Waveform anomaly detection method, electronic device, readable storage medium and program product

Abstract

The application discloses a waveform anomaly detection method, an electronic device, a computer readable storage medium and a computer program product, and relates to the technical field of computer vision; the method comprises the following steps: acquiring a waveform image and a plurality of waveform index parameters in the waveform image; waveform anomaly type weight information corresponding to each preset waveform anomaly type is generated according to the plurality of waveform index parameters; and waveform anomaly detection is performed on the waveform in the waveform image according to the waveform image and the waveform anomaly type weight information. The application can solve the technical problem of low precision of waveform anomaly detection.

Description

Technical Field

[0001] This application relates to the field of computer vision technology, and in particular to a waveform anomaly detection method, electronic device, computer-readable storage medium, and computer program product. Background Technology

[0002] In the hardware functional testing of communication equipment, waveform testing is usually used to test signal integrity.

[0003] Currently, image matching algorithms or large models are typically used for waveform anomaly detection during waveform testing. The specific principle is to identify the type of waveform anomaly by comparing waveform shape features or visual features.

[0004] However, for some similar waveform anomaly types, it is difficult to accurately identify them based solely on waveform shape features or visual features, resulting in low accuracy in waveform anomaly detection. Summary of the Invention

[0005] The main objective of this application is to provide a waveform anomaly detection method, electronic device, computer-readable storage medium, and computer program product, aiming to solve the technical problem of low accuracy in waveform anomaly detection in related technical solutions.

[0006] To achieve the above objectives, this application provides a waveform anomaly detection method, the method comprising: Acquire waveform images from an oscilloscope and multiple waveform index parameters in the waveform images; generate waveform anomaly type weight information corresponding to each preset waveform anomaly type based on the multiple waveform index parameters; Based on the waveform image and the waveform anomaly type weight information, waveform anomaly detection is performed on the waveform in the waveform image.

[0007] In addition, to achieve the above objectives, this application also provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the waveform anomaly detection method as described above.

[0008] In addition, to achieve the above objectives, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the waveform anomaly detection method described above.

[0009] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the waveform anomaly detection method described above.

[0010] In this application, a waveform image from an oscilloscope and multiple waveform index parameters within the waveform image can be acquired. Then, based on these multiple waveform index parameters, waveform anomaly type weight information corresponding to each preset waveform anomaly type can be generated. This allows for the determination of the detection importance of each preset waveform anomaly type during waveform anomaly detection. Furthermore, based on the waveform image and the waveform anomaly type weight information, waveform anomaly detection is performed on the waveform in the waveform image. Since the waveform anomaly type weight information is generated based on multiple waveform index parameters, this application adds additional guiding feature information during the waveform anomaly detection process for some similar waveform anomaly types. Therefore, it can accurately distinguish or identify similar waveform anomaly types, thus improving the accuracy of waveform anomaly detection in the waveform image. Attached Figure Description

[0011] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0012] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating a waveform anomaly detection method in one embodiment of this application; Figure 2 This is a schematic diagram of the process for generating waveform anomaly type weight information corresponding to each preset waveform anomaly type in one embodiment of this application; Figure 3 This is a schematic diagram illustrating the process of detecting waveform anomalies in a waveform image according to one embodiment of this application; Figure 4 This is a schematic diagram illustrating the process of detecting waveform anomalies in a waveform image in another embodiment of this application; Figure 5 This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the waveform anomaly detection method in one embodiment of this application.

[0014] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0015] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0016] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0017] It should be noted that the execution subject of this embodiment may include, but is not limited to, a server, a computer, or any electronic device capable of performing the above functions. This embodiment does not specifically limit this.

[0018] Currently, waveform anomaly detection typically employs image matching algorithms or large models in waveform testing. The principle behind this is to identify waveform anomaly types by comparing waveform shape features or visual characteristics. However, for some similar waveform anomaly types, accurate identification based solely on waveform shape features or visual characteristics is difficult, resulting in low accuracy in waveform anomaly detection.

[0019] In this application, a waveform image from an oscilloscope and multiple waveform index parameters within the waveform image can be acquired. Then, based on these multiple waveform index parameters, waveform anomaly type weight information corresponding to each preset waveform anomaly type can be generated. This allows for the determination of the detection importance of each preset waveform anomaly type during waveform anomaly detection. Furthermore, based on the waveform image and the waveform anomaly type weight information, waveform anomaly detection is performed on the waveform in the waveform image. Since the waveform anomaly type weight information is generated based on multiple waveform index parameters, this application adds additional guiding feature information during the waveform anomaly detection process for some similar waveform anomaly types. Therefore, it can accurately distinguish or identify similar waveform anomaly types, thus improving the accuracy of waveform anomaly detection in the waveform image and solving the technical problem of low accuracy in waveform anomaly detection.

[0020] To address this, this embodiment provides a waveform anomaly detection method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the waveform anomaly detection method proposed in this embodiment. The waveform anomaly detection method includes steps 202 to 206: Step 202: Obtain the waveform image and multiple waveform index parameters in the waveform image.

[0021] The waveform image can be an image displaying a waveform, which can be output by an oscilloscope or captured by a camera on the oscilloscope's display interface; in this embodiment, the method of acquiring the waveform image is not limited; the waveform index parameters can be numerical parameters displayed in the waveform image, such as level value, slope, and overshoot value.

[0022] As an example, in this embodiment, text recognition can be performed on the waveform image to obtain multiple waveform index parameters in the waveform image; alternatively, multiple waveform index parameters in the waveform image can be directly read through the preset communication interface of the oscilloscope that outputs the waveform image.

[0023] Step 204: Generate waveform anomaly type weight information corresponding to each preset waveform anomaly type based on multiple waveform index parameters.

[0024] In this embodiment, a mapping relationship between each waveform index parameter and each preset waveform anomaly type is pre-set. That is, for each preset waveform anomaly type, there will be one or more associated target waveform index parameters.

[0025] As an example, step 204 includes: determining at least one target indicator parameter associated with each preset waveform anomaly type from multiple waveform indicator parameters; and generating waveform anomaly type weight information corresponding to each preset waveform anomaly type based on the at least one target indicator parameter associated with each preset waveform anomaly type.

[0026] As an example, waveform anomaly type weight information can be the target anomaly type weight. In this embodiment, the mapping relationship between the target indicator parameter and the anomaly type weight under each preset waveform anomaly type can be pre-set. In this way, the corresponding anomaly type weight can be queried according to each target indicator parameter. When the preset waveform anomaly type corresponds to multiple target indicator parameters, the anomaly type weights corresponding to the multiple target indicator parameters can be weighted and averaged to obtain the target anomaly type weight. When the preset waveform anomaly type corresponds to a single target indicator parameter, the anomaly type weight corresponding to the single target indicator parameter can be determined as the target anomaly type weight.

[0027] Step 206: Based on the waveform image and waveform anomaly type weight information, perform waveform anomaly detection on the waveform in the waveform image.

[0028] In this embodiment, a waveform detection model can be used to detect waveform anomalies. The waveform detection model can be a preset large model or a neural network model, and there is no limitation here.

[0029] As an example, step 206 includes: using waveform anomaly type weight information as guiding feature information, and using a waveform detection model to detect waveform anomalies in the waveform image as input.

[0030] As an example, the results of waveform anomaly detection may include the predicted waveform anomaly type, which may be a collapse type, a ringing type, or a back channel type, etc.; the results of waveform anomaly detection may also include the location of the edge where the waveform anomaly is located and the location of the anomaly detection box, etc.

[0031] In this embodiment, a waveform image from an oscilloscope and multiple waveform index parameters within the waveform image can be acquired. Then, based on these multiple waveform index parameters, waveform anomaly type weight information corresponding to each preset waveform anomaly type can be generated. This determines the detection importance of each preset waveform anomaly type during the current waveform anomaly detection. Therefore, based on the waveform image and the waveform anomaly type weight information, waveform anomaly detection is performed on the waveform in the waveform image. Since the waveform anomaly type weight information is generated based on multiple waveform index parameters, this embodiment adds additional guiding feature information during the waveform anomaly detection process for some similar waveform anomaly types. Therefore, similar waveform anomaly types can be accurately distinguished or identified, thus improving the accuracy of waveform anomaly detection in the waveform image.

[0032] Reference Figure 2 In one feasible implementation, waveform anomaly type weight information corresponding to each preset waveform anomaly type is generated based on multiple waveform index parameters, including: Step 302: Determine the target index parameter corresponding to each preset waveform anomaly type from multiple waveform index parameters.

[0033] In this embodiment, a mapping relationship between each waveform index parameter and each preset waveform anomaly type is pre-set. That is, for each preset waveform anomaly type, there will be one or more associated target waveform index parameters.

[0034] Step 304: Based on the target index parameters corresponding to each preset waveform anomaly type, generate the anomaly type weight for each preset waveform anomaly type to obtain waveform anomaly type weight information.

[0035] The waveform anomaly type weight information may include the anomaly type weight corresponding to each preset waveform anomaly type.

[0036] As an example, step 304 includes: for each preset waveform anomaly type, generating anomaly type weight corresponding to the preset waveform anomaly type based on the difference between the target index parameter corresponding to the preset waveform anomaly type and the corresponding standard index parameter.

[0037] As an example, based on the target index parameters corresponding to each preset waveform anomaly type, the anomaly type weights for each preset waveform anomaly type are generated, including: For each preset waveform anomaly type: determine the target index parameters corresponding to the preset waveform anomaly type, and the standard index parameters corresponding to each target index parameter; generate the anomaly type weight corresponding to the preset waveform anomaly type based on each target index parameter and each standard index parameter.

[0038] In this embodiment, the target indicator parameters corresponding to the preset waveform anomaly type can be determined from multiple waveform indicator parameters based on the mapping relationship between the preset waveform anomaly type and the indicator parameter type. The standard indicator parameters corresponding to each target indicator parameter can be obtained, and the differences between each target indicator parameter and the corresponding standard indicator parameter can be weighted and merged to obtain the anomaly type weight corresponding to the preset waveform anomaly type.

[0039] It should be noted that the weighted merging method can be either weighted summation or weighted average, and no limitation is made here.

[0040] As an example, based on the target indicator parameters and the standard indicator parameters, anomaly type weights corresponding to preset waveform anomaly types are generated, including: Determine the standard library corresponding to each standard indicator parameter, and the priority of each standard library; determine the deviation value between each target indicator parameter and the corresponding standard indicator parameter; generate the anomaly type weight corresponding to the preset waveform anomaly type based on each deviation value and the priority of each standard library.

[0041] In this embodiment, at least some of the standard indicator parameters can come from different indicator standard libraries. These indicator standard libraries can be user-defined standard libraries, product-level standard libraries, and industry-general standard libraries, etc. In this embodiment, different standard library priorities can be set for each indicator standard library. For example, the standard library priorities can be set from largest to smallest as user-defined standard libraries, product-level standard libraries, and industry-general standard libraries, etc. The standard library priority setting strategy for each indicator standard library in this embodiment can be flexibly set according to actual needs and is not limited here.

[0042] Specifically, the standard library to which each standard indicator parameter belongs, and the standard library priority of each standard library, are determined. The deviation between each target indicator parameter and its corresponding standard indicator parameter is calculated. It should be noted that a target indicator parameter can correspond to one or more standard indicator parameters, and these standard indicator parameters can come from different standard libraries. Based on the deviation values ​​and the standard library priorities of each standard library, anomaly type weights corresponding to preset waveform anomaly types are generated. Thus, in this embodiment, the anomaly type weights corresponding to preset waveform anomaly types can be quantitatively generated based on the standard library priorities of each standard library and multiple standard indicator parameters, ensuring the accuracy of the anomaly type weight generation.

[0043] As an example, based on the deviation values ​​and the corresponding standard library priorities of each indicator standard library, anomaly type weights corresponding to preset waveform anomaly types are generated, including: For multiple deviation values ​​corresponding to the same target indicator parameter, the deviation value corresponding to the indicator with the highest priority in the standard library is selected as the target deviation value; the target deviation values ​​corresponding to each target indicator parameter are weighted and merged to obtain a weighted merged value; the weighted merged value is normalized to obtain the anomaly type weight corresponding to the preset waveform anomaly type.

[0044] As an example, based on the deviation values ​​and the corresponding standard library priorities of each indicator standard library, anomaly type weights corresponding to preset waveform anomaly types are generated, including: For multiple deviation values ​​corresponding to the same target indicator parameter, a weighted weight is set for each deviation value according to the priority of the indicator standard library corresponding to each deviation value. The weighted weight is proportional to the priority of the standard library; that is, the higher the priority of the standard library, the larger the weighted weight, and the lower the priority of the standard library, the smaller the weighted weight. Based on each weighted weight, the deviation values ​​are summed to obtain the target deviation value. The target deviation values ​​corresponding to each target indicator parameter are then weighted and merged to obtain a weighted merged value. The weighted merged value is then normalized to obtain the anomaly type weight corresponding to the preset waveform anomaly type.

[0045] As an example, the calculation process of the weights of each of the above-mentioned anomaly types can be implemented based on a rule engine in this embodiment, which will not be elaborated here.

[0046] In this embodiment, the target index parameters corresponding to each preset waveform anomaly type can be determined from multiple waveform index parameters. Then, based on the target index parameters corresponding to each preset waveform anomaly type, the anomaly type weight of each preset waveform anomaly type can be generated to obtain waveform anomaly type weight information. This achieves reliable and accurate generation of waveform anomaly weight information, providing reliable data support for subsequent waveform anomaly detection.

[0047] Reference Figure 3 In one feasible implementation, waveform anomaly detection is performed on the waveform in the waveform image based on the waveform image and waveform anomaly type weight information, including: Step 402: Determine the detection priority of each preset waveform anomaly type based on the waveform anomaly type weight information.

[0048] The waveform anomaly type weight information can include the anomaly type weight of each preset waveform anomaly type; the anomaly type weight is directly proportional to the detection priority, that is, the larger the anomaly type weight, the higher the detection priority, and the smaller the anomaly type weight, the lower the detection priority.

[0049] Step 404: Generate corresponding anomaly detection prompts based on detection priority.

[0050] In this embodiment, the mapping relationship between the detection priority of each preset waveform anomaly type and the anomaly detection prompt word can be preset. For example, if the detection priority of preset waveform anomaly type A is the highest, the anomaly detection prompt word generated is "aaa". If the detection priority of preset waveform anomaly type b is the highest, the anomaly detection prompt word generated is "bbb". In this embodiment, the anomaly detection prompt word can be generated according to the highest detection priority among the detection priorities of each preset waveform anomaly type.

[0051] As an example, assuming that the preset waveform anomaly type with the highest detection priority is ringing, the generated anomaly detection prompt could be: "Please prioritize comparing ringing features, focusing on the shape of the waveform peak region; if high-frequency oscillations are detected in the peak region, the possibility of ringing is high; at the same time, please also check for other anomalies such as hooks, steps, and collapses."

[0052] Step 406: Using the waveform image and anomaly detection prompts as input, perform waveform anomaly detection on the waveform in the waveform image using a preset large model.

[0053] In this embodiment, a preset large model can be used to detect waveform anomalies in the waveform image. In this way, the waveform image can be used as the prediction input of the preset large model, and the anomaly detection prompt words can be used as the guiding feature input. The preset large model is used to detect waveform anomalies in the waveform image, and finally the preset large model outputs the waveform anomaly detection result.

[0054] In this embodiment, the detection priority of each preset waveform anomaly type can be determined based on the waveform anomaly type weight information. Then, according to the detection priority, corresponding anomaly detection prompt words can be generated. In this way, based on the prompt word project built into the large model, the waveform image and the anomaly detection prompt words are used as input, and the preset large model is used to perform waveform anomaly detection on the waveform in the waveform image. This adds additional guiding feature information to the waveform anomaly detection process, so it can accurately distinguish or identify similar waveform anomaly types, which helps to improve the accuracy of waveform anomaly detection.

[0055] Reference Figure 4 In one feasible implementation, the waveform anomaly type weight information includes the anomaly type weight corresponding to each preset waveform anomaly type; based on the waveform image and the waveform anomaly type weight information, waveform anomaly detection is performed on the waveform in the waveform image, including: Step 502: Based on the weights of each anomaly type, adjust the output weights of the classification heads corresponding to each preset waveform anomaly type in the preset large model.

[0056] The preset large model can include the output weights of the classification head corresponding to each preset waveform anomaly type. The larger the output weight of the classification head, the more the preset large model pays attention to the preset waveform anomaly type corresponding to the output weight of the classification head when performing waveform anomaly detection. The smaller the output weight of the classification head, the less the preset large model pays attention to the preset waveform anomaly type corresponding to the output weight of the classification head when performing waveform anomaly detection.

[0057] As an example, step 502 includes: generating corresponding weight adjustment information according to the weight of each anomaly type, wherein the weight adjustment information may include the weight adjustment magnitude and the weight adjustment direction, wherein the weight adjustment direction may be an increasing direction or a decreasing direction; and adjusting the output weight of the classification head corresponding to each preset waveform anomaly type in the preset large model according to the weight adjustment information.

[0058] As an example, based on the weight of each anomaly type, corresponding weight adjustment information is generated. This weight adjustment information may include the weight adjustment magnitude and the weight adjustment direction, including: For each anomaly type weight, if the anomaly type weight is greater than the preset weight threshold, the weight adjustment direction is set to increase; if the anomaly type weight is not greater than the preset weight threshold, the weight adjustment direction is set to decrease. The difference between the anomaly type weight and the preset weight threshold is obtained as the weight difference value. The weight adjustment range is set according to the absolute value of the weight difference value, where the absolute value of the weight difference value is proportional to the weight adjustment range.

[0059] As an example, based on the weights of each anomaly type, the output weights of the classification layer corresponding to each preset waveform anomaly type in the preset large model are adapted and adjusted, including: The weights of each anomaly type are normalized to obtain normalized weights. Each normalized weight is then assigned to the initial bias term of the classification layer corresponding to each preset waveform anomaly type in the preset large model, so as to adapt and adjust the output weights of each classification layer.

[0060] The preset large model can include multiple classification layers, and each classification layer corresponds one-to-one with a preset abnormal waveform type. The output weight of the classification head corresponding to each classification layer can be adjusted by adjusting the initial bias term of the classification layer.

[0061] Specifically, the weights of each anomaly type are normalized to obtain normalized weights; each normalized weight is assigned to the initial bias term of the classification layer corresponding to each preset waveform anomaly type in the preset large model to obtain the output weights of each classification layer after adaptation and adjustment, wherein the output weights of the classification layer can be the output weights of the classification head.

[0062] Step 504: Using the waveform image as input, perform waveform anomaly detection on the waveform in the waveform image using a preset large model.

[0063] As an example, in this embodiment, a preset large model with adjusted weights for each classifier can be input into the waveform image. The preset large model is then used to perform waveform anomaly detection on the waveform image, and the waveform anomaly detection result is output by the preset large model.

[0064] In this embodiment, the output weights of the classification heads corresponding to each preset waveform anomaly type in the preset large model can be adapted and adjusted according to the weights of each anomaly type. In this way, the preset large model will pay more attention to the preset waveform anomaly types with larger anomaly type weights during the waveform anomaly detection process. This achieves the purpose of adding extra guiding feature information to the preset large model. Thus, using a waveform image as input, the preset large model can perform waveform anomaly detection on the waveform in the waveform image. The preset large model is more likely to distinguish or identify similar waveform anomaly types, thereby improving the accuracy of waveform anomaly detection on the waveform in the waveform image.

[0065] In some embodiments, waveform anomaly detection is performed on the waveform in the waveform image based on the waveform image and waveform anomaly type weight information, including: Based on the waveform image, waveform anomalies are classified in the waveform image to obtain the first classification probability distribution of each preset waveform anomaly type; based on the anomaly type weight information, the first classification probability distribution is weighted to obtain the second classification probability distribution.

[0066] In this embodiment, a preset large model or a conventional neural network model can be used for waveform anomaly detection. The result of waveform anomaly detection can be a probability distribution corresponding to each preset waveform anomaly type. This probability distribution can include the predicted probability value corresponding to each preset waveform anomaly type. The larger the predicted probability value, the higher the probability that the waveform in the waveform image belongs to the preset waveform anomaly type corresponding to the predicted probability value. The anomaly type weight information includes the anomaly type weight corresponding to each preset waveform anomaly type.

[0067] Specifically, based on the waveform image, the waveform in the waveform image is classified into waveform anomalies using a waveform anomaly detection model to obtain the first classification probability distribution of each preset waveform anomaly type; according to the weight of each anomaly type, the predicted probability values ​​corresponding to each preset waveform anomaly type in the first classification probability distribution are weighted to obtain the second classification probability distribution; thus, based on the second classification probability distribution, the probability of the waveform in the waveform image belonging to each preset waveform anomaly type can be determined.

[0068] In this embodiment, waveform anomaly weight information can be directly applied to the first classification probability distribution output by the waveform anomaly detection model. This allows waveform anomaly detection to be performed based on additional guiding feature information and waveform images, making it easier for the waveform anomaly detection model to distinguish or identify similar waveform anomaly types. This improves the accuracy of waveform anomaly detection in waveform images.

[0069] It should be noted that the above embodiments / implementations are only used to assist in understanding this application and do not constitute a limitation on the waveform anomaly detection method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0070] In addition, please refer to Figure 5 , Figure 5 This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the waveform anomaly detection method in this application embodiment.

[0071] The aforementioned electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the waveform anomaly detection method in the above embodiments.

[0072] The following is for reference. Figure 5 The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of this application. The electronic devices in these embodiments may include, but are not limited to, mobile terminals such as computers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0073] like Figure 5 As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays, speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tape, hard disks, etc.; and communication devices 1009. The communication device 1009 allows the electronic device to communicate wirelessly or wiredly with other devices to exchange data. Although the diagrams show electronic devices with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented alternatively.

[0074] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0075] The electronic device provided in this application, employing the waveform anomaly detection method in the above embodiments, can solve the technical problem of low accuracy in waveform anomaly detection. Compared with related technologies, the beneficial effects of the electronic device provided in this application are the same as those of the waveform anomaly detection method provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0076] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0077] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the above claims.

[0078] In addition, this application also provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to perform the steps of the waveform anomaly detection method in the above embodiments.

[0079] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, radio frequency, etc., or any suitable combination thereof.

[0080] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0081] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by an electronic device, the electronic device causes the electronic device to: acquire a waveform image and multiple waveform index parameters in the waveform image; generate waveform anomaly type weight information corresponding to each preset waveform anomaly type based on the multiple waveform index parameters; and perform waveform anomaly detection on the waveform in the waveform image based on the waveform image and the waveform anomaly type weight information.

[0082] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer, for example, via the Internet using an Internet service provider.

[0083] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0084] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0085] The computer-readable storage medium provided in this application stores computer-readable program instructions for performing the steps of the waveform anomaly detection method in the above embodiments, which can solve the technical problem of low accuracy in waveform anomaly detection. Compared with related technologies, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the waveform anomaly detection method provided in the above embodiments, and will not be repeated here.

[0086] Furthermore, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the waveform anomaly detection method described in the above embodiments.

[0087] The computer program product provided in this application can solve the technical problem of low accuracy in waveform anomaly detection. Compared with related technologies, the beneficial effects of the computer program product provided in this application are the same as those of the waveform anomaly detection method provided in the above embodiments, and will not be repeated here.

[0088] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A waveform anomaly detection method, characterized in that, The method includes: Acquire a waveform image and multiple waveform index parameters in the waveform image; Based on the multiple waveform index parameters, generate waveform anomaly type weight information corresponding to each preset waveform anomaly type; Based on the waveform image and the waveform anomaly type weight information, waveform anomaly detection is performed on the waveform in the waveform image.

2. The waveform anomaly detection method according to claim 1, characterized in that, The step of generating waveform anomaly type weight information corresponding to each preset waveform anomaly type based on the multiple waveform index parameters includes: Determine the target indicator parameter corresponding to each preset waveform anomaly type from the plurality of waveform indicator parameters; Based on the target index parameters corresponding to each preset waveform anomaly type, anomaly type weights are generated for each preset waveform anomaly type to obtain waveform anomaly type weight information.

3. The waveform anomaly detection method according to claim 2, characterized in that, The step of generating anomaly type weights for each preset waveform anomaly type based on the target index parameters corresponding to each preset waveform anomaly type includes: For each of the aforementioned preset waveform anomaly types: Determine the target index parameters corresponding to the preset waveform anomaly type, and the standard index parameters corresponding to each target index parameter; Based on the target index parameters and the standard index parameters, anomaly type weights corresponding to the preset waveform anomaly types are generated.

4. The waveform anomaly detection method according to claim 3, characterized in that, The step of generating anomaly type weights corresponding to the preset waveform anomaly types based on the target index parameters and the standard index parameters includes: Determine the standard library corresponding to each of the standard indicator parameters, and the priority of the standard library corresponding to each of the standard indicator libraries; Determine the deviation value between each of the target index parameters and the corresponding standard index parameters; Based on the deviation values ​​and the standard library priorities corresponding to the respective indicator standard libraries, anomaly type weights corresponding to the preset waveform anomaly types are generated.

5. The waveform anomaly detection method according to claim 1, characterized in that, The step of detecting waveform anomalies in the waveform image based on the waveform image and the waveform anomaly type weight information includes: Based on the waveform anomaly type weight information, the detection priority of each preset waveform anomaly type is determined; Based on the detection priority, generate corresponding anomaly detection prompts; Using the waveform image and the anomaly detection prompt as input, a preset large model is used to perform waveform anomaly detection on the waveform in the waveform image.

6. The waveform anomaly detection method according to claim 1, characterized in that, The waveform anomaly type weight information includes the anomaly type weight corresponding to each preset waveform anomaly type; The step of detecting waveform anomalies in the waveform image based on the waveform image and the waveform anomaly type weight information includes: Based on the weights of each of the aforementioned anomaly types, the output weights of the classification heads corresponding to each of the aforementioned preset waveform anomaly types in the preset large model are adapted and adjusted respectively. Using the waveform image as input, the preset large model is used to detect waveform anomalies in the waveform image.

7. The waveform anomaly detection method according to claim 6, characterized in that, The step of adapting and adjusting the classification layer output weights corresponding to each preset waveform anomaly type in the preset large model according to the weights of each anomaly type includes: The weights of each of the aforementioned anomaly types are normalized to obtain the normalized weights. Each of the normalized weights is assigned to the initial bias term of the classification layer corresponding to each of the preset waveform anomaly types in the preset large model, so as to adapt and adjust the output weights of each of the classification layers.

8. The waveform anomaly detection method according to claim 1, characterized in that, The step of detecting waveform anomalies in the waveform image based on the waveform image and the waveform anomaly type weight information includes: Based on the waveform image, the waveforms in the waveform image are classified into waveform anomalies to obtain the first classification probability distribution of each preset waveform anomaly type; Based on the anomaly type weight information, the first classification probability distribution is weighted to obtain the second classification probability distribution.

9. The waveform anomaly detection method according to claim 1, characterized in that, Obtaining multiple waveform index parameters from the waveform image, including: Perform text recognition on the waveform image to obtain multiple waveform index parameters; or Multiple waveform parameters are read from the oscilloscope that outputs the waveform image through a preset communication interface.

10. An electronic device, characterized in that, The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed, implements the steps of the waveform anomaly detection method as described in any one of claims 1-9.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the waveform anomaly detection method as described in any one of claims 1 to 9.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the waveform anomaly detection method as described in any one of claims 1 to 9.

Images