Since the reliability of the avionics module is crucial for aircraft safety, the fault diagnosis and health management of this module are particularly significant. While deep learning-based ...prognostics and health management (PHM) methods exhibit highly accurate fault diagnosis, they have disadvantages such as inefficient data feature extraction and insufficient generalization capability, as well as a lack of avionics module fault data. Consequently, this study first employs fault injection to simulate various fault types of the avionics module and performs data enhancement to construct the P2020 communications processor fault dataset. Subsequently, a multichannel fault diagnosis method, the Hybrid Attention Adaptive Multi-scale Temporal Convolution Network (HAAMTCN) for the integrated functional circuit module of the avionics module, is proposed, which adaptively constructs the optimal size of the convolutional kernel to efficiently extract features of avionics module fault signals with large information entropy. Further, the combined use of the Interaction Channel Attention (ICA) module and the Hierarchical Block Temporal Attention (HBTA) module results in the HAAMTCN to pay more attention to the critical information in the channel dimension and time step dimension. The experimental results show that the HAAMTCN achieves an accuracy of 99.64% in the avionics module fault classification task which proves our method achieves better performance in comparison with existing methods.
Automatic meter reading via deep learning and computer vision have become feasible for ensuring safe and stable substation operation. Meter model classification is a difficult task because of the ...great similarity in appearance of the different meters. Existing methods mostly focus on detecting and reading individual types of meters, while neglecting to put the classification of multi-class meters on the map. For that, we propose a salient feature fusion convolutional network (SFFCN) for meter model classification and detecting multi-class meters, both of which are crucial steps for further meter readings. The central fusion pyramid network within SFFCN is an improved FPN that efficiently extracts and fuses multi-scale features, thereby enhancing feature saliency and diversity. To improve the accuracy of localization and classification, we introduce a classification-weighted localization attention module (CLAM) to the detection heads. CLAM guides the location branch based on the feature map of the classification branch, mitigating the mismatch between classification confidence and localization accuracy. The experimental results demonstrate that the proposed SFFCN achieves 88.8 mAP on the power meter image dataset and reaches a speed of 68.1 FPS on RTX 3080Ti, effectively improving the detection and classification of multi-class meters.
In recent years, face recognition (FR) has made great achievements with the development of deep convolutional neural networks (CNNs). To obtain highly discriminative features, the existing approaches ...train CNNs equipped with well-designed loss function to constrain their distributions in various feature spaces. (e.g., Euclidean or angular space). However, the prior methods still suffer from two issues: (1) The angular loss functions cannot produce radial binding force to push features away from the origin; (2) the Euclidean ones lack tangential restraints to separate features from different classes. To address the aforementioned issues, we design a simple and efficient loss function, named Origin loss, which integrates multi-action forms supervising forces from Euclidean and angular spaces and additionally produces a radial repulsive force to supplement it. A series of extensive experiments are conducted to demonstrate the effects on feature learning and prove the notable improvement on popular facial benchmarks, in which the proposed loss consistently outperforms than the state-of-the-art (SOTA) methods.
Skeleton‐based action recognition algorithms have been widely applied to human action recognition. Graph convolutional networks (GCNs) generalize convolutional neural networks (CNNs) to non‐Euclidean ...graphs and achieve significant performance in skeleton‐based action recognition. However, existing GCN‐based models have several issues, such as the topology of the graph is defined based on the natural skeleton of the human body, which is fixed during training, and it may not be applied to different layers of the GCN model and diverse datasets. Besides, the higher‐order information of the joint data, for example, skeleton and dynamic information is not fully utilised. This work proposes a novel multi‐stream adaptive spatial‐temporal attention GCN model that overcomes the aforementioned issues. The method designs a learnable topology graph to adaptively adjust the connection relationship and strength, which is updated with training along with other network parameters. Simultaneously, the adaptive connection parameters are utilised to optimise the connection of the natural skeleton graph and the adaptive topology graph. The spatial‐temporal attention module is embedded in each graph convolution layer to ensure that the network focuses on the more critical joints and frames. A multi‐stream framework is built to integrate multiple inputs, which further improves the performance of the network. The final network achieves state‐of‐the‐art performance on both the NTU‐RGBD and Kinetics‐Skeleton action recognition datasets. The simulation results prove that the proposed method reveals better results than existing methods in all perspectives and that shows the superiority of the proposed method.
Medical image fusion (MIF) has received painstaking attention due to its diverse medical applications in response to accurately diagnosing clinical images. Numerous MIF methods have been proposed to ...date, but the fused image suffers from poor contrast, non-uniform illumination, noise presence, and improper fusion strategies, resulting in an inadequate sparse representation of significant features. This paper proposes the morphological preprocessing method to address the non-uniform illumination and noise by the bottom-hat-top-hat strategy. Then, grey-principal component analysis (grey-PCA) is used to transform RGB images into gray images that can preserve detailed features. After that, the local shift-invariant shearlet transform (LSIST) method decomposes the images into the low-pass (LP) and high-pass (HP) sub-bands, efficiently restoring all significant characteristics in various scales and directions. The HP sub-bands are fed to two branches of the Siamese convolutional neural network (CNN) by process of feature detection, initial segmentation, and consistency verification to effectively capture smooth edges, and textures. While the LP sub-bands are fused by employing local energy fusion using the averaging and selection mode to restore the energy information. The proposed method is validated by subjective and objective quality assessments. The subjective evaluation is conducted by a user case study in which twelve field specialists verified the superiority of the proposed method based on precise details, image contrast, noise in the fused image, and no loss of information. The supremacy of the proposed method is further justified by obtaining 0.6836 to 0.8794, 0.5234 to 0.6710, and 3.8501 to 8.7937 gain for QFAB, CRR, and AG and noise reduction from 0.3397 to 0.1209 over other methods for objective parameters.
Deep learning (DL) has achieved significant attention in the field of infrared (IR) and visible (VI) image fusion, and several attempts have been made to enhance the quality of the final fused image. ...It produces better results than conventional methods; however, the captured image cannot acquire useful information due to environments with poor lighting, fog, dense smoke, haze, and the noise generated by sensors. This paper proposes an adaptive fuzzy-based preprocessing enhancement method that automatically enhances the contrast of images with adaptive parameter calculation. The enhanced images are then decomposed into base and detail layers by anisotropic diffusion-based edge-preserving filters that remove noise while smoothing the edges. The detailed parts are fed into four convolutional layers of the VGG-19 network through transfer learning to extract features maps. These features maps are fused by multiple fusion strategies to obtain the final fused detailed layer. The base parts are fused by the PCA method to preserve the energy information. Experimental results reveal that our proposed method achieves state-of-the-art performance compared with existing fusion methods in a subjective evaluation through the visual experience of experts and statistical tests. Moreover, the objective assessment parameters are conducted by various parameters (FMI, SSIMa, API, EN, QFAB, and NFAB) which were used in the comparison method. The proposed method achieves 0.2651 to 0.3951, 0.5827 to 0.8469, 56.3710 to 71.9081, 4.0117 to 7.9907, and 0.6538 to 0.8727 gain for FMI, SSIMa, API, EN, and QFAB, respectively. At the same time, the proposed method has more noise reduction (0.3049 to 0.0021) that further justifies the efficacy of the proposed method than conventional methods.
Due to long-term exposure to the wild, insulators are prone to various defects that affect the safe operation of the power system. In recent years, the combination of drones and deep learning has ...provided a more intelligent solution for insulator automatic defect inspection. Positioning insulators is an important prerequisite step for defect detection, and the accuracy of insulator positioning greatly affects defect detection. However, traditional horizontal detectors lose directional information and it is difficult to accurately locate tilted insulators. Although oriented detectors can predict detection boxes with rotation angles to solve this problem, these models are complex and difficult to apply to edge devices with limited computing power. This greatly limits the practical application of deep learning methods in insulator detection. To address these issues, we proposed a lightweight insulator oriented detector. First, we designed a lightweight insulator feature pyramid network (LIFPN). It can fuse features more efficiently while reducing the number of parameters. Second, we designed a more lightweight insulator oriented detection head (LIHead). It has less computational complexity and can predict rotated detection boxes. Third, we deployed the detector on edge devices and further improved its inference speed through TensorRT. Finally, a series of experiments demonstrated that our method could reduce the computational complexity of the detector by approximately 49 G and the number of parameters by approximately 30 M while ensuring almost no decrease in the detection accuracy. It can be easily deployed to edge devices and achieve a detection speed of 41.89 frames per second (FPS).
Image enhancement is a challenging task in image analysis particularly, it is more challenging in performing image fusion. Image fusion is the process of combining multiple images to produce quality ...output without any variation in contrast, blurring, and noise. Many image fusion algorithms have been implemented, but their final fused images suffer from variations in background contrast, uneven illumination, blurring, and the presence of noise. To overcome the aforementioned issues, this paper proposed a new image fusion method, which improves image contrast and also gives appropriate details of the image. Our method is based on a set of conventional techniques such as amalgamated histogram equalization and fast gray-scale grouping to handle the problems mentioned, and we improve overall fusion strategies by proposing a novel principal component analysis technique to convert RGB types images to high gray-scale contrast image as the final output image. We have carried out many experiments on different common databases used by various researchers. Our proposed method gives good subjective and objective performances compared to other statuses. Our proposed method can be used in different monitoring applications.
Poor illumination, less background contrast and blurring effects makes the medical, satellite and camera images difficult to visualize. Image fusion plays the vital role to enhance image quality by ...resolving the above issues and reducing the image quantity. The combination of spatial and spectral technique Discrete Wavelet Transform and Principal Component Analysis (DWT-PCA) decrease processing time and reduce number of dimensions but down sampling causes lack of shift invariance that results in poor quality final fused image. At first this work uses combined median and average filter that eliminates noise in the image which is caused by illumination, camera circuitry and sensor at preprocessing stage. Then, hybrid Stationary Wavelet Transform and Principal Component Analysis (SWT-PCA) technique is implemented to increase output image accuracy by eliminating down sampling and is not influenced by artifacts and blurring effects. Further, it can overcome the trade-off of Heisenberg’s uncertainty principle by improving accuracy in both domains, time (spatial) as well as frequency (spectral). The proposed combined median and average filter with hybrid SWT-PCA algorithm measures quality parameters, such as peak signal to noise ratio (PSNR), mean squared error (MSE) and normalized cross correlation (NCC) and improved results depict the superiority of the algorithm than existing techniques.
Measuring the temperature of insulator in infrared image and diagnosing whether there is abnormal temperature rise are important for electric power inspection. Traditional object detection uses ...horizontal detection boxes. When they are applied to insulator temperature measurement, their rough positioning results in measurement deviation. Although some detectors with rotated box have been used to tackle this problem, their rotation angle prediction is still not precise enough, resulting in the positioning deviation of insulator core rod, which also leads to temperature measurement deviation. In this article, we propose a complete solution to address these problems. First, we design an adaptive balanced feature pyramid network (Adaptive BFPN) to fuse localization and semantic information more efficiently, which can improve the detection integrity of large-scale insulators. Second, we design a novel <inline-formula> <tex-math notation="LaTeX">\theta </tex-math></inline-formula>-intersection over union (<inline-formula> <tex-math notation="LaTeX">\theta </tex-math></inline-formula>-IoU) loss to enhance the accuracy (Acc) of the detector in predicting rotation angle and insulator positioning. Third, we propose a method to locate and automatically correct the position of insulator core rod, measure the temperature, and diagnose defects. It enables faster temperature measurement and more automatic defect diagnosis. Finally, comparative experiments show that our method has a higher accuracy in diagnosing overheating defects than other advanced methods, reaching 97.07%. Our method provides a more intelligent and automated solution for inspecting insulator overheating defects.