0009] In a preferred embodiment, when performing surface recognition on the cylinder head, the cylinder head image data is converted into a color space consistent with the standard image data, and the cylinder head image data after the color space conversion is subjected to noise processing to obtain an image to be proofread; Vectorizing the pixel values ​​in the image to be proofread and the standard image data to obtain a vector to be proofread and a standard vector; Obtaining a comparison function, and inputting the vector to be checked and the standard vector into the comparison function together, and recording the output result of the comparison function as a comparison score; Obtaining an evaluation interval, and comparing the comparison score with the evaluation interval; If the comparison score belongs to the evaluation interval, it indicates that the defect of the cylinder head surface is within the allowable range, and the detection state of the corresponding cylinder head is output as a normal state; If the comparison score does not belong to the evaluation interval, it indicates that there are obvious defects on the surface of the cylinder head, and the detection state of the corresponding cylinder head is output as a defective state.

[0010] In a preferred solution, after the detection state of the cylinder head is output as a normal state, a verification process is synchronously performed, and the specific process is as follows: Counting the pixel coordinates of the pixel point deviation positions between the image to be proofread and the standard image data, and the pixel deviation values ​​at each pixel coordinate; Dividing the deviation area between the image to be proofread and the standard image according to the pixel coordinates, and counting the number of the deviation areas and recording them as the first verification condition parameter; Collecting an average value of the pixel deviation values ​​in each of the deviation areas and recording it as a second verification condition parameter; Obtaining a verification function, and inputting the first verification condition parameter and the second verification condition parameter into the verification function, and recording an output result of the verification function as a verification score; Obtaining a verification threshold, and comparing the verification threshold with a verification score; When the verification score is higher than the verification threshold, it indicates that the corresponding cylinder head has failed the verification, and the detection state of the corresponding cylinder head is converted from a normal state to a defective state; When the verification score is lower than or equal to the verification threshold, it indicates that the corresponding cylinder head has passed the verification, and the detection state of the corresponding cylinder head is maintained in a normal state.

[0011] In a preferred embodiment, the step of outputting the allowable deviation between each cylinder head and the standard image data according to the image feature database comprises: Based on the current detection node, reversely construct the monitoring period; Setting a plurality of sampling nodes within the monitoring period, wherein the sampling nodes correspond to time nodes at which the cylinder head passes the verification process; The second verification condition parameter corresponding to each cylinder head image data at each sampling node is extracted from the image feature database and recorded as the allowable deviation between the cylinder head and the standard image data.

[0012] In a preferred embodiment, the step of determining the deviation type of the cylinder head according to the allowable deviation, and determining the improvement position of the cylinder head production process according to the deviation type, comprises: Acquiring the allowable deviations and arranging the allowable deviations according to the detection sequence of the cylinder head; Obtaining a condition function, inputting the allowable deviation into the condition function, and recording an output result of the condition function as an evaluation condition parameter; Randomly select multiple groups of adjacent allowable deviations; Obtaining an evaluation function, and inputting the allowable deviation and the evaluation condition parameter into the evaluation function, and recording the output result of the evaluation function as a classification parameter; A classification threshold is obtained, and the classification threshold is compared with a classification parameter. When the classification parameter is greater than the classification threshold, the deviation area corresponding to the allowable deviation is recorded as a risk area, and the production link corresponding to the risk area is recorded as a risk link. Otherwise, the deviation area corresponding to the allowable deviation is recorded as a normal area, and the production link corresponding to the normal area is recorded as a normal link.

[0013] The present invention also provides a cylinder head detection system for a new energy vehicle, using the above-mentioned cylinder head detection method for a new energy vehicle, comprising: A data acquisition module, the data acquisition module is used to collect cylinder head image data of a new energy vehicle, wherein the cylinder head image data includes cylinder head surface feature information and edge feature information; a defect recognition module, the defect recognition module being used to perform defect recognition on the cylinder head according to the cylinder head image data and preset standard image data, and determine a detection state of the cylinder head according to a result of the defect recognition, wherein the detection state includes a normal state and a defect state; a deviation degree measuring module, the deviation degree measuring module is used to extract the image features of all cylinder heads in the normal state, establish an image feature database of the cylinder heads in the normal state, and output the allowable deviation between each cylinder head and the standard image data according to the image feature database; A deviation tracing module, the deviation tracing module is used to determine the deviation type of the cylinder head according to the allowable deviation, and determine the improvement position of the cylinder head production link according to the deviation type; A defect output module is used to extract the defect area and defect type of the cylinder head in the defect state, and determine the repair level of the cylinder head according to the defect area and defect type, and then match the corresponding repair solutions one by one according to the repair level of the cylinder head, and upload them to the management end synchronously.

[0014] And, an electronic device, the electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; Among them, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned new energy vehicle cylinder head detection method.

[0015] The technical effects achieved by the present invention are: The present invention realizes the precise identification and positioning of cylinder head defects by carrying out refined detection and management of cylinder heads of new energy vehicles, effectively improving the efficiency and accuracy of cylinder head quality detection. By utilizing image processing and comparison technology, it can quickly and accurately identify tiny defects on the surface of the cylinder head, avoiding missed detection or false detection that may occur in traditional detection methods. At the same time, it can also quantitatively evaluate the difference between the cylinder head and the standard state, providing strong support for the continuous optimization of the cylinder head quality. In addition, it will deeply analyze the reasons why the cylinder head deviates from the standard, help producers accurately locate problems in the production process, so as to implement improvement measures in a targeted manner, and finally automatically match the corresponding repair plan according to the severity of the defect, thereby improving the repair efficiency and quality of the cylinder head and ensuring the reliability and safety of cylinder head production. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flow chart of the method of the present invention; Figure 2 It is a schematic diagram of the system module of the present invention; Figure 3 It is a schematic diagram of the structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0019] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" that appears in different places in this specification does not refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0020] See also Figure 1 As shown, the present invention provides a cylinder head detection method for a new energy vehicle, comprising: S1. Collecting cylinder head image data of a new energy vehicle, wherein the cylinder head image data includes surface feature information and edge feature information of the cylinder head; In step S1, when inspecting the cylinder head of a new energy vehicle, it is first necessary to comprehensively collect image data of the cylinder head of the new energy vehicle. The cylinder head image data not only covers the color information of the cylinder head surface, but also includes the texture information of the cylinder head surface. At the same time, the edge features of the cylinder head are extracted to ensure the integrity and accuracy of the image data, and provide basic data support for subsequent inspection steps. Among them, when collecting the cylinder head image data of the new energy vehicle, it includes global collection and local collection. The global collection is used to obtain the overall image information of the cylinder head, and its purpose is to ensure that the overall structure and shape of the cylinder head can be fully recorded. The local collection is used to obtain detailed image information of key parts of the cylinder head, such as the bolt holes, sealing surfaces and other vulnerable parts of the cylinder head, so as to perform more detailed inspection and analysis of these key parts. The combination of global collection and local collection can greatly improve the accuracy and comprehensiveness of cylinder head inspection and provide more reliable data support for subsequent inspection steps.

[0021] S2, performing defect recognition on the cylinder head according to the cylinder head image data and the preset standard image data, and determining the detection state of the cylinder head according to the result of the defect recognition, wherein the detection state includes a normal state and a defect state; In step S2, after the cylinder head image data is acquired, a comparison and analysis is performed based on the acquired cylinder head image data and the preset standard image data, so as to realize the defect recognition of the cylinder head, and determine the detection state of the cylinder head according to the result of defect recognition. The detection state here is divided into two categories: normal state and defect state. The cylinder head in the normal state means that all its parameters meet the preset standards, and no repair or replacement is required, and it can be put into normal use. The cylinder head in the defect state means that there are some problems that do not meet the preset standards, and further detection and analysis are required. Defect recognition includes edge recognition and surface recognition. Edge recognition is used to detect edge defects of the cylinder head. Surface recognition is used to detect surface defects of the cylinder head, mainly including scratches, cracks, rust, etc. Edge recognition is used to detect the edge integrity of the cylinder head, such as whether there are gaps, deformations, etc. Among them, the priority of edge recognition is higher than that of surface recognition. Its purpose is to ensure the structural integrity and safety of the cylinder head, because edge defects often directly affect the sealing performance and overall strength of the cylinder head. Once a gap or deformation occurs on the edge, it may cause serious problems such as leakage or rupture of the cylinder head during use, thereby posing a threat to the performance and safety of the vehicle.

[0022] Secondly, the step of identifying defects of the cylinder head according to the cylinder head image data and the preset standard image data includes: Acquire cylinder head image data and standard image data, and perform edge curve extraction to obtain a curve to be calibrated corresponding to the cylinder head image data and a standard curve corresponding to the standard image data; Compare the curve to be calibrated with the standard curve to determine whether the curve to be calibrated coincides with the standard curve; If the curve to be calibrated coincides with the standard curve, it indicates that the corresponding cylinder head is qualified, and the surface identification of the cylinder head continues to be performed, and after the cylinder head passes the surface identification, its corresponding detection state is recorded as a normal state; If the curve to be calibrated does not coincide with the standard curve, it indicates that the corresponding cylinder head is unqualified, and the detection status of the corresponding cylinder head is directly recorded as a defective status; Specifically, when performing defect identification based on the cylinder head image data and the preset standard image data, first obtain the cylinder head image data to be detected and the preset standard image data, then perform edge curve extraction processing on the cylinder head image data and the standard image data, so as to obtain the curve to be proofread corresponding to the cylinder head image data and the standard curve corresponding to the standard image data, then perform detailed comparison processing on the extracted curve to be proofread and the standard curve, and judge whether the curve to be proofread completely overlaps with the standard curve through comparison analysis, if the curve to be proofread completely overlaps with the standard curve, it can be determined that the edge quality of the cylinder head meets the standard and is a qualified product, at this time, the cylinder head will continue to be subjected to surface recognition inspection to ensure that there are no other defects on its surface, after the cylinder head successfully passes the surface recognition inspection, its corresponding inspection status will be recorded as a normal status, on the contrary, if the curve to be proofread does not overlap with the standard curve, it indicates that the cylinder head is defective and does not meet the quality standard, at this time, the inspection status of the cylinder head will be directly recorded as a defective status for subsequent corresponding processing.

[0023] Secondly, when performing surface recognition on the cylinder head, the cylinder head image data is converted into a color space consistent with the standard image data, and the cylinder head image data after the color space conversion is subjected to noise processing to obtain an image to be proofread; Vectorization is performed on pixel values ​​in the image to be proofread and the standard image data to obtain a vector to be proofread and a standard vector; Obtaining a comparison function, and inputting the vector to be checked and the standard vector into the comparison function together, and recording the output result of the comparison function as a comparison score; Obtain an evaluation interval, and compare the comparison score with the evaluation interval; If the comparison score belongs to the evaluation interval, it indicates that the defects on the cylinder head surface are within the allowable range, and the detection status of the corresponding cylinder head is output as a normal state; If the comparison score does not fall within the evaluation interval, it indicates that there are obvious defects on the surface of the cylinder head, and the detection status of the corresponding cylinder head is output as a defective state; Specifically, in the process of surface recognition of the cylinder head, it is first necessary to convert the acquired cylinder head image data into a color space consistent with the standard image data to ensure the accuracy and consistency of subsequent processing. After the color space conversion is completed, the converted cylinder head image data is subjected to necessary noise processing to eliminate interference factors in the image, thereby obtaining a clear and clean image to be proofread. Then, each pixel value in the image to be proofread and the standard image data is vectorized, with the purpose of converting the image data into a vector form that can be numerically calculated, thereby obtaining a vector to be proofread and a standard vector. Then, a pre-equipped comparison function is introduced, and the obtained vector to be proofread and the standard vector are input into the comparison function together. Through the calculation and processing of the comparison function, a comparison result is output, and this result is recorded as a comparison score, wherein the expression of the comparison function is: , where represents the comparison score, Indicates the number of feature points in the vector to be compared and the standard vector. and They represent the vector to be compared and the standard vector respectively, and then a preset evaluation interval will be obtained, and the obtained comparison score will be compared with this evaluation interval for analysis. If the comparison score falls within the evaluation interval, it means that the degree of defect on the cylinder head surface is within the preset allowable range, and it can be considered that the surface quality of the cylinder head meets the requirements. At this time, the detection status of the corresponding cylinder head is output as a normal state. On the contrary, if the comparison score does not fall within the evaluation interval, it means that there are obvious defects on the cylinder head surface, which exceeds the allowable range. At this time, the detection status of the corresponding cylinder head is output as a defective state for further inspection and processing.

[0024] In addition, after the detection status of the cylinder head is output as a normal state, the verification process is performed synchronously, and the specific process is as follows: Counting the pixel coordinates of the pixel point deviation positions between the image to be proofread and the standard image data, as well as the pixel deviation values ​​at each pixel coordinate; Dividing the deviation area between the image to be verified and the standard image according to the pixel coordinates, and counting the number of the deviation areas, and recording them as the first verification condition parameter; Collecting the average value of the pixel deviation values ​​in each deviation area and recording it as the second verification condition parameter; Obtaining a verification function, and inputting the first verification condition parameter and the second verification condition parameter into the verification function, and recording an output result of the verification function as a verification score; Obtaining a verification threshold, and comparing the verification threshold with the verification score; When the verification score is higher than the verification threshold, it indicates that the corresponding cylinder head has failed the verification, and the detection state of the corresponding cylinder head is converted from a normal state to a defective state; When the verification score is lower than or equal to the verification threshold, it indicates that the corresponding cylinder head has passed the verification, and the detection state of the corresponding cylinder head is kept in a normal state; In the above, after the detection status of the cylinder head is determined by the system and output as a normal state, it is necessary to perform corresponding verification processing on it synchronously to ensure the accuracy and reliability of the detection results. First, the image to be verified will be compared and analyzed with the pre-set standard image data, and the specific pixel coordinates at the pixel point deviation position between the two will be counted. At the same time, the pixel deviation value at each pixel coordinate point will be recorded one by one for subsequent analysis and processing. Then, based on the pixel coordinates that have been counted, the specific deviation area between the image to be verified and the standard image is further divided. This can be achieved by drawing a deviation distribution map, so that the position and size of each deviation area can be clearly displayed. And distribution, then count the total number of deviation areas, and record this number as the first verification condition parameter as an indicator for evaluating the degree of deviation of the cylinder head image, and then calculate the average value of the pixel deviation values ​​in each divided deviation area, and record the average value calculation result as the second verification condition parameter, so as to form a complete verification condition together with the first verification condition parameter, and then call the preset verification function, and input the recorded first verification condition parameter and second verification condition parameter into the verification function for comprehensive calculation, and finally record the output result of the verification function in detail as the verification score as a prerequisite indicator for judging whether the cylinder head is qualified, wherein the expression of the verification function is: , where represents the verification score, and represent the weight factors of the first verification condition parameter and the second verification condition parameter respectively, and They respectively represent the first verification condition parameter and the second verification condition parameter, and then the preset verification threshold will be obtained, and the verification threshold will be compared and analyzed with the verification score. When the verification score is higher than the preset verification threshold, it will be determined that the corresponding cylinder head has not passed the verification. At this time, the detection status of the corresponding cylinder head will be converted from the original normal state to a defective state for subsequent further processing and repair. On the contrary, when the verification score is lower than or equal to the preset verification threshold, it will be determined that the corresponding cylinder head has successfully passed the verification. At this time, the detection status of the corresponding cylinder head will be maintained in a normal state to determine that it meets the quality standards.

[0025] S3, extracting the image features of all cylinder heads in normal state, establishing an image feature database of cylinder heads in normal state, and outputting the allowable deviation between each cylinder head and the standard image data according to the image feature database; In step S3, for all cylinder heads determined to be in normal state, their image features are systematically extracted, and based on this, a complete image feature database of the normal state of the cylinder heads is established. At the same time, the allowable deviation between each cylinder head and the standard image data is determined based on the image feature database to provide data support for subsequent deviation type determination. The step of outputting the allowable deviation between each cylinder head and the standard image data based on the image feature database includes: Based on the current detection node, reversely construct the monitoring period; A plurality of sampling nodes are set within the monitoring period, wherein the sampling nodes correspond to the time nodes at which the cylinder head passes the verification process; Extracting the second verification condition parameter corresponding to each cylinder head image data at each sampling node from the image feature database, and recording it as the allowable deviation between the cylinder head and the standard image data; Specifically, when determining the allowable deviation between the cylinder head image data and the standard image data, first, using the current detection node as a reference point, trace back and construct a complete monitoring period to ensure that the monitoring period can cover all time periods that need to be analyzed, and then set multiple sampling nodes within the constructed monitoring period. The selection of sampling nodes needs to strictly correspond to the specific time nodes at which each cylinder head passes the verification process to ensure the validity of data collection, and then extract the second verification condition parameters corresponding to each cylinder head image data under each sampling node from the image feature database one by one, and record the extracted second verification condition parameters as the allowable deviation between the cylinder head and the standard image data.

[0026] S4. Determine the deviation type of the cylinder head according to the allowable deviation, and determine the improvement position of the cylinder head production process according to the deviation type; In step S4, after the allowable deviation is determined, the deviation type of the cylinder head is determined according to the obtained allowable deviation. The deviation type includes risk deviation and normal deviation. The risk deviation corresponds to the risk area, and the normal deviation corresponds to the normal area. The specific position of the cylinder head that needs to be improved in the production process is located according to the deviation type, so as to optimize the production process and avoid defects in the cylinder heads produced subsequently. The steps of determining the deviation type of the cylinder head according to the allowable deviation and determining the improvement position of the cylinder head production process according to the deviation type include: Obtain the allowable deviation and arrange the allowable deviation according to the inspection sequence of the cylinder head; Obtaining a condition function, inputting the allowable deviation into the condition function, and recording the output result of the condition function as an evaluation condition parameter; Randomly select multiple groups of adjacent allowable deviations; Obtaining an evaluation function, and inputting the allowable deviation and the evaluation condition parameters into the evaluation function, and recording the output result of the evaluation function as a classification parameter; Obtain a classification threshold, and compare the classification threshold with the classification parameter. When the classification parameter is greater than the classification threshold, the deviation area corresponding to the allowable deviation is recorded as a risk area, and the production link corresponding to the risk area is recorded as a risk link. Otherwise, the deviation area corresponding to the allowable deviation is recorded as a normal area, and the production link corresponding to the normal area is recorded as a normal link. Specifically, when determining the deviation type of the cylinder head based on the allowable deviation, it is first necessary to obtain the allowable deviation of the cylinder head and arrange the allowable deviation in order according to the timing of the cylinder head during the detection process, and then obtain the conditional function for processing these data, and input the arranged allowable deviation data into the conditional function, and record the output result after the conditional function is processed as the conditional parameter for subsequent evaluation. The expression of the conditional function is: , where represents the evaluation condition parameter, Indicates the amount of allowable deviation, and Represents the allowable deviation under adjacent acquisition nodes, and then randomly selects multiple groups of adjacent allowable deviation data, and combines the evaluation function to calculate the classification parameters used to distinguish the cylinder head deviation types. The expression of the evaluation function is: , where represents the classification parameter, The number of groups that represent the allowed deviation, The group number indicating the allowable deviation, Indicates the allowable deviation of the latter position in the same group. It represents the allowable deviation that ranks high in the same group. Finally, the preset classification threshold is obtained, and the classification threshold is compared with the classification parameter one by one. When the classification parameter is greater than the classification threshold, the deviation area indicated by the corresponding allowable deviation is marked as a risk area, and the production link corresponding to the risk area is recorded as a risk link. On the contrary, if the classification parameter is less than or equal to the classification threshold, the deviation area indicated by the corresponding allowable deviation is marked as a normal area, and the production link corresponding to the normal area is recorded as a normal link. Through this series of steps, potential problems in cylinder head production can be effectively identified and located, so as to make targeted improvements.

[0027] S5. Extract the defect area and defect type of the cylinder head in the defect state, determine the repair level of the cylinder head according to the defect area and defect type, match the corresponding repair solutions one by one according to the repair level of the cylinder head, and upload them to the management end synchronously; In step S5, for the cylinder head determined to be in a defective state, its defect area and specific defect type (such as cracks, wear, corrosion, etc.) will be extracted in detail, and the repair level of the cylinder head will be determined based on the defect area and defect type. Then, based on the repair level of the cylinder head (such as the repair level of the crack type is higher than the repair level of the wear or corrosion type), the most appropriate repair plan is matched one by one to ensure the repair effect. At the same time, all relevant information is synchronously uploaded to the management end to facilitate subsequent data analysis and quality management.

[0028] See also Figure 2 A cylinder head detection system for a new energy vehicle, using the above-mentioned cylinder head detection method for a new energy vehicle, comprises: A data acquisition module, the data acquisition module is used to collect cylinder head image data of new energy vehicles, wherein the cylinder head image data includes cylinder head surface feature information and edge feature information; A defect recognition module, the defect recognition module is used to recognize defects of the cylinder head according to the cylinder head image data and the preset standard image data, and determine the detection state of the cylinder head according to the result of the defect recognition, wherein the detection state includes a normal state and a defect state; The deviation degree measurement module is used to extract the image features of all cylinder heads in normal state, establish an image feature database of the cylinder heads in normal state, and output the allowable deviation between each cylinder head and the standard image data according to the image feature database; Deviation traceability module: The deviation traceability module is used to determine the deviation type of the cylinder head according to the allowable deviation, and determine the improvement position of the cylinder head production link according to the deviation type; Defect output module: The defect output module is used to extract the defect area and defect type of the cylinder head in the defective state, and determine the repair level of the cylinder head based on the defect area and defect type. Then, the corresponding repair plan is matched one by one according to the repair level of the cylinder head, and uploaded to the management end synchronously.

[0029] Specifically, the detection system includes a data acquisition module, a defect recognition module, a deviation measurement module, a deviation tracing module and a defect output module. The main function of the data acquisition module is to accurately collect the cylinder head image data of new energy vehicles. Specifically, the cylinder head image data not only covers various feature information on the cylinder head surface, such as texture, color and other details, but also includes the contour, lines and other feature information of the cylinder head edge to ensure the comprehensiveness and accuracy of the data. The role of the defect recognition module is to compare and analyze the collected cylinder head image data with the pre-set standard image data, so as to achieve accurate identification of cylinder head defects. On this basis, according to the results of defect recognition, the system will determine the detection status of the cylinder head, which is specifically divided into a normal state and a defective state. The main task of the deviation measurement module is to extract from all cylinder head images that are judged to be in a normal state The key image features are identified, and based on this, an image feature database of the normal state of the cylinder head is established. Using this database, the system can calculate the allowable deviation between each cylinder head and the standard image data, and provide data support for the subsequent deviation tracing. The function of the deviation tracing module is to further determine the specific deviation type of the cylinder head based on the calculated allowable deviation. On this basis, the system will accurately locate the specific position that needs to be improved in the cylinder head production link according to the deviation type, thereby providing a strong basis for the optimization of the production process. The defect output module is responsible for extracting the defective area and specific defect type of the cylinder head in a defective state. Based on this information, the system will determine the repair level of the cylinder head, and match the corresponding repair plan one by one according to different repair levels. Finally, these repair plans will be uploaded to the management end synchronously for subsequent repair work and data management.

[0030] See also Figure 3 , an electronic device, the electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; Among them, the memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the above-mentioned new energy vehicle cylinder head detection method.

[0031] The processor of the above-mentioned electronic device can be a high-performance computing unit such as a central processing unit (CPU), a graphics processing unit (GPU) or a field programmable gate array (FPGA) to ensure that the complex new energy vehicle cylinder head detection algorithm can be efficiently run. The memory can be a high-speed random access memory (RAM) for temporarily storing the data and instructions required by the processor during execution. At the same time, it can also be equipped with a large-capacity non-volatile memory, such as a solid-state drive (SSD) or a mechanical hard disk (HDD), for long-term storage of detection data, standard image data, historical detection results and other information. In addition, the electronic device can also include an arithmetic unit, an input device and an output device. The arithmetic unit is used to perform various arithmetic and logical operations to assist the processor in completing complex computing tasks. Input devices such as keyboards and mice are convenient for users to operate and input commands. Output devices such as display screens and printers are used to display detection results and print reports.

[0032] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0033] The above is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art unless otherwise specified and limited.

Claims

1. A new energy vehicle cylinder head detection method, characterized in that: include: Collecting cylinder head image data of a new energy vehicle, wherein the cylinder head image data includes cylinder head surface feature information and edge feature information; Performing defect recognition on the cylinder head according to the cylinder head image data and preset standard image data, and determining the detection state of the cylinder head according to the result of the defect recognition, wherein the detection state includes a normal state and a defect state; Extracting the image features of all cylinder heads in the normal state, establishing an image feature database of the cylinder heads in the normal state, and outputting the allowable deviation between each cylinder head and the standard image data according to the image feature database; Determining the deviation type of the cylinder head according to the allowable deviation, and determining the improvement position of the cylinder head production process according to the deviation type; The defect area and defect type of the cylinder head in the defect state are extracted, and the repair level of the cylinder head is determined according to the defect area and defect type. Then, corresponding repair plans are matched one by one according to the repair level of the cylinder head, and uploaded to the management end synchronously.

2. A cylinder head detection method for new energy vehicles according to claim 1, characterized in that: The acquisition of cylinder head image data of the new energy vehicle includes global acquisition and local acquisition. The global acquisition is used to obtain the overall image information of the cylinder head, and the local acquisition is used to obtain the detailed image information of the key parts of the cylinder head.

3. A cylinder head detection method for a new energy vehicle according to claim 1, characterized in that: The defect recognition includes edge recognition and surface recognition, wherein the surface recognition is used to detect surface defects of the cylinder head, and the edge recognition is used to detect edge defects of the cylinder head; The priority of edge recognition is higher than that of surface recognition.

4. A cylinder head detection method for new energy vehicles according to claim 1, characterized in that: The step of identifying defects of the cylinder head according to the cylinder head image data and preset standard image data comprises: Acquire cylinder head image data and standard image data, and perform edge curve extraction to obtain a curve to be calibrated corresponding to the cylinder head image data and a standard curve corresponding to the standard image data; Comparing the curve to be calibrated with the standard curve to determine whether the curve to be calibrated coincides with the standard curve; If the curve to be calibrated coincides with the standard curve, it indicates that the corresponding cylinder head is qualified, and the surface recognition of the cylinder head continues to be performed, and after the cylinder head passes the surface recognition, its corresponding detection state is recorded as a normal state; If the curve to be calibrated does not coincide with the standard curve, it indicates that the corresponding cylinder head is unqualified, and the detection status of the corresponding cylinder head is directly recorded as a defective status.

5. A cylinder head detection method for new energy vehicles according to claim 4, characterized in that: When performing surface recognition on the cylinder head, the cylinder head image data is converted into a color space consistent with the standard image data, and noise processing is performed on the cylinder head image data after the color space conversion to obtain an image to be proofread; Vectorizing the pixel values ​​in the image to be proofread and the standard image data to obtain a vector to be proofread and a standard vector; Obtaining a comparison function, and inputting the vector to be checked and the standard vector into the comparison function together, and recording the output result of the comparison function as a comparison score; Obtaining an evaluation interval, and comparing the comparison score with the evaluation interval; If the comparison score belongs to the evaluation interval, it indicates that the defect of the cylinder head surface is within the allowable range, and the detection state of the corresponding cylinder head is output as a normal state; If the comparison score does not belong to the evaluation interval, it indicates that there are obvious defects on the surface of the cylinder head, and the detection state of the corresponding cylinder head is output as a defective state.

6. A cylinder head detection method for a new energy vehicle according to claim 5, characterized in that: After the detection state of the cylinder head is output as a normal state, a verification process is performed synchronously, and the specific process is as follows: Counting the pixel coordinates of the pixel point deviation positions between the image to be proofread and the standard image data, and the pixel deviation values ​​at each pixel coordinate; Dividing the deviation area between the image to be proofread and the standard image according to the pixel coordinates, and counting the number of the deviation areas and recording them as the first verification condition parameter; Collecting an average value of the pixel deviation values ​​in each of the deviation areas and recording it as a second verification condition parameter; Obtaining a verification function, and inputting the first verification condition parameter and the second verification condition parameter into the verification function, and recording an output result of the verification function as a verification score; Obtaining a verification threshold, and comparing the verification threshold with a verification score; When the verification score is higher than the verification threshold, it indicates that the corresponding cylinder head has failed the verification, and the detection state of the corresponding cylinder head is converted from a normal state to a defective state; When the verification score is lower than or equal to the verification threshold, it indicates that the corresponding cylinder head has passed the verification, and the detection state of the corresponding cylinder head is maintained in a normal state.

7. A cylinder head detection method for new energy vehicles according to claim 6, characterized in that: The step of outputting the allowable deviation between each cylinder head and the standard image data according to the image feature database comprises: Based on the current detection node, reversely construct the monitoring period; Setting a plurality of sampling nodes within the monitoring period, wherein the sampling nodes correspond to time nodes at which the cylinder head passes the verification process; The second verification condition parameter corresponding to each cylinder head image data at each sampling node is extracted from the image feature database and recorded as the allowable deviation between the cylinder head and the standard image data.

8. The cylinder head detection method for a new energy vehicle according to claim 1, characterized in that: The step of determining the deviation type of the cylinder head according to the allowable deviation, and determining the improvement position of the cylinder head production link according to the deviation type, comprises: Acquiring the allowable deviations and arranging the allowable deviations according to the detection sequence of the cylinder head; Obtaining a condition function, inputting the allowable deviation into the condition function, and recording an output result of the condition function as an evaluation condition parameter; Randomly select multiple groups of adjacent allowable deviations; Obtaining an evaluation function, and inputting the allowable deviation and the evaluation condition parameter into the evaluation function, and recording the output result of the evaluation function as a classification parameter; A classification threshold is obtained, and the classification threshold is compared with a classification parameter. When the classification parameter is greater than the classification threshold, the deviation area corresponding to the allowable deviation is recorded as a risk area, and the production link corresponding to the risk area is recorded as a risk link. Otherwise, the deviation area corresponding to the allowable deviation is recorded as a normal area, and the production link corresponding to the normal area is recorded as a normal link.

9. A cylinder head detection system for new energy vehicles, characterized in that: The cylinder head detection method for a new energy vehicle according to any one of claims 1 to 8 comprises: A data acquisition module, the data acquisition module is used to collect cylinder head image data of a new energy vehicle, wherein the cylinder head image data includes cylinder head surface feature information and edge feature information; a defect recognition module, the defect recognition module being used to perform defect recognition on the cylinder head according to the cylinder head image data and preset standard image data, and determine a detection state of the cylinder head according to a result of the defect recognition, wherein the detection state includes a normal state and a defect state; a deviation degree measuring module, the deviation degree measuring module is used to extract the image features of all cylinder heads in the normal state, establish an image feature database of the cylinder heads in the normal state, and output the allowable deviation between each cylinder head and the standard image data according to the image feature database; A deviation tracing module, the deviation tracing module is used to determine the deviation type of the cylinder head according to the allowable deviation, and determine the improvement position of the cylinder head production link according to the deviation type; A defect output module is used to extract the defect area and defect type of the cylinder head in the defect state, and determine the repair level of the cylinder head according to the defect area and defect type, and then match the corresponding repair solutions one by one according to the repair level of the cylinder head, and upload them to the management end synchronously.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the new energy vehicle cylinder head detection method described in any one of claims 1 to 8.

Images