Tailless threaded sleeve deformation detection system and method based on multispectral imaging

Abstract

The invention relates to the technical field of tailless thread sleeve detection, and provides a multispectral imaging-based tailless thread sleeve deformation detection system and method, and the method comprises the steps: S1, collecting reflection maps of tailless thread sleeve standard parts and deformation flaw parts of different models under different spectral bands through a multispectral imaging device, and constructing a data set according to model classification; s2, carrying out preprocessing and feature extraction on the obtained reflection spectrum; s3, constructing a recognition model based on a convolutional neural network (CNN); s4, verifying the accuracy of the constructed recognition model by using a verification set; s5, updating the trained recognition model to a calculation unit; s6, adding newly added detection data into the training set regularly, retraining the model and updating the model to the calculation unit; according to the invention, various deformation types of the tailless thread sleeve can be rapidly identified, the calculation amount is small, and the detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of tailless screw sleeve detection, and in particular to a tailless screw sleeve deformation detection system and method based on multispectral imaging. Background Art

[0002] A tailless insert, also known as a shankless wire thread insert, is a type of wire thread insert (threaded insert). Each end of the insert features a hook-shaped slot, known as an "inverted shark fin slot." This structure eliminates the need to determine the orientation of the insert during installation, significantly improving work efficiency.

[0003] Tailless screw inserts are widely used in aerospace, automotive manufacturing and other fields, and their deformation directly affects assembly accuracy and product performance. Traditionally, due to differences in processing equipment and processes, various defects are inevitable. Common problems include axial bending, surface cracks, and dimensional deviations. Therefore, it is necessary to inspect the processed screw inserts to reduce the defect rate. Conventional inspection methods involve manual inspection or taking photos with an optical CCD camera. Manual inspection is inefficient and only suitable for spot checks. Optical CCD cameras are a non-contact visual inspection method that can only detect surface defects on the workpiece, but cannot promptly detect problems such as local bending deformation and internal cracks in the screw insert.

[0004] Therefore, there is an urgent need for an efficient and convenient detection method to detect the deformation of the workpiece in order to improve the yield rate of the product when it is shipped. Summary of the Invention

[0005] The embodiments of the present invention provide a system and method for detecting deformation of a tailless screw sleeve based on multispectral imaging, which uses multispectral imaging to identify potential defects of the tailless screw sleeve and mark them, and cooperates with external equipment to remove defective products to improve the product yield.

[0006] Other features and advantages of the present invention will become apparent from the following detailed description, or may be learned in part by practice of the present invention.

[0007] According to one aspect of the present invention, a tailless screw sleeve deformation detection system based on multispectral imaging is provided, comprising a detection platform, a multispectral imaging device, an image processing unit, a computing unit, a data storage unit and a communication unit; The testing platform is built on the transmission production line of the tailless screw sleeve to provide a support platform for the testing equipment to be easily installed and fixed; The multispectral imaging device is used to image the tailless screw sleeve passing through the detection platform and collect image data in different bands; The image processing unit is used to preprocess and extract features from the collected images; The calculation unit is used to calculate the extracted features and determine whether the tailless screw sleeve to be tested has deformation; The data storage unit is used to save the test results and build a historical database; The communication unit is used to communicate with the host computer to feedback the test results and receive the control instructions from the host computer; The tailless screw sleeves circulating on the transmission production line are imaged using multispectral imaging equipment, and image data in different bands are collected. After processing the image data, feature recognition is performed, and the trained neural network model is used in the computing unit to perform feature calculations to determine whether there is deformation.

[0008] Preferably, the tailless screw sleeve is transported on the transmission production line using a customized carrier, and is transported to the detection platform and paused for a period of time for the multi-spectral imaging device to image the tailless screw sleeve.

[0009] Preferably, the spectral band of the multispectral imaging device is 400-1100nm; the multispectral imaging device uses incident light of different spectral bands from low to high to hit the tailless screw sleeve to be detected, and collects an image composed of reflected light to generate a reflection spectrum.

[0010] Preferably, different data sets are established in the historical database for different models of tailless screw inserts; The reflection maps of the standard parts of this model and the reflection maps of the parts with deformation defects are collected in advance; the obtained reflection maps are used as training sets and feature recognition is performed using a neural network model.

[0011] A second aspect of the present invention provides a method for detecting deformation of a tailless screw sleeve based on multispectral imaging, which uses the above-mentioned system for detection, including the following specific steps: S1: Use multispectral imaging equipment to collect reflectance maps of different models of tailless screw sleeve standard parts and deformation defect parts in different spectral bands, and construct a data set according to model classification; S2: Preprocessing and feature extraction of the obtained reflectance map; S3: Construct a recognition model based on a convolutional neural network (CNN); the input layer receives fused feature data; the output layer outputs whether the tailless screw sleeve is "normal" or "deformed", as well as the location of the "deformation" and the specific defect type; S4: Use the validation set to verify the accuracy of the constructed recognition model; if the recognition accuracy is greater than 95%, the model training is considered complete; S5: The trained recognition model is updated to the computing unit, and deformation detection is performed on the produced tailless screw sleeve; the detection results are uploaded to the host computer; S6: Regularly add new detection data to the training set and use the transfer learning strategy to fine-tune the model parameters; when the model misjudgment rate exceeds the preset threshold, retrain the model and update it to the computing unit.

[0012] Preferably, in step S1, at least 6 spectral images of different wavelength bands are collected for each tailless screw sleeve to generate a reflectance spectrum.

[0013] Preferably, in step S2, the reflectance map is subjected to denoising, normalization and registration processing to eliminate ambient light interference and image distortion; and fusion features of the multispectral image are extracted, including texture features, geometric features and spectral reflectance differences.

[0014] Preferably, the network structure of the recognition model based on the convolutional neural network (CNN) constructed in step S3 includes an input layer, at least 4 convolutional layers, 2 pooling layers, 3 fully connected layers, and an output layer; The output layer uses the Softmax function; The CNN model was trained using a labeled dataset consisting of 5,000 images of normal and 3,000 images of deformed screw caps. The deformation types included axial bending, surface cracks, and dimensional deviations. Training parameters: learning rate 0.001, batch size 32, optimizer Adam, loss function cross entropy loss, training epochs 50.

[0015] A third aspect of the present invention provides a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the computer program implements the above-mentioned method for detecting deformation of a tailless screw sleeve based on multispectral imaging.

[0016] A fourth aspect of the present invention provides an electronic device, comprising: one or more processors; A storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned detection method in the tailless screw sleeve deformation detection system based on multispectral imaging.

[0017] According to one aspect of the present invention, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the detection method in the tailless screw sleeve deformation detection system based on multispectral imaging provided in the various optional implementations described above.

[0018] In the technical solution of the present invention, multispectral imaging technology is innovatively used to obtain the reflection spectrum of the tailless screw sleeve and construct a data set; the spectral curves of the characteristic pixel points in the constructed data set are selectively analyzed, and the spectral curves of the standard parts and the defective parts are trained to construct a recognition model; the recognition model is used to quickly identify the deformation problem of the tailless screw sleeve; this solution only analyzes the spectral curves of some pixel points in the reflection spectrum, and does not need to process the entire reflection spectrum, which greatly reduces the amount of calculation and improves the calculation efficiency; it is suitable for rapid detection in industrial production. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are incorporated into and constitute a part of this specification, illustrate embodiments consistent with the present invention, and together with the description, serve to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and it is clear that those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0020] Figure 1 The flowchart of the method for detecting deformation of a tailless screw sleeve based on multispectral imaging in one embodiment of the present invention is schematically shown.

[0021] Figure 2 The structure of a tailless screw sleeve deformation detection system based on multispectral imaging in one embodiment of the present invention is schematically shown.

[0022] Figure 3 A schematic structural diagram of a computer system suitable for implementing an electronic device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0023] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0024] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, it will be appreciated by those skilled in the art that the technical solutions of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.

[0025] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0026] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0027] Example 1 The implementation details of the technical solution of the present invention are described in detail below: Figure 2 A structural diagram of a tailless screw sleeve deformation detection system based on multispectral imaging according to an embodiment of the present invention is shown.

[0028] The present invention provides a tailless screw sleeve deformation detection system based on multispectral imaging, comprising a detection platform, a multispectral imaging device, an image processing unit, a calculation unit, a data storage unit and a communication unit; The detection platform is built on the transmission production line of the tailless screw sleeve, which is used to provide a support platform for the detection equipment to facilitate installation and fixation; the tailless screw sleeve is transported on the transmission production line using a customized carrier, and multiple tailless screw sleeves can be placed in the carrier in batches in a specific order; the carrier follows the transmission production line to the detection platform, at which time it can pause for a period of time to allow the multi-spectral imaging equipment to image the tailless screw sleeve placed on the customized carrier; the time is generally set to 1-2s; the specific parameters can be adjusted according to the number of tailless screw sleeves imaged at one time.

[0029] The multispectral imaging device is used to image the tailless screw sleeve passing through the detection platform and collect image data in different bands. In this embodiment, the spectral band of the multispectral imaging device is 400-1100nm. The multispectral imaging device uses incident light of different spectral bands from low to high to hit the tailless screw sleeve to be detected, and collects an image composed of reflected light to generate a reflection spectrum. The six different band spectra of 400, 500, 600, 800, 900 and 1100nm are preferentially used for imaging to obtain the reflection spectrum.

[0030] The image processing unit is used to preprocess and extract features from the collected images; different data sets are established in the historical database for different models of tailless screw sleeves; the reflection maps of standard parts of this model and the reflection maps of parts with deformation defects are collected in advance; the obtained reflection maps are used as training sets, and feature recognition is performed using a neural network model.

[0031] The calculation unit is used to calculate the extracted features and output whether the product is "normal" or "deformed". A judgment threshold is pre-set, and products exceeding the threshold are considered to have deformation and the specific location of the deformation is output. Products that do not exceed the threshold are judged to be "normal" and qualified. The data storage unit is used to save the test results and build a historical database; the historical database is divided into different storage units according to the product model; each storage unit stores the reflection spectrum data of the "normal" tailless screw sleeve and the reflection spectrum data of the "deformed" tailless screw sleeve; The communication unit is used to communicate with the host computer to feedback the detection results and receive control instructions from the host computer; the tailless screw sleeves with "deformation" are rejected, while the reflection spectrum data is retained to facilitate subsequent analysis of the cause of the deformation, optimize production materials and processes, and improve product yield.

[0032] In this embodiment, a multi-spectral imaging device is used to image the tailless screw sleeves circulating on the transmission production line, and image data in different bands is collected. After processing the image data, feature recognition is performed, and a trained neural network model is used in the computing unit to perform feature calculation to further determine whether there is deformation; thereby improving recognition efficiency.

[0033] Example 2 Figure 1 A flow chart of a method for detecting deformation of a tailless screw sleeve based on multispectral imaging according to an embodiment of the present invention is shown. The present invention provides a method for detecting deformation of a tailless screw sleeve based on multispectral imaging, which uses the system in Example 1 for detection, including the following specific steps: S1: Use a multispectral imaging device to collect reflection spectra of different models of tailless screw sleeve standard parts and deformation defect parts in different spectral bands, and construct a data set according to model classification; in step S1, at least 6 spectral images in different bands are collected for each tailless screw sleeve to generate a reflection spectrum; S2: Preprocessing and feature extraction of the obtained reflection map; in step S2, denoising, normalization and registration processing are performed on the reflection map to eliminate ambient light interference and image distortion; Select several pixels with significant features in the reflectance spectrum, obtain the spectral curve of the pixel, analyze the spectral curves of the same type of pixels and perform feature extraction to obtain the peak wavelength shift characteristics, spectral curve slope characteristics, and reflectivity difference characteristics between bands; Specifically, the peak wavelength shift feature indicates that the refractive index of the material itself has changed due to cracks or stress concentration in the identification area; The slope characteristics of the spectral curve indicate that the change rate of reflectivity with wavelength changes due to the increase in surface roughness in the identification area; The inter-band reflectivity difference characteristic refers to the ratio of the near-infrared to visible light bands. The abnormality of this ratio reflects the change of internal stress or density in the identified area.

[0034] Therefore, the presence of axial bending of the tailless screw sleeve to be tested can be determined by the peak wavelength shift characteristics and the reflectivity difference characteristics between the bands; the presence of surface cracks can be determined by the slope characteristics of the spectral curve and the reflectivity difference characteristics between the bands.

[0035] As for dimensional deviation, the ratio of the modulus length of the vector between the selected pixel points to the modulus length of the vector between the same pixel points of the standard part is calculated to determine whether there is a dimensional deviation phenomenon.

[0036] S3: Constructing a recognition model based on a convolutional neural network (CNN); the input layer receives fused feature data; the output layer outputs whether the tailless screw sleeve is "normal" or "deformed" as well as the location of the "deformation" and the specific defect type; the network structure of the recognition model based on the convolutional neural network (CNN) constructed in step S3 includes an input layer, at least 4 convolutional layers, 2 pooling layers, 3 fully connected layers, and an output layer; The output layer uses the Softmax function; The CNN model was trained using a labeled dataset consisting of 5,000 images of normal and 3,000 images of deformed screw caps. The deformation types included axial bending, surface cracks, and dimensional deviations. Training parameters: learning rate 0.001, batch size 32, optimizer Adam, loss function cross entropy loss, training epochs 50.

[0037] S4: Use the validation set to verify the accuracy of the constructed recognition model; if the recognition accuracy is greater than 95%, the model training is considered complete; S5: The trained recognition model is updated to the computing unit, and deformation detection is performed on the produced tailless screw sleeve; the detection results are uploaded to the host computer; S6: Regularly add new detection data to the training set and use the transfer learning strategy to fine-tune the model parameters; when the model misjudgment rate exceeds the preset threshold, retrain the model and update it to the computing unit.

[0038] The following uses the 2TNM-2.5X.45C-2.5 tailless screw insert as an example to introduce this solution in detail: A tailless threaded insert, also known as a shankless wire threaded insert, is a type of wire threaded insert (threaded insert). It is primarily used to repair damaged threaded holes. Generally, it can be installed manually, pneumatically, or with power tools for large-scale installation. It offers strong corrosion resistance, protecting the threaded hole and bolt from rust. Once installed in a threaded hole, it evenly distributes stress, expanding the thread's bearing surface, thereby improving the connection between the threads and increasing the load-bearing capacity and fatigue resistance of the threaded connection. Furthermore, locking-type tailless threaded inserts prevent the screw from loosening.

[0039] Common problems in the production and processing process, such as uneven material or hardness of the screw sleeve, will lead to axial bending and even cracks in severe cases due to uneven stress during the processing. In addition, due to the accumulation of errors in the mechanical processing equipment during the processing, dimensional deviations will occur. The above problems directly affect the product quality of the tailless screw sleeve, resulting in the inability to install normally and loose assembly of the screws. Conventional manual detection methods are inefficient, and although contactless detection methods such as CCD cameras can improve detection efficiency, they can only detect surface cracks and are not very good at identifying axial bending and dimensional deviations. In this embodiment, multispectral imaging technology is used to construct a data set by acquiring reflection maps of different bands. Pixels with significant features are selected from the reflection map to analyze the spectral curve, and the spectral curves of standard parts and defective parts are used for training to construct a recognition model. The recognition model is used to quickly identify the deformation problem of the tailless screw sleeve. This solution only analyzes the spectral curves of some pixels in the reflection map, and does not need to process the entire reflection map, which greatly reduces the amount of calculation and improves the calculation efficiency. It is suitable for rapid detection in industrial production.

[0040] The specific steps are as follows: 1: Use multispectral imaging equipment to collect reflectance maps of standard parts and deformed parts of this model of tailless screw sleeve in different spectral bands to construct a data set; the number of standard tailless screw sleeves is no less than 1000, and the number of deformed parts is no less than 300. The specific number is divided according to the probability of each fault occurring in the current enterprise production process; for example, 120 deformed parts with axial bending, 130 deformed parts with surface cracks, and 50 deformed parts with dimensional deviations are selected; at least 6 spectral images in different bands are collected for each tailless screw sleeve to generate a reflectance map; 2: Preprocessing and feature extraction of the obtained reflection map; in step S2, denoising, normalization and registration are performed on the reflection map to eliminate ambient light interference and image distortion; Several pixel points with significant features are selected in the reflectance spectrum, and the spectral curves of the pixel points are obtained. By analyzing the spectral curves of the same type of pixel points and performing feature extraction, the peak wavelength shift characteristics, spectral curve slope characteristics, and reflectivity difference characteristics between bands are obtained; the number of threads of the same model of tailless screw sleeves is the same, and the spacing between the threads also remains the same; when using a multispectral imaging device to image them, the imaging device is vertically installed directly above the detection platform; the vertex of the thread edge has more significant features than other points, and this point is preferentially selected as a pixel point for analyzing its spectral curve; other points on the thread edge, such as pixels on the side or inside the thread, can also be selected for analysis.

[0041] In this embodiment, specifically, the peak wavelength shift characteristic indicates that the refractive index of the material itself changes due to cracks or stress concentration in the identification area; the peak wavelength shift characteristic satisfies the following formula: Where, Represents the wavelength corresponding to the peak reflectivity of the standard tailless screw sleeve in a specific band. The wavelength corresponding to the reflectivity peak of the defective tailless screw sleeve in the same band, Represents the shift of the peak wavelength, with positive values ​​representing redshift and negative values ​​representing blueshift.

[0042] The slope characteristic of the spectral curve indicates that the reflectivity of the identification area changes with the wavelength due to the increase in surface roughness. The slope characteristic of the spectral curve satisfies the following formula: Where, , In this embodiment, the two independently selected band center wavelengths are selected as 550nm (green light) and 850nm (near infrared light). , Indicated by wavelength , The reflectivity value at the selected wavelength is S, which is the slope of the spectral curve in the selected wavelength range, reflecting the rate at which the reflectivity changes with wavelength.

[0043] The inter-band reflectivity difference characteristic refers to the ratio of the near-infrared to visible light bands. The abnormality of this ratio reflects the change of stress or density within the identified area. The inter-band reflectivity difference characteristic satisfies the following formula: Where, Indicates the central wavelength of the visible light band (550nm), Indicates the center wavelength of the near-infrared band (850nm); , It represents the reflectivity value in the visible light and near-infrared bands; BR represents the reflectivity ratio of the near-infrared and visible light bands, reflecting the difference in optical response of the material in different bands.

[0044] When the stretching or compression caused by axial bending causes the stress of the material to change, the refractive index of the material will be changed through the photoelastic effect, thereby affecting the peak position of the reflection spectrum.

[0045] Near-infrared (NIR) light has strong penetrating power and can reveal the internal state of a material (such as density and stress). Visible light is more sensitive to surface properties. Internal defects (such as density unevenness due to dimensional tolerances) can alter the absorption and reflection characteristics of the NIR band, while surface cracks primarily affect the visible band. Therefore, the presence of axial bending in the tested screw thread can be determined by peak wavelength shift and inter-band reflectivity differences. The presence of surface cracks can also be determined by peak wavelength shift and the slope of the spectral curve.

[0046] Specifically: Axial bending: The refractive index changes significantly in the stress concentration area, causing the reflectivity peak wavelength to shift. Among them, tensile stress causes red shift, and compressive stress causes blue shift. Uneven stress distribution may change the optical anisotropy of the material, resulting in abnormal reflectivity in specific bands. Through experimental analysis, it is determined that the reflectivity peak shift in the stress concentration area ≥2nm; if the reflectivity BR between bands decreases ≤5%, it is determined that the tailless screw sleeve to be tested has axial bending.

[0047] Surface cracks: The surface roughness at the crack increases, which enhances the scattering of short wavelengths (such as blue light), causing the peak wavelength to shift toward the long wavelength direction (red shift); the increase in surface roughness will enhance the scattering of short wavelength light (Mie scattering), resulting in a decrease in the reflectivity of the blue light band, while the near-infrared band is less affected by scattering and has a relatively high reflectivity, thereby increasing the slope. Through experimental analysis, it is determined that the peak reflectivity shift in the stress concentration area ≥2nm, spectral curve slope S≥0.06 nm -1 , it is determined that the tailless screw sleeve to be tested has surface cracks.

[0048] For dimensional deviations, the ratio of the modulus of the vector between the selected pixels to the modulus of the vector between the same pixels on the standard part is calculated to determine whether there is a dimensional deviation. For example, if a thread has 10 turns, 10 pixels with significant features are selected. The two-dimensional coordinates of the 10 pixels are obtained, and the modulus of the vector between the first and last pixels, as well as the modulus of the vector between any two adjacent pixels, are calculated and compared with the modulus between the same pixels on the standard part. If all deviations are less than 0.5%, it is determined that there is no dimensional deviation. If the deviation of a vector modulus is greater than or equal to 0.5%, all modulus lengths are summed and compared with the sum of the modulus lengths of the standard part. If the deviation is still greater than or equal to 0.5%, it is considered a dimensional deviation, and the vector with the largest deviation value is marked as the fault point. This method can reduce errors caused by resolution and the probability of misjudgment. For the production and processing of tailless screw sleeves, dimensional deviations are generally caused by the accumulation of errors in the mechanical processing equipment during the processing. Therefore, the modulus of the vector calculated between adjacent pixel points can determine the specific location where the dimensional deviation is large, which facilitates the subsequent adjustment of the parameters of the processing equipment to reduce errors.

[0049] 3: Construct a recognition model based on convolutional neural network (CNN); the input layer receives fused feature data; the fused feature data includes the offset of the peak wavelength , the slope of the spectral curve S, the reflectivity ratio BR of the near-infrared and visible light bands, and the modulus of the vector formed between the pixels; The output layer outputs whether the tailless screw sleeve is "normal" or "deformed", as well as the location of the "deformation" and the specific defect type; the network structure of the recognition model based on the convolutional neural network (CNN) constructed in step 3 includes an input layer, at least 4 convolutional layers, 2 pooling layers, 3 fully connected layers, and an output layer; The output layer uses the Softmax function; The CNN model was trained using a labeled dataset consisting of 5,000 images of normal and 3,000 images of deformed screw caps. The deformation types included axial bending, surface cracks, and dimensional deviations. Training parameters: learning rate 0.001, batch size 32, optimizer Adam, loss function cross entropy loss, training epochs 50.

[0050] 4: Use the validation set to verify the accuracy of the constructed recognition model; if the recognition accuracy is greater than 95%, the model training is considered complete; 5: Update the trained recognition model to the computing unit and perform deformation detection on the produced tailless screw sleeves; upload the detection results to the host computer; 6: Regularly add new detection data to the training set and use transfer learning strategies to fine-tune model parameters; when the model misjudgment rate exceeds the preset threshold, retrain the model and update it to the computing unit.

[0051] During the model training process, a certain number of standard parts and deformed defective parts of the tailless screw sleeve can be extracted to form a validation set in a ratio of 7 to 3; the deformation of the tailless screw sleeve is detected using the recognition model pre-updated in the calculation unit to determine the accuracy of the recognition.

[0052] In actual use, after the model training is completed, it is updated into the computing unit. Then, the detection platform is built on the transmission production line of the tailless screw insert to provide a support platform for the detection equipment to be easily installed and fixed. The detection equipment needs to be vertically set directly above the support platform to collect the reflection spectrum of the tailless screw insert to be tested from top to bottom. Tailless screw inserts can be transported on the transmission production line using customized carriers, and multiple tailless screw inserts can be placed in the carrier in batches in a specific order; the carrier follows the transmission production line to the detection platform, at which time it can pause for a period of time to allow the multispectral imaging equipment to image the tailless screw inserts placed on the customized carrier; the time is generally set to 1-2 seconds; the specific parameters can be adjusted according to the number of tailless screw inserts imaged at one time.

[0053] The test results are output to the host computer, which locates defective products so that subsequent equipment can remove defective threaded inserts. Simultaneously, the data is stored in the data storage unit via the communication unit, facilitating subsequent analysis of the cause of the defect, optimizing processing equipment and processes, and improving the yield rate. Furthermore, the probability of the same defect type occurring during the processing of the same batch of products is high. The calculation unit can be adjusted to only judge a specific defect, such as axial bending, surface cracks, or dimensional deviations on the threaded insert being tested. This reduces the amount of calculation required and improves detection efficiency.

[0054] Example 3 Figure 3 A schematic structural diagram of a computer system suitable for implementing an electronic device according to an embodiment of the present invention is shown.

[0055] It should be noted that the computer system of the electronic device in this embodiment is only an example and should not bring any limitation to the functions and scope of use of the embodiment of the present invention.

[0056] In the present embodiment, computer system includes central processing unit (CPU) 401, which can be loaded into the random access memory (RAM) 403 from storage portion 408 according to the program stored in the read-only memory (ROM) 402 and various suitable actions and processes, such as performing the tailless screw sleeve deformation detection system and method based on multispectral imaging described in the above-described embodiment. In the random access memory (RAM) 403, various programs and data required for system operation are also stored. Central processing unit (CPU) 401, read-only memory (ROM) 402 and random access memory (RAM) 403 are connected to each other by bus 404. Input / output interface (I / O) 405 is also connected to bus 404.

[0057] The following components are connected to the input / output interface 405: an input section 406 including a keyboard, mouse, and the like; an output section 407 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and speakers; a storage section 408 including devices such as a hard disk; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the input / output interface 405 as needed. Removable media 411, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 410 as needed, so that computer programs read from the removable media can be installed in the storage section 408 as needed.

[0058] In particular, according to an embodiment of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present invention includes a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 409 and / or installed from a removable medium 411. When the computer program is executed by the central processing unit 401, the various functions defined in the system of the present invention are performed.

[0059] It should be noted that the computer-readable medium described in the embodiments of the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable storage media may be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a 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, device, or component. In the present invention, a computer-readable signal medium may include a data signal transmitted in baseband or as part of a carrier wave, which carries a computer-readable computer program. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. A computer program embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0060] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in an order different from that marked in the accompanying drawings. For example, two boxes shown 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 box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0061] The units involved in the embodiments of the present invention may be implemented in software or hardware, and the units described may also be provided in a processor. In some cases, the names of these units do not limit the units themselves.

[0062] According to one aspect of the present invention, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described above.

[0063] As another aspect, the present invention further provides a computer-readable medium, which may be included in the electronic device described in the above embodiments, or may exist independently without being incorporated into the electronic device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device implements the system and method for detecting deformation of a tailless screw sleeve based on multispectral imaging as described in the above embodiments.

[0064] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0065] Through the description of the above embodiments, it will be readily understood by those skilled in the art that the exemplary embodiments described herein can be implemented via software or via a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present invention can be embodied in the form of a software product, which can be stored on a non-volatile storage medium (such as a CD-ROM, USB flash drive, or mobile hard drive) or on a network and includes instructions for causing a computing device (such as a personal computer, server, touch terminal, or network device) to execute the methods according to the embodiments of the present invention.

[0066] Other embodiments of the present invention will readily occur to those skilled in the art after considering the specification and practicing the embodiments disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein.

[0067] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for detecting deformation of a tailless screw sleeve based on multispectral imaging, characterized in that: The specific steps include: S1: Use multispectral imaging equipment to collect reflectance spectra of different types of tailless screw sleeve standard parts and deformation defect parts in different spectral bands, and construct data sets according to model classification; S2: preprocessing and feature extraction of the obtained reflection spectrum; S3: Construct a recognition model based on convolutional neural network (CNN); the input layer receives fused feature data; the output layer outputs whether the tailless screw sleeve is normal or deformed, as well as the location of the deformation and the specific defect type; S4: Use the validation set to verify the accuracy of the constructed recognition model; if the recognition accuracy is greater than 95%, the model training is considered complete; S5: Update the trained recognition model to the computing unit, perform deformation detection on the produced tailless screw sleeve, and upload the detection result to the host computer; S6: Regularly add new detection data to the training set and use transfer learning strategy to fine-tune model parameters; when the model misjudgment rate exceeds the preset threshold, retrain the model and update it to the computing unit.

2. The method for detecting deformation of a tailless screw sleeve based on multispectral imaging according to claim 1, characterized in that: In step S1, at least 6 spectral images of different wavelength bands are collected for each tailless screw sleeve to generate a reflectance spectrum.

3. The method for detecting deformation of a tailless screw sleeve based on multispectral imaging according to claim 1, characterized in that: In step S2, the reflectance map is denoised, normalized and registered to eliminate ambient light interference and image distortion; and the fusion features of the multispectral image are extracted, including texture features, geometric features and spectral reflectance differences.

4. The method for detecting deformation of a tailless screw sleeve based on multispectral imaging according to claim 1, characterized in that: The network structure of the recognition model based on the convolutional neural network (CNN) constructed in step S3 includes an input layer, at least 4 convolutional layers, 2 pooling layers, 3 fully connected layers, and an output layer; The output layer uses the Softmax function; The CNN model was trained using a labeled data set, which included 5,000 sets of normal tailless screw insert images and 3,000 sets of deformed tailless screw insert images. The deformation types included axial bending, surface cracks, and dimensional tolerance. Training parameters: learning rate 0.001, batch size 32, optimizer is Adam, loss function is cross entropy loss, training rounds are 50 times.

5. A deformation detection system for a tailless screw sleeve based on multispectral imaging, using the method described in any one of claims 1 to 4 to detect the tailless screw sleeve, characterized in that: It includes a detection platform, a multispectral imaging device, an image processing unit, a computing unit, a data storage unit and a communication unit; The testing platform is built on the transmission production line of the tailless screw sleeve to provide a support platform for the testing equipment to be easily installed and fixed; The multispectral imaging device is used to image the tailless screw sleeve passing through the detection platform and collect image data in different bands; The image processing unit is used to preprocess and extract features from the collected images; The calculation unit is used to calculate the extracted features to determine whether the tailless screw sleeve to be tested has deformation; The data storage unit is used to save the test results and build a historical database; The communication unit is used to communicate with the host computer to feedback the detection results and receive the control instructions from the host computer; The tailless screw sleeves circulating on the transmission production line are imaged by multi-spectral imaging equipment, and image data in different bands are collected. After processing the image data, feature recognition is performed, and the trained neural network model is used in the computing unit to perform feature calculations to determine whether there is deformation.

6. The tailless screw sleeve deformation detection system based on multispectral imaging according to claim 5 is characterized in that: The tailless screw sleeve is transported on the transmission production line using a customized carrier. It is transported to the detection platform and paused for a period of time to allow the multi-spectral imaging equipment to image the tailless screw sleeve.

7. The tailless screw sleeve deformation detection system based on multispectral imaging according to claim 5 or 6, characterized in that: The spectral band of the multi-spectral imaging device is 400-1100nm; the multi-spectral imaging device uses incident light of different spectral bands from low to high to hit the tailless screw sleeve to be detected, and collects images composed of reflected light to generate a reflection spectrum.

8. The tailless screw sleeve deformation detection system based on multispectral imaging according to claim 7 is characterized in that: Different data sets are established in the historical database for different types of tailless screw inserts; The reflection maps of the standard parts of this model and the reflection maps of the parts with deformation defects are collected in advance; the obtained reflection maps are used as training sets and feature recognition is performed in a neural network model.

9. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the detection method in the tailless screw sleeve deformation detection system based on multi-spectral imaging as described in any one of claims 5 to 8 is implemented.

10. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the detection method in the tailless screw sleeve deformation detection system based on multi-spectral imaging as described in any one of claims 5 to 8.

Images