Deep learning-based stealth coating material multi-source data nondestructive testing method

Abstract

The invention discloses a stealth coating material multi-source data nondestructive testing method based on deep learning. The stealth coating material multi-source data nondestructive testing method comprises the following steps: acquiring an infrared thermal imaging image and an optical image; the infrared thermal imaging image is preprocessed; marking a defect position in the infrared thermal imaging image; taking the infrared image as a fourth channel, and carrying out stacking operation on the infrared image and the optical image; performing data enhancement to expand data samples, and dividing a training set and a test set; by adjusting hyper-parameters of the convolutional neural network YOLO-V8, an optimized convolutional neural network YOLO-V8 model used for detecting the infrared thermal imaging image is obtained; the defect area is obtained through network calculation, damage grade grading is carried out according to internal specifications, and damage is mapped to an optical image of a test piece. According to the method, the problems of inaccurate detection, complex detection method and overlong time caused by single-source data can be solved, the position and the area of the defect of the coating material can be conveniently and quickly obtained, and the accuracy, the efficiency and the robustness of detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of stealth coating material defect detection, and in particular to a multi-source data nondestructive detection method for stealth coating materials based on deep learning. Background Art

[0002] Stealth coating is a coating applied to the surface of an aircraft that can absorb electromagnetic waves emitted by radar, and has the function of improving the aircraft's radar stealth performance and infrared stealth performance. However, the stealth coating on the surface of an aircraft may wear, peel or crack due to environmental factors (such as ultraviolet radiation, temperature changes, humidity, etc.), physical impact, friction, aging, etc., thereby reducing its ability to absorb or scatter radar waves, causing the aircraft's RCS value to increase and the stealth function to be greatly reduced. Therefore, it is particularly important to know the status of the aircraft's stealth coating at the first time.

[0003] At present, the main methods for detecting stealth coating defects include visual method, eddy current method, ultrasonic detection method, etc. The methods have the following problems:

[0004] (1) The visual method cannot observe the defects inside the coating and cannot directly obtain the degree of attenuation of the coating's microwave absorption capacity. The eddy current method cannot detect non-conductive materials and is difficult to detect objects with complex shapes. The ultrasonic detection method requires the use of a coupling agent, which will have a certain impact on the coating's microwave absorption effect.

[0005] (2) Infrared thermal imaging technology has the advantages of being non-contact and having a large detection area. However, the contrast between defects and non-defects in the thermal image is not high, and it is easy to misjudge by the naked eye. Therefore, the complete detection time is long and the accuracy is low.

[0006] (3) Judging the damage degree from images is mostly based on the experience of maintenance personnel, and there is currently no systematic damage degree classification. Summary of the invention

[0007] In view of this, the purpose of the present invention is to provide a multi-source data nondestructive testing method for stealth coating materials based on deep learning, which uses infrared thermal imaging detection technology combined with optical image detection technology to obtain preliminary data, and uses deep learning technology to obtain a detection model to solve the problems of inaccurate detection caused by single-source data and complex and time-consuming detection methods, so as to quickly obtain the location and area of ​​coating material defects and improve the accuracy, efficiency and robustness of detection. The area and category of damage are graded according to internal specifications, so that front-line maintenance personnel can perform timely maintenance according to corresponding maintenance specifications.

[0008] The present invention solves the technical problem by adopting the following technical solution:

[0009] A method for nondestructive testing of stealth coating materials using multi-source data based on deep learning, comprising the following steps:

[0010] Step S10: thermally excite the test piece to obtain an infrared thermal imaging image, and collect an optical image of the coating of the test piece at the same angle;

[0011] Step S20: preprocessing the infrared thermal imaging image;

[0012] Step S30: The acquired infrared thermal imaging image is combined with the change of the corresponding optical image to determine the damage position and shape of the optical image of the damaged specimen, and mark it;

[0013] Step S40: Calculate the mean and variance of the pixel values ​​of the four channels of the acquired thermal imaging image and the optical image, and perform normalization processing on the brightness and contrast; after the processing, stack the infrared image as the fourth channel with the optical image;

[0014] Step S50: performing data enhancement on the infrared thermal imaging image and the corresponding optical image obtained in step S10 to expand the data samples, and dividing the expanded data samples into a training set and a test set in a ratio of 8:2;

[0015] Step S60: input the training set obtained in step S40 into the convolutional neural network YOLO-V8 for training, input the test set into the trained convolutional neural network YOLO-V8 during the training process, monitor the detection accuracy of the test set on the convolutional neural network YOLO-V8 in real time, optimize the convolutional neural network YOLO-V8 by adjusting the hyperparameters of the convolutional neural network YOLO-V8, and obtain an optimized convolutional neural network YOLO-V8 model for detecting infrared thermal imaging images;

[0016] Step S70: The infrared thermal imaging image and optical image to be tested are obtained in the manner of step S10, processed through steps S20, S30, and S40, and then input into the optimized convolutional neural network YOLO-V8 model for detecting mixed images obtained in step S60 to detect whether the mixed image contains damage. If so, the position coordinates and area of ​​the damage are obtained, the damage level is graded according to internal specifications, and the damage is mapped to the optical image of the specimen.

[0017] Furthermore, step S10 specifically includes the following steps:

[0018] Step S11: Setting up a light-stimulated infrared thermal imaging nondestructive testing system in the maintenance yard, using a halogen lamp as the excitation light source, with a color temperature of 4500-5500K, fixing the infrared thermal imager on a lamp stand, and aligning it with the area to be tested of the specimen;

[0019] Step S12: Place the optically excited infrared thermal imaging nondestructive testing system at a position of 0.8 to 1.2 meters in the normal direction of the center of the test area, with an excitation current of 9A, that is, a heating time of 2000ms, a power supply voltage of 230V, and an excitation signal of a phase-locked pulse signal to scan the area;

[0020] Step S13: using an industrial camera to collect an optical image of the coating from the normal direction of the center of the area to be tested;

[0021] Step S14: importing the acquired coating infrared thermal imaging image and optical image into the light stimulated infrared thermal imaging nondestructive testing system.

[0022] Furthermore, in step S20, the method for preprocessing the infrared thermal imaging image is as follows: the infrared thermal imaging image and the optical image obtained in S10 are size-aligned, the resolution of the industrial camera used is (1920, 1080, 3), and the resolution of the infrared thermal imaging image is (640, 480, 1); bilinear interpolation is performed using OpenCV to adjust the infrared image to (1920, 1080, 1).

[0023] Furthermore, in step S30, the method of marking the damage position and shape of the optical image is: using a rectangular frame to frame the damage, and the marking format is [X min ,Y min ,X max ,Y max ], where X min , Y min is the coordinate of the upper left corner of the rectangle, X max , Y max The coordinates of the lower right corner of the rectangle.

[0024] Furthermore, in step S40, the brightness and contrast are normalized using the following formula:

[0025]

[0026] Where μ is the mean of the pixel values ​​of the four channels, σ is the variance of the pixel values ​​of the four channels, f(x, y) represents the image pixel after adjustment, g(x, y) represents the image pixel before adjustment, and (x, y) represents the pixel coordinate position.

[0027] Furthermore, in step S50, the Mixup algorithm is used to perform data enhancement, and the formula is as follows:

[0028]

[0029] Among them, x new is the newly generated input sample, y new is the newly generated label, x Aand x B are two randomly selected input samples from the original data set, y A and B Corresponding to x A and x B where λ is a random weight between 0 and 1, sampled from a Beta distribution and defined as:

[0030] λ~Beta(α,α)

[0031] Among them, α is a hyperparameter used to control the distribution of weights; if α is large, the value of λ is close to 0.5, and the generated new sample is more inclined to be a uniform mixture of the two samples; if α is small, the value of λ tends to be closer to 0 or 1, and the new sample will be closer to a certain original sample.

[0032] Further, in step S60, the method for optimizing and improving the convolutional neural network YOLO-V8 is as follows: on the basis of YOLOv8s, the C2f module in Backbone is replaced with the RepBlock module to enhance the network's feature extraction and feature expression capabilities; the ECA attention mechanism is introduced to adaptively adjust the channel weights so that the model focuses on important features, and the head part of YOLOv8s adopts a decoupled head structure; in terms of loss function, YOLOv8 introduces the following TaskAlignedAssigner positive sample allocation strategy to improve the allocation of positive and negative samples and improve the training effect; at the same time, the distributed focal loss is introduced to further optimize the calculation of the loss function:

[0033]

[0034] Among them, IoU(b,bgt) is the IoU between the predicted box b and the real box bgt, is the probability score of the predicted category c;

[0035]

[0036] Among them, L DFL is the distributed focal loss, w i is the corresponding weight term, which is used to adjust the importance of each interval, p i is the distribution probability predicted by the model, y i is the true distribution probability.

[0037] Furthermore, the initial learning rate of the convolutional neural network YOLO-V8 is 0.01, and the momentum and weight decay are set to 0.937 and 0.0005 respectively; the optimizer used in training is SGD, the batch size is 32, and the Dropout rate is set to 0.5 to reduce overfitting.

[0038] The present invention discloses a method for nondestructive testing of stealth coating materials using multi-source data based on deep learning, which has the following beneficial effects:

[0039] 1. Compared with the existing infrared thermal imaging detection technology, the present invention uses deep learning technology to obtain the detection model, directly mapping the damage location to the optical image, which is convenient for maintenance personnel to handle, reduces detection time, and improves maintenance efficiency

[0040] 2. Compared with the existing technology of using deep learning to detect object defects, the present invention can not only retain its original advantages, but also classify the damage types, so that front-line maintenance personnel can intuitively understand the damage situation and perform repairs according to the situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of the method of the present invention;

[0042] Figure 2 This is the schematic diagram of the optimized YOLO-V8 model. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] refer to Figure 1 The present invention discloses a method for nondestructive testing of stealth coating materials using multi-source data based on deep learning, comprising the following steps:

[0045] Step S10: thermally excite the test piece to obtain an infrared thermal imaging image, and collect an optical image of the coating of the test piece at the same angle;

[0046] Step S20: preprocessing the infrared thermal imaging image;

[0047] Step S30: The acquired infrared thermal imaging image is combined with the change of the corresponding optical image to determine the damage position and shape of the optical image of the damaged specimen, and mark it;

[0048] Step S40: Calculate the mean and variance of the pixel values ​​of the four channels of the acquired thermal imaging image and the optical image, and perform normalization processing on the brightness and contrast; after the processing, stack the infrared image as the fourth channel with the optical image;

[0049] Step S50: performing data enhancement on the infrared thermal imaging image and the corresponding optical image obtained in step S10 to expand the data samples, and dividing the expanded data samples into a training set and a test set in a ratio of 8:2;

[0050] Step S60: input the training set obtained in step S40 into the convolutional neural network YOLO-V8 for training, input the test set into the trained convolutional neural network YOLO-V8 during the training process, monitor the detection accuracy of the test set on the convolutional neural network YOLO-V8 in real time, optimize the convolutional neural network YOLO-V8 by adjusting the hyperparameters of the convolutional neural network YOLO-V8, and obtain an optimized convolutional neural network YOLO-V8 model for detecting infrared thermal imaging images;

[0051] Step S70: The infrared thermal imaging image and optical image to be tested are obtained in the manner of step S10, processed through steps S20, S30, and S40, and then input into the optimized convolutional neural network YOLO-V8 model for detecting mixed images obtained in step S60 to detect whether the mixed image contains damage. If so, the position coordinates and area of ​​the damage are obtained, the damage level is graded according to internal specifications, and the damage is mapped to the optical image of the specimen.

[0052] To further optimize the technical solution, step S10 specifically includes the following steps:

[0053] Step S11: Setting up a light-stimulated infrared thermal imaging nondestructive testing system in the maintenance yard, using a halogen lamp as the excitation light source, with a color temperature of 4500-5500K, fixing the infrared thermal imager on a lamp stand, and aligning it with the area to be tested of the specimen;

[0054] Step S12: Place the optically excited infrared thermal imaging nondestructive testing system at a position of 0.8 to 1.2 meters in the normal direction of the center of the test area, with an excitation current of 9A, that is, a heating time of 2000ms, a power supply voltage of 230V, and an excitation signal of a phase-locked pulse signal to scan the area;

[0055] Step S13: using an industrial camera to collect an optical image of the coating from the normal direction of the center of the area to be tested;

[0056] Step S14: importing the acquired coating infrared thermal imaging image and optical image into the light stimulated infrared thermal imaging nondestructive testing system.

[0057] To further optimize the technical solution, in step S20, the method for preprocessing the infrared thermal imaging image is as follows: align the sizes of the infrared thermal imaging image and the optical image obtained in S10, the resolution of the industrial camera used is (1920, 1080, 3), and the resolution of the infrared thermal imaging image is (640, 480, 1); use OpenCV for bilinear interpolation to adjust the infrared image to (1920, 1080, 1).

[0058] To further optimize the technical solution, in step S30, the method of marking the damage position and shape of the optical image is: using a rectangular frame to frame the damage, and the marking format is [X min ,Y min ,X max ,Y max ], where X min , Y min is the coordinate of the upper left corner of the rectangle, X max , Y max The coordinates of the lower right corner of the rectangle.

[0059] To further optimize the technical solution, in step S40, the brightness and contrast are normalized using the following formula:

[0060]

[0061] Where μ is the mean of the pixel values ​​of the four channels, σ is the variance of the pixel values ​​of the four channels, f(x, y) represents the image pixel after adjustment, g(x, y) represents the image pixel before adjustment, and (x, y) represents the pixel coordinate position.

[0062] To further optimize the technical solution, in step S50, data enhancement includes methods such as random image rotation, random cropping, random scaling, adding Gaussian noise, mosaic (mosaic data enhancement), etc. On the basis of the above methods, the Mixup algorithm is used to perform data enhancement, and the formula is as follows:

[0063]

[0064] Among them, x new is the newly generated input sample, y new is the newly generated label, x A and x B are two randomly selected input samples from the original data set, y A and B Corresponding to x A and x B where λ is a random weight between 0 and 1, sampled from a Beta distribution and defined as:

[0065] λ~Beta(α,α)

[0066] Among them, α is a hyperparameter used to control the distribution of weights; if α is large, the value of λ is close to 0.5, and the generated new sample is more inclined to be a uniform mixture of the two samples; if α is small, the value of λ tends to be closer to 0 or 1, and the new sample will be closer to a certain original sample.

[0067] To further optimize the technical solution, in step S60, the stacked images of the training set obtained in step S50 are input into the convolutional neural network YOLO-V8, and the prediction result of the convolutional neural network YOLO-V8 is output, that is, [Xmin, Ymin, Xmax, Ymax] is compared with the real mark [Xmin, Ymin, Xmax, Ymax] of the optical image of the infrared thermal imaging image, and the loss function is calculated. The gradient of the loss function is calculated, and then the model is corrected to reduce the loss function value. At least 1000 rounds of training are performed, and after every 10 rounds of training, the test set is input into the trained convolutional neural network YOLO-V8, and the detection accuracy of the test set on the convolutional neural network YOLO-V8 is monitored in real time, and the optimized convolutional neural network YOLO-V8 model for detecting infrared thermal imaging images is obtained by adjusting the hyperparameters.

[0068] The present invention adopts an improved yolov8 model. YOLOv8s is a lightweight version of the YOLOv8 model, which is specially designed to perform target detection and segmentation tasks in resource-limited environments. YOLOv8s is smaller in model scale than other versions (such as YOLOv8m and YOLOv8l), but still retains strong performance and high efficiency, and is suitable for scenarios such as embedded devices and mobile devices. The improved algorithm proposed in the present invention replaces the C2f module in Backbone with the RepBlock module on the basis of YOLOv8s to enhance the feature extraction and feature expression capabilities of the network. The content-aware feature recombination module (CARAFE) is used to replace the upsampling method to improve the segmentation accuracy of small targets. The ECA attention mechanism is introduced to focus the model on important features through adaptive adjustment of channel weights, thereby improving the ability to extract key information in complex scenes. In addition, the head part of YOLOv8s is significantly different from YOLOv5, and a decoupled head structure is adopted to improve detection accuracy and efficiency. In terms of loss function, YOLOv8 introduces the following TaskAlignedAssigner positive sample allocation strategy to improve the allocation of positive and negative samples and improve the training effect; at the same time, it also introduces the distributed focal loss (DistributionFocalLoss) to further optimize the calculation of the loss function.

[0069]

[0070] Among them, IoU(b,bgt) is the IoU between the predicted box b and the real box bgt, is the probability score of the predicted category c;

[0071]

[0072] Among them, L DFL is the distributed focal loss, w i is the corresponding weight term, which is used to adjust the importance of each interval, p i is the distribution probability predicted by the model, y i is the true distribution probability (the discrete distribution generated by the true coordinate values).

[0073] To further optimize the technical solution, the initial learning rate of the convolutional neural network YOLO-V8 is 0.01, and the momentum and weight decay are set to 0.937 and 0.0005 respectively; the optimizer used for training is SGD, the batch size is 32, and the Dropout rate is set to 0.5 to reduce overfitting.

[0074] Compared with the existing infrared thermal imaging detection technology, the present invention uses deep learning technology to obtain a detection model, directly mapping the damage location to the optical image, which is convenient for maintenance personnel to handle, reduces detection time, and improves maintenance efficiency. Compared with the existing technology of using deep learning to detect object defects, the present invention can also classify the damage type on the basis of retaining its original advantages, so that front-line maintenance personnel can intuitively understand the damage situation and repair it according to the situation.

[0075] Example

[0076] This embodiment provides a method for nondestructive testing of stealth coating materials using multi-source data based on deep learning. The specific process is as follows:

[0077] Step S10: light excitation equipment, infrared thermal imaging and optical imaging equipment are set up in the experimental field, with a halogen lamp power of 2000W, a power supply current of 6A, a voltage of 230V, a spacing of 1 meter, and a height of 1.5 meters from the ground;

[0078] Move the test system to 1 meter in the normal direction of the area to be tested, the excitation system outputs phase-locked excitation, the halogen lamp excitation time is 2000ms, the infrared thermal imager obtains the thermal imaging image of the measurement area in real time, and stores the measurement results as the target thermal imaging data file;

[0079] Use an industrial camera to collect optical images of the coating at the same angle, and store the acquired images in the system;

[0080] In this embodiment, 2000 infrared thermal imaging images and optical images of the stealth coating are obtained, of which 1400 are damaged and 600 are undamaged. The damage mainly includes physical scale damage and corrosion and oxidation damage. The physical scale damage is manifested as shedding, cracking and other phenomena; the corrosion and oxidation damage is manifested as rust and blistering and other phenomena.

[0081] Step S20: Align the sizes of the infrared image and the optical image obtained in S10. The resolution of the industrial camera used is (1920, 1080, 3), and the resolution of the infrared image is (640, 480, 1). Use OpenCV for bilinear interpolation to adjust the infrared image to (1920, 1080, 1).

[0082] Step S30: Based on the stacked image obtained in step S20 and the changes in the corresponding infrared image and optical image, determine the damage position and shape of the optical image of the damaged specimen, and frame the damage with a rectangular frame. The marking format is [X min ,Y min ,X max ,Y max ], where X min , Y min is the coordinate of the upper left corner of the rectangle, X max , Y max The coordinates of the lower right corner of the rectangle.

[0083] Step S40: normalize the stacked image obtained in step S20: respectively calculate the mean μ and variance σ of the pixel values ​​of the four channels of the stacked image obtained in step S10, let g(x, y) represent the image pixel before adjustment, f(x, y) represent the image pixel after adjustment, and use f(x, y) to normalize the brightness and contrast of the stacked image obtained in step S10:

[0084]

[0085] The image pixel value distribution is made to approximate the standard normal distribution, where x and y represent the pixel coordinate positions. In this embodiment, the mean μ is [0.452, 0.436, 0.445] and the variance σ is [0.211, 0.217, 0.215]. The advantage of normalization is to reduce the difficulty of model training, improve the generalization ability of the model, and prevent the occurrence of gradient explosion.

[0086] Step S50: At the same time, data enhancement is performed on the stacked image obtained in step S20 to expand the data sample, and the number of the expanded data samples is divided into a training set and a test set according to a ratio of 8:2; in this embodiment, random rotation, random cropping, random scaling, adding Gaussian noise and mosaic data enhancement are used to enhance the data of the sample. On the basis of the above method, the Mixup algorithm is used to enhance the data, and the formula is as follows:

[0087]

[0088] Among them, x new is the newly generated input sample, y new is the newly generated label, x A and x B are two randomly selected input samples from the original data set, y A and B Corresponding to x A and x B where λ is a random weight between 0 and 1, sampled from a Beta distribution and defined as:

[0089] λ~Beta(α,α)

[0090] Among them, α is a hyperparameter used to control the distribution of weights. If α is large, the value of λ is close to 0.5, and the generated new sample is more inclined to be a uniform mixture of the two samples; if α is small, the value of λ tends to be closer to 0 or 1, and the new sample will be closer to a certain original sample.

[0091] Step S60: Input the training set obtained in step S50 into the convolutional neural network YOLO-V8 for training. During the training process, input the test set into the trained convolutional neural network YOLO-V8. By adjusting the hyperparameters of the convolutional neural network YOLO-V8 (including n max and n min ), monitor the detection accuracy of the test set on the convolutional neural network YOLO-V8 in real time, optimize the convolutional neural network YOLO-V8, and obtain the optimized convolutional neural network YOLO-V8 model for detecting infrared thermal imaging images, refer to Figure 2 ;

[0092] The initial learning rate of the original convolutional neural network YOLO-V8 model is 0.01, and the learning rate is changed using the "step" method. This embodiment uses the exponential decay method to change the learning rate, which is defined as the following formula:

[0093] η t =η0e -λt

[0094] Where η0 is the initial learning rate, λ is the decay rate, and t is the number of training steps or epochs.

[0095] The training of the convolutional neural network model YOLOv8 is as follows: the infrared thermal imaging image of the training set is input into the convolutional neural network YOLOv8, and the prediction result of the convolutional neural network YOLOv8 is output, that is, [x min ,y min ,x max ,y max ](predict) the true label of the optical image corresponding to the infrared thermal imaging image [x min ,y min ,x max ,y max ] for comparison, find the loss function, find the gradient of the loss function, and then correct the model to reduce the loss function value. Perform at least 1000 rounds of training, and input the test set into the trained convolutional neural network YOLOv8 after every 10 rounds of training. Monitor the detection accuracy of the test set on the convolutional neural network YOLOv8 in real time, and obtain the optimized convolutional neural network YOLO-V8 model for detecting infrared thermal imaging images by adjusting the hyperparameters.

[0096] The convolutional neural network YOLOv8 is the latest version of the real-time target detection algorithm YOLO. Its basic framework integrates CSPNet and other lightweight structures, and has efficient feature extraction capabilities. The YOLOv8 algorithm in this experiment uses an improved backbone network, replacing the C2f module in the original Backbone with the RepBlock module to enhance the network's feature extraction and feature expression capabilities. The SPPF (Spatial Pyramid Pooling-Fast) module is used to extract multi-scale features, retaining efficient feature expression capabilities while reducing computational overhead. YOLOv8 downsamples through convolution with a step size of 2, undergoes multiple downsampling processes, and combines PAN (Path Aggregation Network) for feature fusion, and also performs multiple detections in the network structure. The content-aware feature recombination module (CARAFE) is used to replace the upsampling method to improve the segmentation accuracy of small targets. The ECA attention mechanism is introduced to adaptively adjust the channel weights so that the model focuses on important features and improves the ability to extract key information in complex scenes. At the same time, YOLOv8 uses the Anchor-Free detection mechanism to simplify the network training and reasoning process and reduce the complexity of hyperparameters. In addition, YOLOv8 has made improvements on the loss function, introducing the TaskAlignedAssigner positive sample allocation strategy and distributed focal loss to further optimize the calculation of the loss function and make the bounding box regression more accurate.

[0097] In the training process of the convolutional neural network YOLOv8 model, the SiLU (Swish) activation function can effectively prevent the gradient from disappearing and accelerate network training. The SiLU function multiplies the input by the output of the sigmoid function, making the activation function smoother while maintaining nonlinear expression capabilities, thereby improving the accuracy of the model.

[0098] The SiLU activation function is defined as:

[0099]

[0100] The loss function adopts the TaskAlignedAssigner positive sample allocation strategy to improve the allocation of positive and negative samples and improve the training effect. At the same time, the distributed focal loss (Distribution Focal Loss) is introduced to further optimize the calculation of the loss function as shown in the following formula:

[0101]

[0102] Among them, IoU(b,bgt) is the IoU between the predicted box b and the real box bgt, is the probability score of the predicted category c;

[0103]

[0104] Among them, L DFL is the distributed focal loss, w i is the corresponding weight term, which is used to adjust the importance of each interval, p i is the distribution probability predicted by the model, y i is the true distribution probability (the discrete distribution generated by the true coordinate values).

[0105] The purpose of network training is to find the parameter value that minimizes the loss function; in the present invention, the parameter value is updated, and the updating method is shown in the following formula:

[0106]

[0107] Where v t is the estimate of the first-order momentum of the gradient at step t; s t is the estimate of the second-order momentum of the gradient at step t; v t-1 is the estimated value of the first-order momentum of the gradient at step t-1; s t-1 is the estimate of the second-order momentum of the gradient at step t-1; β1 is the exponentially weighted average coefficient of the first-order momentum estimate, which is usually taken as 0.9 and affects the smoothness of the gradient accumulation; β2 is the exponentially weighted average coefficient of the second-order momentum estimate, which is usually taken as 0.999 and affects the smoothness of the mean square gradient; vt corret is the first-order momentum estimate of the corrected gradient; s t corret is the second-order momentum estimate of the corrected gradient; ω t is the model parameter at the tth step; ω t+1 is the parameter at step t+1; η is the learning rate, which controls the step size of parameter update; ∈ is a very small positive number (∈=10-8), which is used to prevent division by zero during update. It ensures that the denominator always has a certain value to avoid numerical instability; λ is the weight attenuation coefficient.

[0108] Step S70: The infrared thermal imaging image and the optical image to be detected are obtained in the manner of step S10, processed through steps S20, S30, and S40, and then input into the optimized convolutional neural network YOLO-V8 model for detecting the coating obtained in step S60 to detect whether the stealth coating contains damage. If it contains damage, the position coordinates and the damage area of ​​the damage are obtained, the damage is marked on the optical image of the coating, and the damage level is divided according to the degree of damage according to the internal grading specifications, so that the damage position and degree can be visualized on the optical image.

[0109] The present invention combines infrared thermal imaging technology with optical image detection technology, and develops a multi-source data nondestructive testing scheme based on a deep learning model. The detection process first uses a halogen lamp to phase-lock the coating area, uses an infrared thermal imager to collect thermal imaging images of the coating in real time, and uses an industrial camera to obtain optical images of the corresponding angle. The collected infrared thermal imaging and optical images are normalized for contrast, and a variety of data enhancement methods (such as random rotation, cropping, scaling, adding Gaussian noise and mosaic data enhancement) are used to expand the data sample. The processed data is divided into a training set and a test set according to a ratio of 8:2, and then input into the convolutional neural network YOLO-V8 for training. During the model training process, the test set is input in real time and its detection accuracy is monitored, and the model hyperparameters (such as initial learning rate, momentum and weight decay) are adjusted for optimization, thereby gradually improving the accuracy of detection and the robustness of the model. In order to further improve the detection effect, the present invention introduces the distributed focal loss (Distribution FocalLoss) and TaskAlignedAssigner positive sample allocation strategy in the training process of YOLO-V8 to optimize the allocation of positive and negative samples and improve the accuracy of bounding box regression. The optimized YOLO-V8 model is used to detect newly acquired infrared thermal imaging images and optical images, and the coordinates of the defect area and its corresponding area are directly output through the deep learning model. The detection results map the location of the damage to the optical image, and classify the damage according to the preset internal standards, so as to help front-line maintenance personnel intuitively judge the severity of the defect and then formulate corresponding repair plans.

[0110] Compared with traditional detection methods, the present invention has significant advantages. Traditional detection methods such as visual method, eddy current method, ultrasonic detection, etc. all have their own limitations, and the present invention effectively overcomes these technical obstacles through deep learning technology and multi-source data fusion, greatly improving the detection accuracy and efficiency. At the same time, by combining infrared thermal imaging images with internal grading specifications, the present invention realizes damage grading of different types of defects (such as coating peeling, cracks and corrosion, etc.), which not only simplifies the maintenance process, but also significantly improves the maintenance efficiency and reduces the possibility of subjective human misjudgment.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-source data nondestructive testing method for stealth coating materials based on deep learning, characterized in that: The steps include: Step S10: thermally excite the test piece to obtain an infrared thermal imaging image, and collect an optical image of the coating of the test piece at the same angle; Step S20: preprocessing the infrared thermal imaging image; Step S30: The acquired infrared thermal imaging image is combined with the change of the corresponding optical image to determine the damage position and shape of the optical image of the damaged specimen, and mark it; Step S40: Calculate the mean and variance of the pixel values ​​of the four channels of the acquired thermal imaging image and optical image, and perform normalization processing on the brightness and contrast; After processing, the infrared image is used as the fourth channel to be stacked with the optical image; Step S50: performing data enhancement on the infrared thermal imaging image and the corresponding optical image obtained in step S10 to expand the data samples, and dividing the expanded data samples into a training set and a test set in a ratio of 8:2; Step S60: input the training set obtained in step S40 into the convolutional neural network YOLO-V8 for training, input the test set into the trained convolutional neural network YOLO-V8 during the training process, monitor the detection accuracy of the test set on the convolutional neural network YOLO-V8 in real time, optimize the convolutional neural network YOLO-V8 by adjusting the hyperparameters of the convolutional neural network YOLO-V8, and obtain an optimized convolutional neural network YOLO-V8 model for detecting infrared thermal imaging images; Step S70: The infrared thermal imaging image and optical image to be tested are obtained in the manner of step S10, processed through steps S20, S30, and S40, and then input into the optimized convolutional neural network YOLO-V8 model for detecting mixed images obtained in step S60 to detect whether the mixed image contains damage. If so, the position coordinates and area of ​​the damage are obtained, the damage level is graded according to internal specifications, and the damage is mapped to the optical image of the specimen.

2. According to claim 1, a method for nondestructive testing of stealth coating materials based on multi-source data based on deep learning is characterized in that: Step S10 specifically includes the following steps: Step S11: Setting up a light-stimulated infrared thermal imaging nondestructive testing system in the maintenance yard, using a halogen lamp as the excitation light source, with a color temperature of 4500-5500K, fixing the infrared thermal imager on a lamp stand, and aligning it with the area to be tested of the specimen; Step S12: Place the optically excited infrared thermal imaging nondestructive testing system at a position of 0.8 to 1.2 meters in the normal direction of the center of the test area, with an excitation current of 9A, that is, a heating time of 2000ms, a power supply voltage of 230V, and an excitation signal of a phase-locked pulse signal to scan the area; Step S13: using an industrial camera to collect an optical image of the coating from the normal direction of the center of the area to be tested; Step S14: importing the acquired coating infrared thermal imaging image and optical image into the light stimulated infrared thermal imaging nondestructive testing system.

3. According to claim 2, a method for nondestructive testing of stealth coating materials based on multi-source data based on deep learning is characterized in that: In step S20, the method for preprocessing the infrared thermal imaging image is as follows: align the sizes of the infrared thermal imaging image and the optical image obtained in S10, the resolution of the industrial camera used is (1920, 1080, 3), and the resolution of the infrared thermal imaging image is (640, 480, 1); use OpenCV for bilinear interpolation to adjust the infrared image to (1920, 1080, 1).

4. According to claim 3, a method for nondestructive testing of stealth coating materials based on multi-source data based on deep learning is characterized in that: In step S30, the method of marking the damage position and shape of the optical image is: using a rectangular frame to frame the damage, and the marking format is [X min ,Y min ,X max ,Y max ], where X min , Y min is the coordinate of the upper left corner of the rectangle, X max , Y max The coordinates of the lower right corner of the rectangle.

5. According to claim 4, a method for nondestructive testing of stealth coating materials based on multi-source data based on deep learning is characterized in that: In step S40, the brightness and contrast are normalized using the following formula: Where μ is the mean of the pixel values ​​of the four channels, σ is the variance of the pixel values ​​of the four channels, f(x, y) represents the image pixel after adjustment, g(x, y) represents the image pixel before adjustment, and (x, y) represents the pixel coordinate position.

6. The method for nondestructive testing of stealth coating materials based on multi-source data based on deep learning according to claim 5 is characterized in that: In step S50, the Mixup algorithm is used to perform data enhancement, and the formula is as follows: Among them, x new is the newly generated input sample, y new is the newly generated label, x A and x B are two randomly selected input samples from the original data set, y A and B Corresponding to x A and x B where λ is a random weight between 0 and 1, sampled from a Beta distribution and defined as: λ~Beta(α,α) Among them, α is a hyperparameter used to control the distribution of weights; if α is large, the value of λ is close to 0.5, and the generated new sample is more inclined to be a uniform mixture of the two samples; if α is small, the value of λ tends to be closer to 0 or 1, and the new sample will be closer to a certain original sample.

7. The method for nondestructive testing of stealth coating materials based on multi-source data based on deep learning according to claim 6 is characterized in that: In step S60, the method for optimizing and improving the convolutional neural network YOLO-V8 is as follows: on the basis of YOLOv8s, the C2f module in Backbone is replaced with the RepBlock module to enhance the network's feature extraction and feature expression capabilities; the ECA attention mechanism is introduced to adaptively adjust the channel weights so that the model focuses on important features, and the head part of YOLOv8s adopts a decoupled head structure; in terms of loss function, YOLOv8 introduces the following TaskAlignedAssigner positive sample allocation strategy to improve the allocation of positive and negative samples and improve the training effect; at the same time, the distributed focal loss is introduced to further optimize the calculation of the loss function: Among them, IoU(b,bgt) is the IoU between the predicted box b and the real box bgt, is the probability score of the predicted category c; Among them, L DFL is the distributed focal loss, w i is the corresponding weight term, which is used to adjust the importance of each interval, p i is the distribution probability predicted by the model, y i is the true distribution probability.

8. The method for nondestructive testing of stealth coating materials based on multi-source data based on deep learning according to claim 7 is characterized in that: The initial learning rate of the convolutional neural network YOLO-V8 is 0.01, and the momentum and weight decay are set to 0.937 and 0.0005 respectively; the optimizer used for training is SGD, the batch size is 32, and the Dropout rate is set to 0.5 to reduce overfitting.

Images