Method and system for identifying electromagnetic fingerprint characteristics of automobile parts

Abstract

The invention discloses a method and a system for identifying electromagnetic fingerprint features of automobile parts. The method comprises the following steps: determining identification tags of the automobile parts based on traversal of identification areas of the automobile parts; the electromagnetic fingerprint features of the automobile parts are determined according to the identification tags of the automobile parts, and accurate control over the electromagnetic fingerprint features of the automobile parts is achieved. Furthermore, the identification of the electromagnetic fingerprint features is triggered according to the identification mode, the corresponding identification signal and the signal attenuation degree are output, and the authenticity of the automobile part is determined according to the identification signal and the signal attenuation degree, so that the overall consideration of the electromagnetic fingerprint features, the automobile part and identification influence factors is compatible; according to the invention, the accuracy of the identification mode is ensured, the identification signal and the signal attenuation degree are introduced, the multi-dimensional control of the identification signal and the signal attenuation degree is realized, the authenticity identification effect of the automobile part is ensured, and the identification precision of the electromagnetic fingerprint feature is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile parts, and in particular to a method and system for identifying electromagnetic fingerprint features of automobile parts. Background Art

[0002] With the development of science and technology, auto parts are gradually used in people's lives. As a part of the car, in the production of auto parts, it is necessary to mark the auto parts and realize online tracking of auto parts. In the existing technology, auto parts are dynamically identified. However, in the dynamic identification of auto parts, the electromagnetic fingerprint characteristics of auto parts cannot be accurately controlled, which affects the recognition accuracy of the electromagnetic fingerprint characteristics. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a method and system for identifying electromagnetic fingerprint features of automobile parts.

[0004] The embodiment of the present invention provides a method for identifying electromagnetic fingerprint features of automobile parts, which is applied to the identification scenario of electromagnetic fingerprint features of automobile parts; The method for identifying electromagnetic fingerprint features of automobile parts comprises: Collect the location of auto parts; Trigger corresponding positioning detection according to the location of the automobile parts, and collect the identification area of ​​the automobile parts according to the positioning detection of the automobile parts; Determining an identification tag of the automobile part based on traversal of the identification area of ​​the automobile part; Determine the electromagnetic fingerprint characteristics of the automobile parts according to the identification tags of the automobile parts; Match the corresponding recognition mode based on electromagnetic fingerprint features, automobile parts and recognition influencing factors; The recognition of the electromagnetic fingerprint feature is triggered according to the recognition mode, and the corresponding recognition signal and signal attenuation degree are output, and the authenticity of the automobile parts is determined according to the recognition signal and the signal attenuation degree.

[0005] In addition, an embodiment of the present invention further provides a system for identifying electromagnetic fingerprint features of automobile parts, the system for identifying electromagnetic fingerprint features of automobile parts comprising: A collection module is used to collect the locations of automobile parts; An identification area module is used to trigger corresponding positioning detection according to the location of the automobile parts, and collect the identification area of ​​the automobile parts according to the positioning detection of the automobile parts; An identification tag module, used to determine the identification tag of the automobile part based on the traversal of the identification area of ​​the automobile part; An electromagnetic fingerprint feature module, used to determine the electromagnetic fingerprint features of automobile parts according to identification tags of automobile parts; Recognition mode module, used to match corresponding recognition modes based on electromagnetic fingerprint features, automobile parts and recognition influencing factors; The recognition module is used to trigger the recognition of electromagnetic fingerprint features according to the recognition mode, and output the corresponding recognition signal and signal attenuation degree, and determine the authenticity of the automobile parts according to the recognition signal and signal attenuation degree.

[0006] In an embodiment of the present invention, the location of automobile parts is collected through the method in the embodiment of the present invention; the corresponding positioning detection is triggered according to the location of the automobile parts, and the identification area of ​​the automobile parts is collected according to the positioning detection of the automobile parts; the identification tag of the automobile part is determined based on the traversal of the identification area of ​​the automobile part; the electromagnetic fingerprint feature of the automobile part is determined according to the identification tag of the automobile part, thereby realizing the precise control of the electromagnetic fingerprint feature of the automobile part.

[0007] Furthermore, a corresponding recognition pattern is matched based on electromagnetic fingerprint features, automobile parts and identification influencing factors; recognition of the electromagnetic fingerprint features is triggered according to the recognition pattern, and a corresponding recognition signal and a signal attenuation degree are output, and the authenticity of the automobile parts is determined according to the recognition signal and the signal attenuation degree, which is compatible with the overall consideration of the electromagnetic fingerprint features, automobile parts and identification influencing factors, ensuring the accuracy of the recognition pattern, and then introducing the recognition signal and the signal attenuation degree, realizing multi-dimensional control of the recognition signal and the signal attenuation degree, ensuring the authenticity of automobile parts recognition, and improving the recognition accuracy of the electromagnetic fingerprint features. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0009] Figure 1 It is a schematic flow chart of a method for identifying electromagnetic fingerprint features of automobile parts in an embodiment of the present invention; Figure 2 It is a schematic diagram of the structure of the electromagnetic fingerprint feature recognition system of automobile parts in an embodiment of the present invention; Figure 3 The figure is a hardware diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0011] See also Figures 1 to 3 , a method for identifying electromagnetic fingerprint features of automobile parts, applied to the identification scenario of electromagnetic fingerprint features of automobile parts; the method for identifying electromagnetic fingerprint features of automobile parts includes: Step S11: collecting the locations of automobile parts; Step S12: triggering corresponding positioning detection according to the location of the automobile parts, and collecting the identification area of ​​the automobile parts according to the positioning detection of the automobile parts; Step S13: determining the identification tag of the automobile part based on the traversal of the identification area of ​​the automobile part; Step S14: determining the electromagnetic fingerprint characteristics of the automobile parts according to the identification tags of the automobile parts; Step S15: matching corresponding recognition patterns based on electromagnetic fingerprint features, automobile parts and recognition influencing factors; Step S16: triggering the recognition of the electromagnetic fingerprint feature according to the recognition mode, and outputting the corresponding recognition signal and the signal attenuation degree, and determining the authenticity of the automobile parts according to the recognition signal and the signal attenuation degree.

[0012] In step S11, the location of automobile parts is collected; In the specific implementation process of the present invention, the specific steps may be: S111: Collect production signals of automobile parts; S112: Determine corresponding position information according to analysis of production signals of automobile parts; S113: Determine the space where the automobile parts are located based on the interpretation of the position information; S114: Determine the location of the automobile parts according to the autonomous detection of the space where the automobile parts are located, so as to collect the location of the automobile parts.

[0013] In the embodiments of the present application, production signals of automobile parts are collected, the production signals of automobile parts are introduced, and the production signals of automobile parts are controlled, thereby performing targeted management and control on the production signals of automobile parts.

[0014] At this time, the corresponding position information is determined according to the analysis of the production signals of the automobile parts, the analysis of the production signals of the automobile parts is realized, the corresponding position information is introduced, and further interpretation of the position information is realized, so that the space where the automobile parts are located is determined based on the interpretation of the position information, thereby ensuring the accuracy of the space where the automobile parts are located.

[0015] Optionally, the collected production signals usually need to be parsed to extract useful information. In the production process of automobile parts, the purpose of parsing production signals is to determine the location information of automobile parts. At this time, the collected production signals (such as RFID tag information) are compared with the information in the database to obtain the storage location or production line location of the automobile parts, and the collected signals are analyzed using algorithms to infer the location of the automobile parts. For example, a preset machine learning algorithm can be used to train sensor data to predict the location of automobile parts.

[0016] At the same time, if the location information is given in the form of coordinates, the system needs to convert it into a specific physical space description, such as "Warehouse No. X", "Position Z on Line Y", etc.; there are preset description information of multiple storage areas or production line segments, and by matching the parsed location information with these preset information, the specific space where the auto parts are located can be determined. Optionally, assume that in the production information management system of an auto parts manufacturing enterprise, each part has a unique serial number, and the serial number is associated with the specific location of the part in the warehouse or production line. When the system parses the serial number of a part, it will first query the preset location information corresponding to the serial number. For example, the serial number "ABC123" corresponds to the storage location of "Warehouse No. 2, Area A, Row 5, Column 3". By interpreting this location information, the system can determine the specific space where the part is located. Therefore, the location of the auto parts is determined based on the autonomous detection of the space where the auto parts are located, so as to collect the location of the auto parts, introduce the autonomous detection of the space where the auto parts are located, realize the refined control of the autonomous detection of the space where the auto parts are located, and ensure the authenticity of the auto parts.

[0017] At this time, on the basis of determining the space where the auto parts are located, autonomous detection is carried out to further accurately determine the location of the auto parts. Optionally, necessary autonomous detection equipment, such as RFID readers, cameras, sensors, etc., are installed in the space where the auto parts are located. These devices are configured to ensure that they can accurately read or sense the information of the auto parts; data from the autonomous detection equipment is collected in real time, including RFID tag information, images captured by the camera, physical parameters read by the sensor, etc., and the collected data is processed and analyzed to extract useful location information. For example, the position of the parts in the camera's field of view is identified through an image recognition algorithm, or the movement trajectory of the parts is inferred through the reading time and position information of the RFID reader. When the position of the parts changes, the system should update its position information in a timely manner to ensure the accuracy and real-time nature of the data.

[0018] In step S12, a corresponding positioning detection is triggered according to the location of the automobile parts, and the identification area of ​​the automobile parts is collected according to the positioning detection of the automobile parts; In the specific implementation process of the present invention, the specific steps may be: S121: freeze the location of automobile parts; S122: triggering corresponding positioning detection based on the location of the automobile parts; S123: Real-time monitoring of positioning detection of automobile parts; S124: collecting multiple area parameters according to the positioning detection of automobile parts; S125: Associating multiple area parameters and automotive parts; S126: Determine the recognition area of ​​the automobile parts based on the multiple area parameters and the automobile parts to collect the recognition area of ​​the automobile parts.

[0019] In an embodiment of the present application, the position of automobile parts is frozen, and the position of automobile parts is further processed, so as to trigger corresponding positioning detection based on the position of automobile parts, thereby realizing positioning detection of automobile parts and further controlling automobile parts.

[0020] Therefore, the positioning detection of automobile parts is monitored in real time; multiple regional parameters are collected according to the positioning detection of automobile parts, multiple regional parameters are introduced, and multiple regional parameters are interacted with each other to ensure the dynamic interaction effect of multiple regional parameters.

[0021] At this time, the location information from different autonomous detection equipment is collected, and the data is cleaned and integrated to eliminate errors and redundancies. Based on the integrated location information, the system determines the precise location coordinates of the automotive parts, which is a three-dimensional space coordinate and also a regional identification code. Once the location coordinates are determined, the system locks them to prevent subsequent location changes caused by data updates or equipment errors. Once the position is fixed, the system immediately triggers the corresponding positioning detection operation.

[0022] At the same time, real-time data from positioning detection equipment is continuously collected, and the collected data is verified to ensure the accuracy and completeness of the data. If the system detects any abnormal data or behavior, it will immediately trigger the alarm mechanism and interrupt the detection process. Furthermore, temperature, humidity, and shelf load-bearing capacity are selected as regional parameters that need to be collected. Temperature sensors, humidity sensors, and weight sensors are installed in the area where the parts are located. The values ​​of these parameters are monitored in real time, and the collected data are recorded in the database and analyzed. For example, when the temperature or humidity is found to be beyond the preset range, the system automatically triggers the alarm mechanism. After receiving the alarm, the relevant personnel immediately take measures, such as adjusting the temperature and humidity in the warehouse, reinforcing the shelves, etc., to ensure the safe storage of parts. . Furthermore, multiple regional parameters and automobile parts are associated; the identification area of ​​the automobile parts is determined based on the multiple regional parameters and automobile parts, so as to collect the identification area of ​​the automobile parts, introduce multiple regional parameters and automobile parts, and perform overall management and control of the multiple regional parameters and automobile parts, thereby ensuring multi-dimensional control of the multiple regional parameters and automobile parts and improving the accuracy of the identification area of ​​the automobile parts.

[0023] At this point, the previously collected multiple regional parameters (such as temperature, humidity, shelf load-bearing capacity, space size, etc.) are associated with specific auto parts. This association enables the system to understand the specific environmental conditions of each component, providing data support for subsequent analysis, management and optimization; at the same time, multiple regional parameters and auto parts are introduced, and auto parts contain information such as shape, surface characteristics, materials and performance; combined with regional parameters and auto parts characteristics, a comprehensive evaluation is conducted. Considering factors such as lighting conditions, temperature and humidity, background environment, spatial layout, personnel flow and equipment movement, the optimal recognition area is selected. Experimental verification is carried out in the initially determined recognition area. Use recognition equipment to collect images and perform preliminary image processing and recognition tests. Adjust the size, position and lighting scheme of the recognition area according to the test results.

[0024] Optionally, the gear has a specific shape and size, including parameters such as the number of teeth, module, and pressure angle. These parameters are critical to the recognition process, so it is necessary to ensure that the recognition area can fully capture the image of the gear; the surface features of the gear include tooth surface roughness, coating, and markings. These features vary due to different treatment methods and wear levels, so it is necessary to select a recognition area that can clearly show these features; at this time, a comprehensive evaluation is performed in combination with parameters such as lighting conditions, temperature and humidity in the warehouse, and the shape and size of the gear. An area with uniform lighting, stable temperature and humidity, and a simple background is selected as the preliminary recognition area; within the preliminarily determined recognition area, a high-resolution industrial camera is used to photograph the gear. The collected image is preprocessed and feature extracted, including steps such as filtering, denoising, edge detection, and texture analysis. According to the processing results, the parameters such as the size, position, and angle of the recognition area are adjusted to ensure that the image and features of the gear can be accurately captured; based on the results of experimental verification, the recognition area is further optimized and adjusted. For example, the focal length and exposure time of the camera are adjusted to improve image quality; the position and angle of the recognition area are adjusted to reduce background interference; or additional lighting equipment is added to improve lighting uniformity.

[0025] In step S13, the identification tag of the automobile part is determined based on the traversal of the identification area of ​​the automobile part; In the specific implementation process of the present invention, the specific steps may be: S131: freeze the identification area of ​​automobile parts; S132: Determine a corresponding traversal mode based on the recognition area of ​​the automobile part and the outline of the automobile part; S133: triggering traversal of the identification area of ​​the automobile parts according to the identification area of ​​the automobile parts and the corresponding traversal mode; S134: Determine the identification tag of the automobile part based on the traversal of the identification area of ​​the automobile part, and mark the location of the identification tag of the automobile part.

[0026] In the embodiment of the present application, the recognition area of ​​the automobile parts is frozen, the recognition area of ​​the automobile parts is introduced, the recognition area of ​​the automobile parts is controlled as a whole, and the further processing of the recognition area of ​​the automobile parts is ensured.

[0027] Furthermore, the corresponding traversal mode is determined based on the recognition area of ​​automobile parts and the outer contour of automobile parts, the recognition area of ​​automobile parts and the outer contour of automobile parts are introduced, the recognition area of ​​automobile parts and the outer contour of automobile parts are considered as a whole, the multi-dimensional control of the recognition area of ​​automobile parts and the outer contour of automobile parts is realized, and the accuracy of the traversal mode is ensured.

[0028] At this point, carefully study the outline of the automotive parts to understand their structural characteristics and key features. This helps determine the complexity and directionality of the traversal pattern, and design a suitable traversal path or traversal pattern based on the size and shape of the recognition area and the outline of the parts. While ensuring full coverage of the recognition area, minimize redundancy and repeated scanning to improve traversal efficiency.

[0029] Therefore, the traversal of the identification area of ​​the automobile part is triggered according to the identification area of ​​the automobile part and the corresponding traversal mode; the identification tag of the automobile part is determined based on the traversal of the identification area of ​​the automobile part, and the position of the identification tag of the automobile part is marked, thereby realizing the traversal of the identification area of ​​the automobile part and ensuring the accuracy of the identification tag of the automobile part, so as to facilitate further processing of the identification tag of the automobile part.

[0030] At this time, according to the traversal mode, the traversal function in the image recognition or processing software is started, and the recognition area is scanned point by point or face by face according to the path or sequence of the traversal mode. This involves pixel-level traversal of the image, traversal of feature points, etc. During the traversal process, the scanned image data or feature information is recorded in real time for subsequent analysis. The feature information collected during the traversal process is matched with the features of the automotive parts pre-stored in the database. Based on the matching results, the unique identification tag that matches the current automotive part is determined, and the location of the identification tag is marked in the image, usually by drawing a border on the image, annotating text, etc.

[0031] Optionally, taking the valve assembly of a car engine as an example, the feature information collected during the traversal process is matched with the valve assembly features in the database. Through matching, we determine the unique identification label of the current valve assembly, such as "valve assembly-XX model-YY batch". Then, we mark the location of the identification label in the image, that is, draw a border around the identification area of ​​the valve assembly, and annotate the text information of the identification label in the border. In this way, we can clearly know what the car parts in the current image are, which model and batch they belong to, and other information.

[0032] In step S14, the electromagnetic fingerprint characteristics of the automobile parts are determined according to the identification tags of the automobile parts; In the specific implementation process of the present invention, the specific steps may be: S141: Identification labels for fixed-frame automotive parts; S142: Identification tags of associated automobile parts and the space where the automobile parts are located; S143: matching a corresponding label recognition pattern according to the identification label of the automobile component and the space where the automobile component is located; S144: associating identification tags of automobile parts and corresponding tag identification modes; S145: Determine the electromagnetic fingerprint feature of the automobile component according to the identification tag of the automobile component and the corresponding tag identification mode.

[0033] In an embodiment of the present application, the location of automobile parts is collected; corresponding positioning detection is triggered according to the location of the automobile parts, and the identification area of ​​the automobile parts is collected according to the positioning detection of the automobile parts; the identification tag of the automobile part is determined based on the traversal of the identification area of ​​the automobile part; the electromagnetic fingerprint feature of the automobile part is determined according to the identification tag of the automobile part, thereby realizing precise control of the electromagnetic fingerprint feature of the automobile part.

[0034] At this time, the identification tags of the auto parts are fixed, the identification tags of the auto parts are introduced, and the identification tags of the auto parts are further processed. At the same time, the identification tags of the auto parts and the space where the auto parts are located are associated, and the identification tags of the auto parts and the space where the auto parts are located are introduced, realizing the multiple interactions of the identification tags of the auto parts and the space where the auto parts are located. Therefore, according to the matching of the corresponding label recognition mode of the identification tags of the auto parts and the space where the auto parts are located, the identification tags of the auto parts and the space where the auto parts are located are introduced, realizing the dynamic matching of the identification tags of the auto parts and the space where the auto parts are located, and ensuring the accuracy of the label recognition mode.

[0035] At this time, the identification tags of auto parts and the space where auto parts are located are introduced to determine the specific location of auto parts on the vehicle or production line, including the storage area, assembly station or maintenance location of the parts. This helps to understand the readability and recognition difficulty of the label, and whether specific recognition equipment or technology is required. According to the type and spatial location of the identification label, select the appropriate label recognition mode. For example, for barcode or QR code labels, optical scanners can be used for identification; for RFID tags, RFID readers need to be used for reading. At the same time, it is also necessary to consider the interference factors of the recognition environment, such as light, temperature, humidity, etc., to ensure the accuracy and reliability of recognition. According to the selected recognition mode, set the corresponding recognition parameters, such as scanning speed, reading distance, recognition accuracy, etc. The settings of these parameters should be adjusted according to the actual situation to ensure the best recognition effect.

[0036] Furthermore, the identification tags of automobile parts and the corresponding tag identification modes are associated; the electromagnetic fingerprint characteristics of automobile parts are determined according to the identification tags of automobile parts and the corresponding tag identification modes, the identification tags of automobile parts and the corresponding tag identification modes are introduced, the overall consideration of the identification tags of automobile parts and the corresponding tag identification modes is realized, and the multi-dimensional control of the identification tags of automobile parts and the corresponding tag identification modes is further realized, thereby ensuring the accuracy of the electromagnetic fingerprint characteristics of automobile parts.

[0037] At this time, the identification tags of the automobile parts are determined based on the traversal of the identification areas of the automobile parts; the electromagnetic fingerprint features of the automobile parts are determined according to the identification tags of the automobile parts, thereby achieving accurate control of the electromagnetic fingerprint features of the automobile parts.

[0038] At the same time, the identification tags of auto parts usually contain key data such as basic product information, manufacturer, production date, batch number, etc. The tag recognition mode refers to reading the identification tag information on auto parts through specific technical means. Different tag types require different recognition technologies. For example, barcodes and QR codes are usually recognized using optical scanning technology, while radio frequency tags are read using radio frequency identification (RFID) technology.

[0039] Electromagnetic fingerprint features are the features of electromagnetic signals. These features are detectable, stable and identifiable, and can be used to uniquely identify an electromagnetic signal source (here, automotive parts). When determining the electromagnetic fingerprint features of automotive parts, it is first necessary to use the tag recognition mode to obtain the basic information of the parts. Then, the parts are tested with professional electromagnetic testing equipment to collect their electromagnetic radiation signals. Finally, the characteristic parameters of the electromagnetic signal (such as frequency, amplitude, phase, etc.) are extracted using signal processing and analysis technology as the electromagnetic fingerprint features of the parts.

[0040] In step S15, the corresponding recognition pattern is matched based on the electromagnetic fingerprint features, the automobile parts and the recognition influencing factors; In the specific implementation process of the present invention, the specific steps may be: S151: freeze-frame electromagnetic fingerprint features; S152: Determine corresponding identification influencing factors based on the detection of electromagnetic fingerprint features; S153: Correlation of electromagnetic fingerprint features, automotive components, and identification influencing factors; S154: Determine a first recognition coefficient based on the electromagnetic fingerprint feature and the automobile parts; S155: Determine a second recognition coefficient based on the electromagnetic fingerprint feature and the recognition influencing factors; S156: Associating the first recognition coefficient, the second recognition coefficient, and the recognition matching table, and matching the corresponding recognition pattern according to the first recognition coefficient, the second recognition coefficient, and the recognition matching table.

[0041] In the embodiments of the present application, the electromagnetic fingerprint feature is frozen, the electromagnetic fingerprint feature is introduced, and the electromagnetic fingerprint feature is further controlled. At the same time, the corresponding identification influencing factors are determined based on the detection of the electromagnetic fingerprint feature; the electromagnetic fingerprint feature, automobile parts and identification influencing factors are associated, the electromagnetic fingerprint feature, automobile parts and identification influencing factors are introduced, and multiple interactions of the electromagnetic fingerprint feature, automobile parts and identification influencing factors are realized.

[0042] At this time, the freeze-frame electromagnetic fingerprint feature refers to the extraction of stable and unique features from the electromagnetic signal, which can be used to identify and recognize different electromagnetic signal sources. Electromagnetic fingerprint features usually come from hardware defects of wireless devices. These defects are inevitable during the device manufacturing process, resulting in unique characteristics of the electromagnetic signal emitted by each device. These features can be extracted from basic physical properties such as amplitude, frequency, and phase of the signal. At the same time, a signal receiver is used to capture the electromagnetic signal emitted by the device to be identified, and stable and unique features are extracted from the captured signal. These features should meet the requirements of independence, stability, and measurability. The extracted features are combined to construct an electromagnetic fingerprint feature library for subsequent identification tasks.

[0043] The electromagnetic fingerprint of the device to be identified is matched with the features in the fingerprint library to determine the identity of the device. However, the identification process may be affected by many factors, resulting in inaccurate identification results. Therefore, it is necessary to determine these identification influencing factors and take corresponding measures to improve them. Identification influencing factors may include noise interference, signal sampling rate, and feature stability.

[0044] Furthermore, a first recognition coefficient is determined based on electromagnetic fingerprint features and automobile parts; a second recognition coefficient is determined based on electromagnetic fingerprint features and recognition influencing factors; the first recognition coefficient, the second recognition coefficient and a recognition matching table are associated, and corresponding recognition patterns are matched according to the first recognition coefficient, the second recognition coefficient and the recognition matching table. The first recognition coefficient, the second recognition coefficient and the recognition matching table are introduced, and multiple interactions of the first recognition coefficient, the second recognition coefficient and the recognition matching table are realized, thereby ensuring the accuracy of the recognition pattern.

[0045] At this point, the electromagnetic fingerprint features are combined with the information of the auto parts to determine a preliminary identification coefficient, namely the first identification coefficient. This coefficient is mainly based on the uniqueness and stability of the electromagnetic fingerprint features and is used to preliminarily distinguish different auto parts.

[0046] First, stable features are extracted from the electromagnetic radiation or conduction signals of automotive parts. These features can be frequency, phase, amplitude, etc. The extracted features are normalized to eliminate the differences in feature values ​​under different test conditions, and are quantified for subsequent calculations and comparisons. The normalized and quantified electromagnetic fingerprint features are combined with the basic information of automotive parts (such as model, batch, manufacturer, etc.) to form a comprehensive feature vector. Based on the above feature vector, the first recognition coefficient is calculated through a specific algorithm (such as a machine learning model).

[0047] At the same time, various influencing factors that may exist in the process of electromagnetic fingerprint feature extraction and recognition, such as environmental noise, the accuracy of the test equipment, the skills of the operator, etc., are considered to determine a more accurate recognition coefficient, namely the second recognition coefficient. This coefficient is intended to further improve the accuracy and robustness of recognition. First, various factors that may affect the extraction and recognition of electromagnetic fingerprint features are analyzed and evaluated in detail. The identification influencing factors obtained by the analysis are quantified to obtain a quantized influencing factor vector. The quantized influencing factor vector is combined with the electromagnetic fingerprint feature to form a more comprehensive feature vector. Based on the above feature vector, a more complex algorithm (such as a multi-feature fusion algorithm) is used to calculate the second recognition coefficient.

[0048] Therefore, the first recognition coefficient and the second recognition coefficient are associated with a pre-established recognition matching table to match the corresponding recognition mode. The recognition matching table is a database or lookup table containing the correspondence between different recognition coefficients and recognition modes; first, based on the known automobile parts information and electromagnetic fingerprint features, a recognition matching table containing the correspondence between different recognition coefficients and recognition modes is constructed. The first recognition coefficient and the second recognition coefficient of the automobile parts to be identified are matched with the values ​​in the recognition matching table. According to the matching results, the corresponding recognition mode is determined from the recognition matching table.

[0049] In step S16, the recognition of the electromagnetic fingerprint feature is triggered according to the recognition mode, and the corresponding recognition signal and signal attenuation degree are output, and the authenticity of the automobile parts is determined according to the recognition signal and the signal attenuation degree; In the specific implementation process of the present invention, the specific steps may be: S161: freeze the recognition mode; S162: Associating the recognition pattern with the electromagnetic fingerprint feature; S163: triggering recognition of the electromagnetic fingerprint feature based on the recognition mode and the electromagnetic fingerprint feature; S164: real-time monitoring of the recognition of electromagnetic fingerprint features, and outputting corresponding recognition signals and signal attenuation levels; S165: Correlate the identification signal and the signal attenuation level; S166: Determine the authenticity of the automotive parts based on the identification signal and the degree of signal attenuation.

[0050] In the specific implementation process of the present invention, a corresponding recognition pattern is matched based on electromagnetic fingerprint features, automobile parts and identification influencing factors; the recognition of the electromagnetic fingerprint features is triggered according to the recognition pattern, and the corresponding recognition signal and signal attenuation degree are output, and the authenticity of the automobile parts is determined according to the recognition signal and the signal attenuation degree, which is compatible with the overall consideration of the electromagnetic fingerprint features, automobile parts and identification influencing factors, ensuring the accuracy of the recognition pattern, and then introducing the recognition signal and the signal attenuation degree, realizing multi-dimensional control of the recognition signal and the signal attenuation degree, ensuring the authenticity of automobile parts, and improving the recognition accuracy of the electromagnetic fingerprint features.

[0051] At this time, the recognition pattern is frozen and introduced. At the same time, the recognition pattern and the electromagnetic fingerprint feature are associated; based on the recognition pattern and the electromagnetic fingerprint feature, the recognition of the electromagnetic fingerprint feature is triggered, the recognition pattern and the electromagnetic fingerprint feature are introduced, and multiple interactions of the recognition pattern and the electromagnetic fingerprint feature are realized, thereby ensuring the triggering of the recognition of the electromagnetic fingerprint feature.

[0052] At the same time, the identification pattern is associated with its corresponding electromagnetic fingerprint feature. This association is the key to ensure the accuracy and traceability of identification. Once the identification pattern is associated with the electromagnetic fingerprint feature, we can verify the authenticity of the parts by comparing the electromagnetic fingerprint feature and the identification pattern in the subsequent identification process. Optionally, after extracting the electromagnetic fingerprint feature of the device to be identified, it is compared with the identification pattern in the database. This step usually involves calculating the similarity or distance between the features to determine whether they match. During the comparison process, a specific matching algorithm may be required to evaluate the similarity of the features. These algorithms may include statistical-based methods, machine learning algorithms, or deep learning models. The choice of algorithm depends on the characteristics of the electromagnetic fingerprint feature and the requirements of the identification system.

[0053] In order to determine whether the features match, a threshold is usually set. If the similarity between the features exceeds the threshold, they are considered to match successfully; otherwise, the match fails. The setting of the threshold needs to take into account factors such as noise, device differences, and recognition accuracy; based on the comparison results and the threshold setting, it is determined whether the device to be identified matches a certain recognition pattern in the database. If the match is successful, the identity or type of the device can be determined based on the information of the recognition pattern; if the match fails, further verification or inspection may be required.

[0054] Therefore, the recognition of electromagnetic fingerprint features is monitored in real time, and the corresponding recognition signal and signal attenuation degree are output; the recognition signal and signal attenuation degree are associated; the authenticity of automobile parts is determined according to the recognition signal and signal attenuation degree, and the recognition signal and signal attenuation degree are introduced, and multiple interactions are performed on the recognition signal and signal attenuation degree to ensure the authenticity of automobile parts.

[0055] At this point, once the electromagnetic fingerprint feature is successfully identified, the system will output a corresponding identification signal, which represents the unique identity of the identified component. At the same time, the system also needs to calculate the attenuation of the signal from transmission to reception, which usually involves the measurement of signal strength and consideration of factors such as distance and obstacles. Further, the electromagnetic fingerprint features captured in real time are compared and analyzed with the records in the database, including a consistency check of the identification signal and the signal attenuation degree. If the electromagnetic fingerprint features captured in real time are highly consistent with the records in the database and the signal attenuation degree is within a reasonable range, the component is determined to be authentic; otherwise, it is determined to be a counterfeit or has quality problems.

[0056] Alternatively, suppose there is a car manufacturer that equips the engine control unit (ECU) with an electromagnetic fingerprint recognition system. The following is the actual application process of the system: The automobile manufacturer's factory has installed an electromagnetic signal receiving device, which captures the electromagnetic signal emitted by the ECU in real time during operation and extracts the electromagnetic fingerprint features. The system recognizes the unique identity of the ECU and outputs an identification signal.

[0057] At the same time, the system calculates the attenuation of the signal from the ECU to the receiving device and records the data. The identification signal and the signal attenuation are associated to form a complete data record and stored in the database.

[0058] When a car needs to be repaired or the ECU needs to be replaced, the maintenance personnel use the electromagnetic fingerprint recognition system to verify the new ECU. The system extracts the electromagnetic fingerprint characteristics of the ECU and compares and analyzes them with the records in the database. Since the new ECU is authentic, its electromagnetic fingerprint characteristics are highly consistent with the records in the database, and the signal attenuation is within a reasonable range. Therefore, the system determines that the ECU is authentic and allows it to be installed in the car.

[0059] In another embodiment of the present invention, the method for identifying the electromagnetic fingerprint characteristics of automobile parts is further regulated as follows: For feature extraction and selection: The hardware circuits produced by different RFID tag manufacturers during the manufacturing process will be different, so the signals responded by these circuits will also be different. The features extracted from the signals responded by the tags at the physical layer are the embodiment of the differences in the hardware circuits. Extracting the appropriate features can distinguish different types of tags. However, the differences between tags are often diverse. For example, the phase deviation of the signal is different in some types of tags, but not in other types of tags. Of course, multi-feature joint classification can be used. After the classifier is trained, the more important features will be given a larger weight, while the less important features will be given a smaller weight. However, in this case, determining the number of features is still a problem to be solved. From the perspective of describing the characteristics of the classification object, the more features the better, because the various characteristics of the object will be shown as much as possible. However, when the number of features is large, redundancy will inevitably occur. This part of the redundant features is not only redundant during classification, but also reduces the classification performance.

[0060] Since the EPC codes of genuine and fake tags are the same, the difference between tag signals is more reflected in their respective noise signals, and the normalized EPC signals of each tag will also be different, such as frequency drift. In actual tests, this paper found that the performance of classifiers with more features is not better than that of classifiers with fewer features. On the other hand, the classification performance of classifiers with fewer features will also decrease. It can be seen that there is actually an optimal value for the number of selected features. Too many will cause redundancy, and too few will cause information loss.

[0061] For classification tests: Traditional tests of the classification performance of physical layer identification methods mostly use the classic cross-validation method, in which each type of data is randomly divided into several parts, one part as a training set and the other part as a test set. The data types in the training set should also be in the test set. Class 1 and Class 2 are true labels and attack labels, respectively. It can be seen that there are attack labels in the training set and there are also such attack labels in the test set. However, this situation may not be guaranteed in practical applications. Since it is impossible to predict what kind of labels the attacker uses, although as many types of labels as possible can be included in the training library, it is inevitable that a type of label used by the attacker is missing. Since the attack label already exists in the training set, the probability of identifying the attack label in the test set is relatively high. From this perspective, the classification accuracy obtained by using the traditional 5-fold cross-validation method to test the physical layer identification method will be artificially high.

[0062] For the algorithm: Traditional methods only extract features from the tag response signal received by the reader. However, the tag's noise signal and expected EPC signal also carry their own information, so features can also be extracted from the above signals.

[0063] First, the tag signal received by the reader and obtained by the RF antenna array is IQ modulated to obtain I and Q signals, and then the modulus is obtained to obtain a(n), where n=1, 2, ...N is the sampling point. Then, the signal is clustered into a center point vector V=[v0,v1] T , expressed as: V = clu[a(n)] Among them, clu[·] is the clustering function, and v0 and v1 are the center points corresponding to the code element taking 0 and 1 respectively. It should be noted that the two center vectors V obtained after clustering do not contain the information of 0 and 1. However, the tag will have a silent period signal before sending its RN16. This signal has been specified in EPC C1 Gen2. Therefore, the EPC signal can be clustered together with this signal to determine v0 and v1. The clustering process is as follows. First, v0' can be obtained by clustering the silent period. Then, the two center points obtained by clustering the EPC signal are determined. The one closer to v0' is determined as v0, and the other is determined as v1. From the cluster center point, the Euclidean distance judgment of a(n) is obtained to obtain the expected signal: a e (n)=dec[a(n)] in, , In theory, the expected signals of the real tag and the cloned tag should be the same because their EPCs are the same, but due to frequency drift, there will be differences in the periods between the tags, and therefore the related features extracted from them will also be different.

[0064] Thirdly, due to the different transmission power, reading distance and tag sensitivity when the reader reads the tag, the amplitude of the tag response signal is different. To eliminate these differences, the signal can be standardized as follows: , Most of the sampling points are near 0 and 1, so the relevant features extracted from the signal will show less amplitude difference of the signal.

[0065] Finally, the noise signal can be obtained by subtracting the normalized signal from the expected signal, which is expressed as: , If the difference in frequency drift is ignored, the EPC expected signal of each tag signal should be the same, so the difference between tag signals will be more reflected in the noise signal.

[0066] After the above processing, this paper will get the expected signal , Standard signal and noise signal , together with the original received signal , a total of 4 groups of signals.

[0067] The expected signal will be , Standard signal and noise signal , together with the original received signal , a total of 4 groups of signals. Perform cross spectral entropy feature extraction. Cross spectral entropy is obtained based on spectral entropy and Wiener-Khinchin theorem, which can effectively represent the degree of correlation between two signals.

[0068] Spectral entropy is a measure that describes the relationship between the power spectrum and the entropy rate. It is expressed as: , in, It is the probability density function of the proportion of the energy of each frequency band to the total energy.

[0069] , Due to the Wiener-Khinchin theorem, the cross-power spectral density has two definitions, one in the frequency domain and the other in the time domain, namely: , The definition of cross spectral entropy is: , Cross spectral entropy indicates the correlation between two time series. The greater the correlation, the stronger the independence. The cross spectral entropy value is between 0 and 1. In order to ensure the integrity of the extracted feature information, a new indicator is introduced: the maximum amplitude , Maximum Phase , Maximum frequency . Its expression is: , At the same time, the second amplitude, phase, and frequency are also defined. A total of 7 features.

[0070] 2.2 MT-BP neural network design The above 7 features are used as sample input data of the MT-BP neural network, and the sample input and output data of the MT-BP neural network are set as column vectors: , The learning process of the MT-BP neural network includes the forward propagation of the working signal and the reverse transmission of the error. During the forward transmission, the input signal is transmitted from the input layer to the neurons of the hidden layer for processing, and finally transmitted to the output layer, while the neurons of the previous layer can only act on the next layer. When the error between the expected value and the actual value of the output is large, the network system will enter the reverse transmission, so that the error signal returns along the original line, and then adjust the weights and thresholds of each neuron so that the error finally reaches the expected minimum error, and finally use the best network for prediction to obtain a reasonable expected value.

[0071] The activation function used in the input layer is the linear function, and the activation function used in the hidden layer is the sigmoid function. The following is the internal relationship between the input and output of the network: 1. For the input layer: In the input layer, the input data is the parameters X of each sealing structure i (t), since the function used is a linear function, the output y t =X i .

[0072] 2. For the hidden layer: In the hidden layer, the input value of the neuron is the input value to which it is linked multiplied by the corresponding weight, and then added to the threshold of the neuron. As shown below: , Where: n represents the number of neurons in the input layer, and j represents the number of neurons in the hidden layer.

[0073] The output of the hidden layer will be operated by the sigmoid function, and its value is as follows: , 3. For the output layer: In the output layer, the calculation method and steps of its neurons are the same as those of the hidden layer, as shown below: , Similarly, the output value of the output layer is as follows: , To summarize, the basic steps of neural network training are as follows: (1) Weight initialization: Initialize the weights w of the input layer and the hidden layer ji And the weights w of the hidden layer and the output layer kj Randomly assign a set of non-zero initial values ​​to each.

[0074] (2) Set the network structure parameters, such as the training target error, maximum number of iterations, learning rate, and momentum coefficient, and then train each sample according to steps ① to ⑥ until the error reaches the convergence condition or the upper limit of the training number is reached.

[0075] ① Calculate the output y of each layer starting from the input layer i ,y j ,y k ; ② Calculate the error between the output layer neuron prediction value and the true value: , ③After calculating the error value of the predicted value, calculate the error value of each neuron in the previous hidden layer: , ④ Adjust the weights and thresholds of each layer according to the error: , in: , Where: The number of iterations m is generally 100~10000; Momentum coefficient m c Generally, it is taken as 0.9~1; Learning rate l r It is generally set between 0.1 and 3, which greatly affects the error convergence and iteration step size.

[0076] ⑤After a complete calculation of the feedforward propagation, compare the error With training goals size.

[0077] ⑥ If , the training is terminated; otherwise, continue to execute ③ and ④, calculate the error of reverse transmission, adjust the weights and thresholds between layers, execute ① and ②, and re-perform feedforward forward propagation until the error meets the requirements and the training is terminated.

[0078] The evaluation indicators of the MT-BP neural network model are divided into two categories according to their properties: classification properties or regression properties. For the regression problem used in this article, there are usually five evaluation indicators: mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), mean relative error (MAPE) and coefficient of determination (R 2 ), the latter two are dimensionless, so they are more accurate in evaluating the performance of neural networks. Their expressions are: , Where: n represents the total number of samples, and k represents the kth sample; represents the predicted value of the kth sample, Represents the corresponding sample reality.

[0079] For the results: The experimental data comes from the passive UHF RFID tags specified by EPC C1 Gen2. 140 tags of 7 common types on the market are used. The 7 types of tags are manufactured by three manufacturers. The details are shown in Table 1. Before collecting data, the same EPC code is written to all 140 tags. The reader used is the UHF100U reader from Guangzhou Wangyuan Electronics. Its system parameters are shown in Table 2.

[0080] , , After all tags are written with the same EPC code, the tag data is collected by an ultra-high frequency RFID system built with a USRP software radio. The system complies with the EPC C1 Gen2 standard and the software is implemented using GNURadio.

[0081] Each time data is collected, only one tag is placed in front of the antenna, and the rest of the tags are not within the magnetic field range of the reading system to reduce the risk of tag conflict. All data collection is not performed in an isolated environment, that is, the environment includes thermal noise, mobile phone noise, wireless network and radio frequency noise. Tags are randomly placed in a rectangular area formed by two antennas, the area of ​​which is 25cmx25cm. Each tag records 10s of data, randomly intercepts the signal containing the silent period and the EPC code, and stores it in MATLAB format as the data to be classified, and a total of 140 tag data are obtained.

[0082] The RFID reader is deployed next to the gantry placed indoors, and the information of the RFID tags within the reading range is collected in real time through the external RFID antenna array. The RFID tags are installed on the automotive parts to be sensed, and the initial data is entered into the MT-BP neural network. The initial data includes learning rate, training rounds, initial weight value, tag classification category, tag location range, etc. RFID tags are divided into different categories of tags, and the tags are placed on each automotive part. This type of tag is a passive RFID tag that is sensitive to antenna transmission power and communicates by backscattering the energy of the RFID antenna.

[0083] The four antennas are evenly deployed on the gantry placed indoors to scan the RFID tags within the reading range from different angles; each antenna can adjust parameters based on the instructions issued by the reader and use different transmission powers to scan tags; at the same time, the reflected signal energy value of the RFID tag is accurately collected and used as an identification reference. The RFID antenna array transmits RF signals, and the RFID reader collects the reflected RF signal data in real time through the RFID antenna array; each set of data consists of the RF signal strength values ​​and the arrival time of the feedback RF signals from the four RFID antennas on the gantry, and the time-frequency division multiplexing technology is used to solve the signal interference problem when multiple tags in a mixed automotive parts group respond simultaneously.

[0084] The MT-BP neural network module processes and trains the acquired reflected RF signal data to obtain a multi-tag RF perception recognition prediction model based on the MT-BP neural network, which can identify any RFID tag category and the position of the RFID tag in the pile of automotive parts through the reflected RF signal data. According to the attenuated signal power of the RFID antenna array, the reference tag in the critical area and its multi-dimensional signal strength information are filtered out, and various information data are input into the MT-BP neural network multi-tag RF perception recognition prediction model. According to the decay law of the feedback RF signal strength obtained by the RFID antenna array, the specific signal attenuation strength of different types of RFID tags is used to predict the tag category of the perceived target tag. Then, the position of the target tag is weightedly calculated by using the known position of the reference tag and the Euclidean distance of the signal strength between the reference tag and the target tag through the arrival time difference of the feedback RF signal. Thus, the result of the multi-tag RF perception recognition method based on the MT-BP neural network is obtained.

[0085] The perception recognition accuracy is optimized through machine learning algorithms. The actual label category and location data are measured to obtain the real label of the target object, and compared with the above perception results, the historical feedback information is obtained and recorded in the historical database, and the model is trained through supervised learning using regression analysis. In the next perception process, the multi-dimensional signal strength information of the target label is compared with the historical feedback information to obtain the historical feedback information with the closest Euclidean distance to the signal strength of the target label, and the recognition result is adjusted using this information and its corresponding regional location. The final recognition result is obtained in a weighted manner as the perception recognition result of the MT-BP neural network multi-label radio frequency perception recognition prediction optimization model method based on regression analysis, thereby further improving the recognition accuracy.

[0086] In an embodiment of the present invention, the location of automobile parts is collected through the method in the embodiment of the present invention; the corresponding positioning detection is triggered according to the location of the automobile parts, and the identification area of ​​the automobile parts is collected according to the positioning detection of the automobile parts; the identification tag of the automobile part is determined based on the traversal of the identification area of ​​the automobile part; the electromagnetic fingerprint feature of the automobile part is determined according to the identification tag of the automobile part, thereby realizing the precise control of the electromagnetic fingerprint feature of the automobile part.

[0087] Furthermore, a corresponding recognition pattern is matched based on electromagnetic fingerprint features, automobile parts and identification influencing factors; recognition of the electromagnetic fingerprint features is triggered according to the recognition pattern, and a corresponding recognition signal and a signal attenuation degree are output, and the authenticity of the automobile parts is determined according to the recognition signal and the signal attenuation degree, which is compatible with the overall consideration of the electromagnetic fingerprint features, automobile parts and identification influencing factors, ensuring the accuracy of the recognition pattern, and then introducing the recognition signal and the signal attenuation degree, realizing multi-dimensional control of the recognition signal and the signal attenuation degree, ensuring the authenticity of automobile parts recognition, and improving the recognition accuracy of the electromagnetic fingerprint features.

[0088] See also Figure 2 , Figure 2 It is a schematic diagram of the structure of the electromagnetic fingerprint feature recognition system of automobile parts in an embodiment of the present invention.

[0089] like Figure 2 As shown, a system for identifying electromagnetic fingerprint features of automobile parts, the system for identifying electromagnetic fingerprint features of automobile parts comprises: A collection module 21 is used to collect the locations of automobile parts; The identification area module 22 is used to trigger the corresponding positioning detection according to the location of the automobile parts, and collect the identification area of ​​the automobile parts according to the positioning detection of the automobile parts; An identification tag module 23, for determining an identification tag of an automobile part based on traversal of an identification area of ​​the automobile part; The electromagnetic fingerprint feature module 24 is used to determine the electromagnetic fingerprint feature of the automobile parts according to the identification tags of the automobile parts; Recognition mode module 25, used for matching corresponding recognition modes based on electromagnetic fingerprint features, automobile parts and recognition influencing factors; The identification module 26 is used to trigger the identification of the electromagnetic fingerprint feature according to the identification mode, and output the corresponding identification signal and the signal attenuation degree, and determine the authenticity of the automobile parts according to the identification signal and the signal attenuation degree.

[0090] See also Figure 3 , refer to the following Figure 3 The electronic device 40 according to this embodiment of the present invention will be described. Figure 3 The electronic device 40 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0091] like Figure 3As shown, the electronic device 40 is in the form of a general computing device. The components of the electronic device 40 may include but are not limited to: at least one processing unit 41, at least one storage unit 42, and a bus 43 connecting different system components (including the storage unit 42 and the processing unit 41).

[0092] The storage unit stores program codes, which can be executed by the processing unit 41, so that the processing unit 41 executes the steps according to various exemplary embodiments of the present invention described in the above “Embodiment Method” section of this specification.

[0093] The storage unit 42 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 421 and / or a cache memory unit 422 , and may further include a read-only memory unit (ROM) 423 .

[0094] The storage unit 42 may also include a program / utility 424 having a set (at least one) of program modules 425, such program modules 425 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination thereof includes an implementation of a network environment.

[0095] Bus 43 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0096] The electronic device 40 may also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable a user to interact with the electronic device 40, and / or any device that enables the electronic device 40 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 44. Furthermore, the electronic device 40 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 45. Figure 3 As shown, the network adapter 45 communicates with other modules of the electronic device 40 via the bus 43. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 40, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup planning systems.

[0097] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0098] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium can include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc. In addition, it stores computer program instructions, and when the computer program instructions are executed by a computer, the computer executes the above method.

[0099] In addition, the above is a detailed introduction to the electromagnetic fingerprint feature recognition method and system for automotive parts provided in the embodiments of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for identifying electromagnetic fingerprint features of automobile parts, characterized in that: Applied to the recognition scenario of electromagnetic fingerprint features of automotive parts; The method for identifying electromagnetic fingerprint features of automobile parts comprises: Collect the location of auto parts; Trigger corresponding positioning detection according to the location of the automobile parts, and collect the identification area of ​​the automobile parts according to the positioning detection of the automobile parts; Determining an identification tag of the automobile part based on traversal of the identification area of ​​the automobile part; Determine the electromagnetic fingerprint characteristics of the automobile parts according to the identification tags of the automobile parts; Match the corresponding recognition mode based on electromagnetic fingerprint features, automobile parts and recognition influencing factors; The recognition of the electromagnetic fingerprint feature is triggered according to the recognition mode, and the corresponding recognition signal and signal attenuation degree are output, and the authenticity of the automobile parts is determined according to the recognition signal and the signal attenuation degree.

2. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 1, characterized in that: The location of the collected automobile parts includes: Collect production signals of automotive parts; Determine the corresponding position information based on the analysis of the production signal of the automobile parts; Determine the space where the automobile parts are located based on the interpretation of the position information; The location of the automobile parts is determined based on the autonomous detection of the space where the automobile parts are located, so as to collect the location of the automobile parts.

3. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 1, characterized in that: The triggering of corresponding positioning detection according to the location of the automobile parts, and collecting the identification area of ​​the automobile parts according to the positioning detection of the automobile parts, includes: Determine the location of auto parts; Trigger corresponding positioning detection based on the location of automobile parts; Real-time monitoring of positioning detection of automotive parts; Collect multiple area parameters based on the positioning detection of automobile parts; Associating multiple area parameters and vehicle parts; The identification area of ​​the automobile parts is determined based on multiple area parameters and the automobile parts to collect the identification area of ​​the automobile parts.

4. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 3, characterized in that: The method of determining the identification tag of the automobile part based on traversing the identification area of ​​the automobile part includes: Freeze the identification area of ​​automotive parts; Determine a corresponding traversal mode based on the recognition area of ​​the automobile parts and the outline of the automobile parts; Triggering the traversal of the identification area of ​​the automobile parts according to the identification area of ​​the automobile parts and the corresponding traversal mode; Based on the traversal of the identification area of ​​the automobile part, the identification tag of the automobile part is determined, and the position of the identification tag of the automobile part is marked.

5. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 4, characterized in that: The method of determining the electromagnetic fingerprint characteristics of the automobile parts according to the identification tags of the automobile parts comprises: Identification labels for fixed-frame automotive parts; an identification tag that associates the auto part with the space where the auto part is located; Matching corresponding label recognition patterns according to the identification labels of the automobile parts and the space where the automobile parts are located; Associate identification tags of automotive parts and corresponding tag identification patterns; The electromagnetic fingerprint characteristics of the automobile parts are determined according to the identification tags of the automobile parts and the corresponding tag recognition patterns.

6. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 5, characterized in that: The identification mode corresponding to the matching of electromagnetic fingerprint features, automobile parts and identification influencing factors includes: Freeze electromagnetic fingerprint features; The corresponding identification influencing factors are determined based on the detection of electromagnetic fingerprint features.

7. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 6, characterized in that: The identification mode corresponding to the matching of electromagnetic fingerprint features, automobile parts and identification influencing factors also includes: Correlate electromagnetic fingerprint features, automotive components, and identification influencing factors; Determine a first recognition coefficient based on the electromagnetic fingerprint feature and the automobile parts; Determining a second recognition coefficient based on the electromagnetic fingerprint feature and the recognition influencing factors; The first recognition coefficient, the second recognition coefficient and the recognition matching table are associated, and the corresponding recognition mode is matched according to the first recognition coefficient, the second recognition coefficient and the recognition matching table.

8. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 7, characterized in that: The method triggers the recognition of electromagnetic fingerprint features according to the recognition mode, outputs corresponding recognition signals and signal attenuation, and determines the authenticity of automobile parts according to the recognition signals and signal attenuation, including: Freeze the recognition pattern; associating the recognition pattern with the electromagnetic fingerprint feature; Recognition of the electromagnetic fingerprint feature is triggered based on the recognition pattern and the electromagnetic fingerprint feature.

9. The method for identifying electromagnetic fingerprint features of automobile parts according to claim 8, characterized in that: The method triggers the recognition of electromagnetic fingerprint features according to the recognition mode, outputs a corresponding recognition signal and a signal attenuation degree, and determines the authenticity of the automobile parts according to the recognition signal and the signal attenuation degree, and further includes: Monitor the recognition of electromagnetic fingerprint features in real time and output the corresponding recognition signal and signal attenuation degree; Correlate identification signals and signal attenuation levels; The authenticity of automotive parts can be determined based on the identification signal and the degree of signal attenuation.

10. A recognition system for electromagnetic fingerprint features of automobile parts, characterized in that: The electromagnetic fingerprint feature recognition system of automobile parts is applied to the electromagnetic fingerprint feature recognition method of automobile parts as claimed in any one of claims 1 to 9, and the electromagnetic fingerprint feature recognition system of automobile parts comprises: A collection module is used to collect the locations of automobile parts; An identification area module is used to trigger corresponding positioning detection according to the location of the automobile parts, and collect the identification area of ​​the automobile parts according to the positioning detection of the automobile parts; An identification tag module, used to determine the identification tag of the automobile part based on the traversal of the identification area of ​​the automobile part; An electromagnetic fingerprint feature module, used to determine the electromagnetic fingerprint features of automobile parts according to identification tags of automobile parts; Recognition mode module, used to match corresponding recognition modes based on electromagnetic fingerprint features, automobile parts and recognition influencing factors; The recognition module is used to trigger the recognition of electromagnetic fingerprint features according to the recognition mode, and output the corresponding recognition signal and signal attenuation degree, and determine the authenticity of the automobile parts according to the recognition signal and signal attenuation degree.

Images