Belt foreign matter dynamic analysis method and device fusing time-frequency characteristics
Through the dynamic analysis method of belt foreign objects with time-frequency characteristics, combined with vibration, infrared and visual information, a comprehensive and detailed analysis of foreign objects during belt transportation is achieved, which improves the recognition accuracy and response speed, and solves the complexity of traditional monitoring methods. Problems of insufficient recognition capabilities in the environment.
Patent Information
- Application Number
- CN202411397684.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-09
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional belt foreign object monitoring methods have limited recognition capabilities in complex industrial environments, and have high false alarm rates and missed detection rates, making it difficult to identify foreign object characteristics in dynamic changes in real time and accurately.
The dynamic analysis method of belt foreign body with fusion time-frequency characteristics is adopted. By obtaining vibration information, infrared radiation images and RGB images of the vibration sensor, the vibration frequency domain, infrared frequency domain and time domain characteristics are extracted, and the training classification is used. The device performs foreign objects classification.
It improves the accuracy of foreign object recognition and the system's adaptability to complex industrial environments, and can respond as soon as foreign objects appear, effectively avoiding accidents.
Smart Images

Figure CN118916796A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of foreign body detection, and in particular to a belt foreign body dynamic analysis method integrating time-frequency features and a belt foreign body dynamic analysis device integrating time-frequency features. Background Art
[0002] In heavy industries such as mining, coal, electricity, and metallurgy, the belt conveyor system is not only the artery for efficient material flow, but also the indispensable core to maintain the continuous operation of the production line. However, with the expansion of production scale and the complexity of the operating environment, the problem of foreign matter mixing in the belt transportation process has become increasingly prominent, becoming a hidden danger that threatens equipment safety, affects production efficiency, and even causes major safety accidents. Although traditional belt foreign matter monitoring methods can play a role to a certain extent, their limitations are becoming increasingly obvious.
[0003] Traditional monitoring methods mostly rely on a single sensor technology. For example, weight sensors are mainly used to detect weight changes of materials on the belt, but their recognition ability is limited for foreign objects with irregular shapes and uneven weight distribution. Infrared sensors detect the presence of foreign objects by detecting infrared radiation from objects, but in dusty and complex light environments, their false alarm and missed detection rates are high. In addition, although simple image processing technology can capture visual information on the belt, it is often limited by computing power and algorithm complexity, making it difficult to accurately identify the characteristics of dynamically changing foreign objects in real time. Summary of the invention
[0004] The object of the present invention is to provide a belt foreign body dynamic analysis method integrating time-frequency characteristics to solve at least one of the above-mentioned technical problems.
[0005] One aspect of the present invention provides a belt foreign body dynamic analysis method integrating time-frequency features, and the belt foreign body dynamic analysis method integrating time-frequency features comprises: Obtain vibration information transmitted by the vibration sensor; Acquire infrared radiation images and RGB images; Extracting the vibration frequency domain features of the vibration information and extracting the infrared frequency domain features of the infrared radiation image; Extracting time domain features of the RGB image; Fusing the vibration frequency domain features, the infrared frequency domain features and the time domain features to obtain fused time-frequency features; Obtaining a trained foreign object classifier; The fused time-frequency features are input into the trained classifier to obtain a foreign body classification result.
[0006] Optionally, before extracting the frequency domain features of the vibration information, the belt foreign body dynamic analysis method integrating time-frequency features further comprises: The vibration information transmitted by the vibration sensor is filtered and denoised using the following formula: Vibration signal filtering and denoising: for ; Where n is the sample index, x[n] is the sampled signal, h[n] is the impulse response of the filter, h[nk] is the value of h[n] when n takes the value nk, N is the length of x[n], and M is the length of h[n].
[0007] Optionally, before extracting the time domain features of the RGB image, the belt foreign body dynamic analysis method integrating time-frequency features further comprises: The infrared radiation image is subjected to image enhancement processing to improve the contrast between the foreign matter and the background.
[0008] Optionally, the following formula is used for image enhancement: Infrared radiation image enhancement: , Where I is the image pixel value and C is the enhanced image pixel value.
[0009] Optionally, extracting the vibration frequency domain features of the vibration information includes: Perform fast Fourier transform on the signal collected by the vibration sensor to convert the time domain signal into the frequency domain; The infrared frequency domain features of the infrared radiation image are extracted including: The ViT model is used to perform frequency domain analysis on infrared radiation images and identify the frequency distribution patterns in the images.
[0010] Optionally, fusing the vibration frequency domain feature, the infrared frequency domain feature and the time domain feature to obtain a fused time-frequency feature includes: Design a fusion layer to concatenate or weightedly fuse the extracted time domain features with the frequency domain features, and optimize the fusion weights through training to ensure that the fused features can fully reflect the physical effects and visual characteristics caused by foreign objects.
[0011] Optionally, the trained foreign body classifier is a random forest classifier.
[0012] Optionally, the foreign matter classification result includes having foreign matter; The belt foreign body dynamic analysis method integrating time-frequency features further comprises: When the foreign matter classification result is that there is foreign matter, an alarm signal is generated.
[0013] The present application also provides a belt foreign body dynamic analysis device integrating time-frequency features, and the belt foreign body dynamic analysis device integrating time-frequency features comprises: A vibration information acquisition module, wherein the vibration information acquisition module is used to acquire vibration information transmitted by the vibration sensor; An image acquisition module, wherein the image acquisition module is used to acquire infrared radiation images and RGB images; A frequency domain feature extraction module, the frequency domain feature extraction module is used to extract the vibration frequency domain features of the vibration information and extract the infrared frequency domain features of the infrared radiation image; A time domain feature extraction module, wherein the time domain feature extraction module is used to extract the time domain features of the RGB image; A fused time-frequency feature acquisition module, which is used to fuse the vibration frequency domain features, the infrared frequency domain features and the time domain features, so as to obtain fused time-frequency features; A classifier acquisition module, wherein the classifier acquisition module is used to acquire a trained foreign body classifier; A classification module is used to input the fused time-frequency features into the trained classifier to obtain a foreign body classification result.
[0014] Beneficial effects: The belt foreign body dynamic analysis method of the present application integrating time-frequency features has the following advantages: This application achieves a more comprehensive and detailed analysis of foreign objects in the belt transportation process by deeply integrating time domain and frequency domain features. Time domain features can capture direct physical effects such as belt vibration and speed changes caused by foreign objects, while frequency domain features reveal deep information such as periodicity and frequency components behind these changes. This cross-domain feature fusion not only improves the accuracy of foreign object identification, but also enhances the system's adaptability to complex industrial environments.
[0015] The dynamic frequency domain analysis of this application shows unique advantages in video processing. By analyzing the frequency domain information in the belt transport video in real time, the system can dynamically capture subtle signals such as tiny vibrations and sound changes caused by foreign objects, and convert them into recognizable characteristic patterns. This processing method not only improves the sensitivity and accuracy of video monitoring, but also greatly shortens the response time, allowing the system to respond as soon as foreign objects appear, effectively avoiding accidents. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 It is a flow chart of a belt foreign body dynamic analysis method integrating time-frequency features according to an embodiment of the present application; Figure 2 It is used to realize Figure 1The electronic device schematic diagram of the belt foreign body dynamic analysis method integrating time-frequency characteristics is shown. DETAILED DESCRIPTION
[0017] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the embodiment of this application will be described in more detail below in conjunction with the drawings in the embodiment of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of this application, not all of them. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain this application, and should not be construed as limitations on this application. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The embodiments of this application are described in detail below in conjunction with the drawings.
[0018] Figure 1 It is a flow chart of a belt foreign body dynamic analysis method integrating time-frequency features according to an embodiment of the present application.
[0019] like Figure 1 The belt foreign body dynamic analysis method integrating time-frequency characteristics shown includes: Step 1: Obtain the vibration information transmitted by the vibration sensor; Step 2: Obtain infrared radiation image and RGB image; Step 3: extracting the vibration frequency domain features of the vibration information and extracting the infrared frequency domain features of the infrared radiation image; Step 4: extracting the time domain features of the RGB image; Step 5: Fusing the vibration frequency domain features, infrared frequency domain features and the time domain features to obtain fused time-frequency features; Step 6: Obtain the trained foreign body classifier; Step 7: Input the fused time-frequency features into the trained classifier to obtain foreign body classification results.
[0020] The belt foreign body dynamic analysis method of the present application integrating time-frequency features has the following advantages: This application achieves a more comprehensive and detailed analysis of foreign objects in the belt transportation process by deeply integrating time domain and frequency domain features. Time domain features can capture direct physical effects such as belt vibration and speed changes caused by foreign objects, while frequency domain features reveal deep information such as periodicity and frequency components behind these changes. This cross-domain feature fusion not only improves the accuracy of foreign object identification, but also enhances the system's adaptability to complex industrial environments.
[0021] The dynamic frequency domain analysis of this application shows unique advantages in video processing. By analyzing the frequency domain information in the belt transport video in real time, the system can dynamically capture subtle signals such as tiny vibrations and sound changes caused by foreign objects, and convert them into recognizable characteristic patterns. This processing method not only improves the sensitivity and accuracy of video monitoring, but also greatly shortens the response time, allowing the system to respond as soon as foreign objects appear, effectively avoiding accidents.
[0022] In this embodiment, before extracting the frequency domain features of the vibration information, the belt foreign body dynamic analysis method integrating time-frequency features further includes: The vibration information transmitted by the vibration sensor is filtered and denoised using the following formula: Vibration signal filtering and denoising: for ; Where n is the sample index, x[n] is the sampled signal, h[n] is the impulse response of the filter, h[nk] is the value of h[n] when n takes the value nk, N is the length of x[n], and M is the length of h[n].
[0023] In this embodiment, before extracting the time domain features of the RGB image, the belt foreign body dynamic analysis method integrating time-frequency features further includes: The infrared radiation image is subjected to image enhancement processing to improve the contrast between the foreign matter and the background.
[0024] In this embodiment, the following formula is used for image enhancement: Infrared radiation image enhancement: , Where I is the image pixel value and C is the enhanced image pixel value.
[0025] In this embodiment, extracting the vibration frequency domain features of the vibration information includes: Perform fast Fourier transform on the signal collected by the vibration sensor to convert the time domain signal into the frequency domain; The infrared frequency domain features of the infrared radiation image are extracted including: The ViT model is used to perform frequency domain analysis on infrared radiation images and identify the frequency distribution patterns in the images.
[0026] In this embodiment, fusing the vibration frequency domain features, the infrared frequency domain features, and the time domain features to obtain the fused time-frequency features includes: Design a fusion layer to concatenate or weightedly fuse the extracted time domain features with the frequency domain features, and optimize the fusion weights through training to ensure that the fused features can fully reflect the physical effects and visual characteristics caused by foreign objects.
[0027] In this embodiment, the trained foreign body classifier is a random forest classifier.
[0028] In this embodiment, the foreign matter classification result includes having foreign matter; The belt foreign body dynamic analysis method integrating time-frequency features further comprises: When the foreign matter classification result is that there is foreign matter, an alarm signal is generated.
[0029] The present application is further described in detail below by way of examples. It should be understood that the examples do not constitute any limitation to the present application.
[0030] The multi-dimensional data of the belt transportation process is collected in real time through a multi-sensor array (including but not limited to vibration sensors, infrared sensors, high-definition cameras, etc.) installed on the belt transportation system. These data include belt vibration signals, infrared radiation images, and high-definition video streams.
[0031] In this embodiment, the time domain signal collected by the vibration sensor needs to be filtered and denoised to remove the interference of environmental noise on the signal quality.
[0032] Vibration signal filtering and denoising: for ; Where n is the sample index, x[n] is the sampled signal, h[n] is the impulse response of the filter, N is the length of x[n], and M is the length of h[n].
[0033] In this embodiment, the infrared radiation image needs to be subjected to image enhancement processing to improve the contrast between the foreign matter and the background.
[0034] The high-definition video stream needs to be frame synchronized and cropped to ensure that each frame covers the entire belt area and removes irrelevant background.
[0035] Infrared radiation image enhancement: , Where I is the image pixel value and C is the enhanced image pixel value.
[0036] S2: Feature extraction and fusion (time-frequency feature fusion based on ViT model) The Vision Transformer (ViT) model is used to extract the time domain features and frequency domain features in the belt conveyor video respectively, and the deep fusion of these two features is achieved through a specific strategy.
[0037] In this embodiment, when extracting time domain features, the preprocessed high-definition video stream is divided into multiple segments according to the time series. Each segment is used as the input of the ViT model, and the time domain features in the key frames, such as motion trajectory, speed change, etc., are extracted through the self-attention mechanism.
[0038] enter: , Where X_video represents the preprocessed high-definition video stream, which is divided into T segments, and the segment size is: H (height) × W (width) × C (number of channels).
[0039] Output: , Where F_temporal represents the temporal features extracted by the ViT model, and D is the dimension of the feature vector.
[0040] In this embodiment, when extracting frequency domain features, a fast Fourier transform (FFT) is performed on the signal collected by the vibration sensor to convert the time domain signal into the frequency domain and analyze the spectrum characteristics of the signal. At the same time, the ViT model is used to perform frequency domain analysis on the infrared radiation image to identify the frequency distribution pattern in the image.
[0041] Vibration signal: , Infrared image features: , Where F_infrared is the extracted feature vector, and the convolutional neural network ViT model is used for feature extraction. I_infrareds is the infrared radiation image, and D' is the dimension of the feature vector.
[0042] In this embodiment, a fusion layer is designed to concatenate or weightedly fuse the extracted time domain features and frequency domain features. The fusion weights are optimized through training to ensure that the fused features can fully reflect the physical effects and visual features caused by foreign matter.
[0043] Fusion features: , The feature fusion here is the feature vector concatenation operation. and It is a learnable fusion weight coefficient, which is optimized through the training process.
[0044] S3: Dynamic frequency domain analysis and anomaly detection Based on the fused features, dynamic frequency domain analysis technology is used to monitor frequency changes during belt transportation in real time to identify and locate foreign objects.
[0045] In this embodiment, in terms of dynamic spectrum monitoring, the frequency domain part of the fusion feature is continuously monitored to analyze the dynamic changes of the frequency components. By comparing the difference between the current spectrum and the historical spectrum, abnormal frequency fluctuations are identified.
[0046] In this embodiment, an anomaly detection algorithm method is used to design an anomaly detection algorithm based on machine learning, and a classifier (random forest) is trained using fusion features to achieve automatic classification and positioning of foreign objects. The algorithm needs to have high sensitivity and low false alarm rate to ensure accurate identification of foreign objects in complex industrial environments.
[0047] In this embodiment, in terms of the real-time response mechanism, once a foreign object is detected, the system immediately triggers an alarm and displays the location, type and other information of the foreign object through a visual interface. At the same time, the emergency response mechanism can be automatically activated, such as slowing down the belt speed, starting the cleaning device, etc., to reduce the damage of foreign objects to the equipment.
[0048] S4: Result verification and optimization The effectiveness and reliability of the present invention are verified through simulation experiments and actual production environment tests, and the algorithm is optimized and adjusted according to the test results.
[0049] In this embodiment, during the simulation experiment stage, a belt conveyor system simulation platform is built in a laboratory environment to simulate the mixing of different types of foreign matter and test indicators such as the recognition accuracy and response speed of the present invention.
[0050] In this embodiment, during the actual production environment testing phase, the present invention is deployed in the actual production environment of industries such as mining, coal, electricity, and metallurgy, and long-term operation tests are performed to collect feedback data for algorithm optimization.
[0051] In this embodiment, during the algorithm optimization phase, the parameter settings, fusion strategy, anomaly detection algorithm, etc. of the ViT model are adjusted according to the test results to improve the overall performance of the system. At the same time, attention is paid to the development of new technologies and advanced technologies are promptly integrated into the implementation of the present invention.
[0052] The present application also provides a belt foreign body dynamic analysis device integrating time-frequency features, the belt foreign body dynamic analysis device integrating time-frequency features comprises a vibration information acquisition module, an image acquisition module, a frequency domain feature extraction module, a time domain feature extraction module, a fused time-frequency feature acquisition module, a classifier acquisition module and a classification module, wherein: The vibration information acquisition module is used to acquire the vibration information transmitted by the vibration sensor; The image acquisition module is used to acquire infrared radiation images and RGB images; The frequency domain feature extraction module is used to extract the vibration frequency domain features of the vibration information and to extract the infrared frequency domain features of the infrared radiation image; The time domain feature extraction module is used to extract the time domain features of the RGB image; The fused time-frequency feature acquisition module is used to fuse the vibration frequency domain feature, the infrared frequency domain feature and the time domain feature, so as to obtain the fused time-frequency feature; The classifier acquisition module is used to obtain the trained foreign body classifier; The classification module is used to input the fused time-frequency features into the trained classifier to obtain the foreign body classification result.
[0053] For underground production, the application of this technology is of epoch-making significance. The underground environment is complex and changeable, with dim light and dust everywhere, and traditional monitoring methods are often unable to cope with it. The dynamic analysis method of belt foreign matter that integrates time-frequency characteristics can work stably in harsh underground environments with its high sensitivity, low false alarm rate, and rapid response, providing strong guarantees for safe production in mines. At the same time, the real-time nature of this technology also ensures the continuous operation of the production line, reduces downtime caused by foreign matter, and improves production efficiency and economic benefits.
[0054] In summary, the belt foreign body dynamic analysis method integrating time-frequency characteristics proposed in the present invention is not only a major innovation in traditional monitoring technology, but also an important step in promoting underground production and safety towards intelligence and efficiency.
[0055] It should be noted that the above explanations of the method embodiment are also applicable to the device of this embodiment and will not be repeated here.
[0056] The present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the computer program, the above-mentioned belt foreign body dynamic analysis method integrating time-frequency features is implemented.
[0057] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it can implement the above-mentioned belt foreign body dynamic analysis method integrating time-frequency features.
[0058] Figure 2 It is used to realize Figure 1 The electronic device schematic diagram of the belt foreign body dynamic analysis method integrating time-frequency characteristics is shown.
[0059] like Figure 2As shown, the electronic device includes an input device 501, an input interface 502, a central processor 503, a memory 504, an output interface 505, and an output device 506. The input interface 502, the central processor 503, the memory 504, and the output interface 505 are interconnected through a bus 507, and the input device 501 and the output device 506 are connected to the bus 507 through the input interface 502 and the output interface 505, respectively, and then connected to other components of the electronic device. Specifically, the input device 501 receives input information from the outside, and transmits the input information to the central processor 503 through the input interface 502; the central processor 503 processes the input information based on the computer executable instructions stored in the memory 504 to generate output information, temporarily or permanently stores the output information in the memory 504, and then transmits the output information to the output device 506 through the output interface 505; the output device 506 outputs the output information to the outside of the electronic device for use by the user.
[0060] That is to say, Figure 2 The electronic device shown can also be implemented as comprising: a memory storing computer executable instructions; and one or more processors, which can implement the combination when executing the computer executable instructions. Figure 1 A dynamic analysis method for foreign bodies in belts by integrating time-frequency features is described.
[0061] In one embodiment, Figure 2 The electronic device shown can be implemented to include: a memory 504 configured to store executable program code; and one or more processors configured to run the executable program code stored in the memory 504 to execute the belt foreign body dynamic analysis method integrating time-frequency features in the above-mentioned embodiment.
[0062] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0063] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
[0064] Computer readable media include permanent and non-permanent, removable and non-removable, and the media can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), data versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.
[0065] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a module, a program segment or a part of a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes identified in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or the total flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0066] The processor referred to in this embodiment may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0067] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the device / terminal equipment by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0068] In this embodiment, if the module / unit integrated in the device / terminal equipment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. Although the present application is disclosed as above in the form of a preferred embodiment, it is not actually intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined in the present application.
[0069] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0070] In addition, it is obvious that the word "comprising" does not exclude other units or steps. Multiple units, modules or devices stated in the device can also be implemented by one unit or the entire device through software or hardware.
[0071] Although the present invention has been described in detail above with general descriptions and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements may be made thereto based on the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection claimed by the present invention.
Claims
1. A belt foreign body dynamic analysis method integrating time-frequency characteristics, characterized in that: The belt foreign body dynamic analysis method integrating time-frequency characteristics comprises: Obtain vibration information transmitted by the vibration sensor; Acquire infrared radiation images and RGB images; Extracting the vibration frequency domain features of the vibration information and extracting the infrared frequency domain features of the infrared radiation image; Extracting time domain features of the RGB image; Fusing the vibration frequency domain features, the infrared frequency domain features and the time domain features to obtain fused time-frequency features; Obtaining a trained foreign object classifier; The fused time-frequency features are input into the trained classifier to obtain a foreign body classification result.
2. The belt foreign body dynamic analysis method integrating time-frequency features as claimed in claim 1 is characterized in that: Before extracting the frequency domain features of the vibration information, the belt foreign body dynamic analysis method integrating time-frequency features further includes: The vibration information transmitted by the vibration sensor is filtered and denoised using the following formula: Vibration signal filtering and denoising: for ; Where n is the sample index, x[n] is the sampled signal, h[n] is the impulse response of the filter, h[nk] is the value of h[n] when n takes the value nk, N is the length of x[n], and M is the length of h[n].
3. The belt foreign body dynamic analysis method integrating time-frequency features as claimed in claim 2 is characterized in that: Before extracting the time domain features of the RGB image, the belt foreign body dynamic analysis method integrating time-frequency features further comprises: The infrared radiation image is subjected to image enhancement processing to improve the contrast between the foreign matter and the background.
4. The belt foreign body dynamic analysis method integrating time-frequency features as claimed in claim 3 is characterized in that: The following formula is used for image enhancement: Infrared radiation image enhancement: , Where I is the image pixel value and C is the enhanced image pixel value.
5. The belt foreign body dynamic analysis method integrating time-frequency features as claimed in claim 4 is characterized in that: The extracting of the vibration frequency domain features of the vibration information comprises: Perform fast Fourier transform on the signal collected by the vibration sensor to convert the time domain signal into the frequency domain; The infrared frequency domain features of the infrared radiation image are extracted including: The ViT model is used to perform frequency domain analysis on infrared radiation images and identify the frequency distribution patterns in the images.
6. The belt foreign body dynamic analysis method integrating time-frequency features as claimed in claim 5 is characterized in that: The fusing of the vibration frequency domain features, the infrared frequency domain features and the time domain features to obtain the fused time-frequency features includes: Design a fusion layer to concatenate or weightedly fuse the extracted time domain features with the frequency domain features, and optimize the fusion weights through training to ensure that the fused features can fully reflect the physical effects and visual characteristics caused by foreign objects.
7. The belt foreign body dynamic analysis method integrating time-frequency features as described in any one of claims 1 to 6, characterized in that: The trained foreign body classifier is a random forest classifier.
8. The belt foreign body dynamic analysis method integrating time-frequency features as claimed in claim 7 is characterized in that: The foreign body classification result includes having foreign body; The belt foreign body dynamic analysis method integrating time-frequency features further comprises: When the foreign matter classification result is that there is foreign matter, an alarm signal is generated.
9. A belt foreign body dynamic analysis device integrating time-frequency characteristics, characterized in that: The belt foreign body dynamic analysis device integrating time-frequency characteristics comprises: A vibration information acquisition module, wherein the vibration information acquisition module is used to acquire vibration information transmitted by the vibration sensor; An image acquisition module, wherein the image acquisition module is used to acquire infrared radiation images and RGB images; A frequency domain feature extraction module, the frequency domain feature extraction module is used to extract the vibration frequency domain features of the vibration information and extract the infrared frequency domain features of the infrared radiation image; A time domain feature extraction module, wherein the time domain feature extraction module is used to extract the time domain features of the RGB image; A fused time-frequency feature acquisition module, which is used to fuse the vibration frequency domain features, the infrared frequency domain features and the time domain features, so as to obtain fused time-frequency features; A classifier acquisition module, wherein the classifier acquisition module is used to acquire a trained foreign body classifier; A classification module is used to input the fused time-frequency features into the trained classifier to obtain a foreign body classification result.
Citation Information
Patent Citations
Railway track foreign matter detection device and method
CN109784214A
Equipment and method for foreign matter detection and pavement detection of airport runway
CN115079293A
Foreign matter detection method and device for underground belt conveyor
CN115601682A
Test stage training face forgery detection method and system containing uncertainty guidance
CN117275068A
Airport runway vibration monitoring system and method
CN117664225A
