Channel Attention Module Research Articles

Improving Deep Learning-based Saliency Detection Using Channel Attention Module

Improving Deep Learning-based Saliency Detection Using Channel Attention Module

An Automated Framework for Histopathological Nucleus Segmentation With Deep Attention Integrated Networks.

Clinical management and accurate disease diagnosis are evolving from qualitative stage to the quantitative stage, particularly at the cellular level. However, the manual process of histopathological analysis is lab-intensive and time-consuming. Meanwhile, the accuracy is limited by the experience of the pathologist. Therefore, deep learning-empowered computer-aided diagnosis (CAD) is emerging as an important topic in digital pathology to streamline the standard process of automatic tissue analysis. Automated accurate nucleus segmentation can not only help pathologists make more accurate diagnosis, save time and labor, but also achieve consistent and efficient diagnosis results. However, nucleus segmentation is susceptible to staining variation, uneven nucleus intensity, background noises, and nucleus tissue differences in biopsy specimens. To solve these problems, we propose Deep Attention Integrated Networks (DAINets), which mainly built on self-attention based spatial attention module and channel attention module. In addition, we also introduce a feature fusion branch to fuse high-level representations with low-level features for multi-scale perception, and employ the mark-based watershed algorithm to refine the predicted segmentation maps. Furthermore, in the testing phase, we design Individual Color Normalization (ICN) to settle the dyeing variation problem in specimens. Quantitative evaluations on the multi-organ nucleus dataset indicate the priority of our automated nucleus segmentation framework.

An Improved Lightweight YOLO Algorithm for Recognition of GPS Interference Signals in Civil Aviation

Considering several sources that cause global position system (GPS) interference in civil aviation and the challenges faced by interference recognition algorithms in terms of efficiency and accuracy, we propose an improved You Only Look Once (YOLO)v7‐CHS algorithm (YOLOv7‐CHS) and investigate its effectiveness in identifying GPS signals and different types of interference signals. First, continuous wavelet transform (CWT) is introduced as a method for processing and analyzing signals in the time–frequency (TF) domain to effectively obtain their temporal and spectral characteristic information. Second, the ConvNeXt structure is integrated into the YOLOv7 backbone network to create a ConvNeXtBlock (CNeB) module to enhance the classification and recognition accuracy of interference signals. Additionally, an attention mechanism is introduced to further improve model recognition accuracy. To effectively improve the capability of signal feature extraction and mitigate the impact of background noise on TF feature suppression, we have integrated the efficient channel attention (ECA) channel attention module with the convolutional block attention module (CBAM) spatial attention module, thereby proposing a hybrid CBAM and ECA (HCE) attention module. Last, to address issues arising from accidental deletion of detection frames and multipath interference negatively affecting model recognition performance, we have employed the soft nonmaximum suppression (Soft‐NMS) algorithm while selecting an optimal loss function through comparative analysis. The comparative evaluation experimental results under different circumstances show that YOLOv7‐CHS achieves recognition accuracies of 98.0% and 99.6% for various types of signals, respectively. These values represent an increase of 1.7% and 1%, respectively, compared to YOLOv7. Moreover, in terms of lightweight indicators, YOLOv7‐CHS exhibits a significant improvement in performance: the frames per second (FPS) is increased by 75.1, the number of parameters (Params) was reduced by 4.75 M, and giga floating point operations per second (GFLOPs) were reduced by 65.9 G while effectively enhancing recognition capabilities. The proposed YOLOv7‐CHS not only improves signal recognition accuracy but also reduces model Params and computational complexity, achieving a lightweight model with promising application prospects in the rapid detection and recognition of GPS interference sources in civil aviation.

Underground Target Classification from Full-polarimetric GPR Data Using Deep Convolutional Neural Network with Channel Attention Module

Underground Target Classification from Full-polarimetric GPR Data Using Deep Convolutional Neural Network with Channel Attention Module

PD-YOLO: A study of daily behavioural detection in housed sheep.

Sheep behavior recognition helps to monitor the health status of sheep and prevent the outbreak of infectious diseases. Aiming at the problems of low detection accuracy and slow speed due to the crowding of sheep in real farming scenarios, which can easily obscure each other, this study proposes a lightweight sheep behavior recognition model based on the YOLOv8n model. First, the Convolutional Block Attention Module (CBAM) is introduced and improved in the YOLOv8n model, and the channel attention module and spatial attention module are changed from serial to parallel to construct a novel attention mechanism, PCBAM, to enhance the network's attention to the sheep and eliminate redundant background information; second, the ordinary convolution in the backbone network is replaced with depth-separable convolution, which effectively reduces the number of parameters in the model and reduces the computational complexity. The study takes the housed breeding sheep as the test object, installs a camera diagonally above the sheep pen to collect images and makes a data set for testing, and in order to verify the superiority of the PD-YOLO model, compares it with a variety of target detection models. The experimental results show that the mean average precision (mAP) of the model proposed in this paper are 95.8%, 98.9%, and 96.2% for the three postures of sheep lying, feeding, and standing, respectively, which are 8.5%, 0.8%, and 0.8% higher than those of YOLOv8n, respectively, and the size of the model has been reduced by 13.3% and the amount of computation has been reduced by 12.1%. The inference speed reaches 52.1 FPS per second, which is better than other models in meeting the real-time detection requirement. To verify the practicality of this research method, the PD-YOLO model was deployed on the RK3399Pro development board for testing, and a high inference speed was achieved. It can provide effective technical support for sheep smart farming.

IMD-Net: Interpretable multi-scale detection network for infrared dim and small objects.

This study proposed an interpretable multi-scale infrared small object detection network (IMD-Net) design method to improve the precision of infrared small object detection and contour segmentation in complex backgrounds. To this end, a multi-scale object enhancement module was constructed, which converted artificially designed features into network structures. The network structure was used to enhance actual objects and extract shallow detail and deep semantic features of images. Next, a global object response, channel attention, and multilayer feature fusion modules were introduced, combining context and channel information and aggregated information, selected data, and decoded objects. Finally, the multiple loss constraint module was constructed, which effectively constrained the network output using multiple losses and solved the problems of high false alarms and high missed detections. Experimental results showed that the proposed network model outperformed local energy factor (LEF), self-regularized weighted sparse model (SRWS), asymmetric contextual modulation (ACM), and other state of the art methods in the intersection-over-union (IoU) and Fmeasure values by 10.8% and 11.3%, respectively. The proposed method performed best on the currently available datasets, achieving accurate detection and effective segmentation of dim and small objects in various infrared complex background images.

Remote sensing image semantic segmentation network based on multi-scale feature enhancement fusion

Semantic segmentation is a crucial method for recognizing and classifying objects in high-resolution remote sensing images (HRRSIs). However, due to the problems of varying target scale and difficulty in determining the edges of small-scale targets in remote sensing images, traditional semantic segmentation models perform poorly. To address this issue, we propose a multi-scale feature enhancement network (MFENet) to improve the segmentation accuracy of small-scale objects in HRRSIs. MFENet considers the differences between objects of different scales and selects more suitable receptive fields to enhance the extraction of multi-scale semantic features. We propose a composite atrous multi-scale feature fusion (CAMFF) module to enhance the extraction of spatial detail and semantic information of features at different scales. In addition, we propose an improved composite atrous spatial pyramid pooling (C-ASPP) module to enhance the network feature extraction capability across multiple scales. We also propose a network structure that combines the C-ASPP module with the efficient channel attention (ECA) module in parallel, which performs better to extract contextual information. Our experimental evaluations on the Potsdam and Vaihingen datasets demonstrate the effectiveness of our Network, It F1 score reaching 93.33% and 94.66% respectively.

LeaNet: Lightweight U-shaped architecture for high-performance skin cancer image segmentation

Skin cancer diagnosis often relies on image segmentation as a crucial aid, and a high-performance segmentation can lower misdiagnosis risks. Part of the medical devices often have limited computing power for deploying image segmentation algorithms. However, existing high-performance algorithms for image segmentation primarily rely on computationally intensive large models, making it challenging to meet the lightweight deployment requirement of medical devices. State-of-the-art lightweight models are not able to capture both local and global feature information of lesion edges due to their model structures, result in pixel loss of lesion edge. To tackle this problem, we propose LeaNet, a novel U-shaped network for high-performance yet lightweight skin cancer image segmentation. Specifically, LeaNet employs multiple attention blocks in a lightweight symmetric U-shaped design. Each blocks contains a dilated efficient channel attention (DECA) module for global and local contour information and an inverted external attention (IEA) module to improve information correlation between data samples. Additionally, LeaNet uses an attention bridge (AB) module to connect the left and right sides of the U-shaped architecture, thereby enhancing the model's multi-level feature extraction capability. We tested our model on ISIC2017 and ISIC2018 datasets. Compared with large models like ResUNet, LeaNet improved the ACC, SEN, and SPEC metrics by 1.09 %, 2.58 %, and 1.6 %, respectively, while reducing the model's parameter number and computational complexity by 570x and 1182x. Compared with lightweight models like MALUNet, LeaNet achieved improvements of 2.07 %, 4.26 %, and 3.11 % in ACC, SEN, and SPEC, respectively, reducing the parameter number and computational complexity by 1.54x and 1.04x.

DATran: Dual Attention Transformer for Multi-Label Image Classification

Multi-label image classification is a fundamental yet challenging task, which aims to predict the labels associated with a given image. Most of previous methods directly exploit the high-level features from the last layer of convolutional neural network for classification. However, these methods cannot obtain global features due to the limited size of convolutional kernels, and they fail to extract multi-scale features to effectively recognize small-scale objects in the images. Recent studies exploit the graph convolution network to model the label correlations for boosting the classification performance. Despite substantial progress, these methods rely on manually pre-defined graph structures. Besides, they ignore the associations between semantic labels and image regions, and do not fully explore the spatial context of images. To address above issues, we propose a novel Dual Attention Transformer (DATran) model, which adopts a dual-stream architecture that simultaneously learns spatial and channel correlations from multi-label images. Firstly, in order to solve the problem that current methods are difficult to recognize small-size objects, we develop a new multi-scale feature fusion (MSFF) module to generate multi-scale feature representation by jointly integrating both high-level semantics and low-level details. Secondly, we design a prior-enhanced spatial attention (PSA) module to learn the long-range correlation between objects from different spatial positions in images to enhance the model performance. Thirdly, we devise a prior-enhanced channel attention (PCA) module to capture the inter-dependencies between different channel maps, thus effectively improving the correlation between semantic categories. It is worth noting that PSA module and PCA module complement and promote each other to further augment the feature representations. Finally, the outputs of these two attention modules are fused to obtain the final features for classification. Performance evaluation experiments are conducted on MS-COCO 2014, PASCAL VOC 2007 and VG-500 datasets, demonstrating that DATran model achieves better performance than current state-of-the-art models.

MH6D: Multi-Hypothesis Consistency Learning for Category-Level 6-D Object Pose Estimation.

Six-degree-of-freedom (6DoF) object pose estimation is a crucial task for virtual reality and accurate robotic manipulation. Category-level 6DoF pose estimation has recently become popular as it improves generalization to a complete category of objects. However, current methods focus on data-driven differential learning, which makes them highly dependent on the quality of the real-world labeled data and limits their ability to generalize to unseen objects. To address this problem, we propose multi-hypothesis (MH) consistency learning (MH6D) for category-level 6-D object pose estimation without using real-world training data. MH6D uses a parallel consistency learning structure, alleviating the uncertainty problem of single-shot feature extraction and promoting self-adaptation of domain to reduce the synthetic-to-real domain gap. Specifically, three randomly sampled pose transformations are first performed in parallel on the input point cloud. An attention-guided category-level 6-D pose estimation network with channel attention (CA) and global feature cross-attention (GFCA) modules is then proposed to estimate the three hypothesized 6-D object poses by extracting and fusing the global and local features effectively. Finally, we propose a novel loss function that considers both the process and the final result information allowing MH6D to perform robust consistency learning. We conduct experiments under two different training data settings (i.e., only synthetic data and synthetic and real-world data) to verify the generalization ability of MH6D. Extensive experiments on benchmark datasets demonstrate that MH6D achieves state-of-the-art (SOTA) performance, outperforming most data-driven methods even without using any real-world data. The code is available at https://github.com/CNJianLiu/MH6D.

RTC_TongueNet: An improved tongue image segmentation model based on DeepLabV3.

Tongue segmentation as a basis for automated tongue recognition studies in Chinese medicine, which has defects such as network degradation and inability to obtain global features, which seriously affects the segmentation effect. This article proposes an improved model RTC_TongueNet based on DeepLabV3, which combines the improved residual structure and transformer and integrates the ECA (Efficient Channel Attention Module) attention mechanism of multiscale atrous convolution to improve the effect of tongue image segmentation. In this paper, we improve the backbone network based on DeepLabV3 by incorporating the transformer structure and an improved residual structure. The residual module is divided into two structures and uses different residual structures under different conditions to speed up the frequency of shallow information mapping to deep network, which can more effectively extract the underlying features of tongue image; introduces ECA attention mechanism after concat operation in ASPP (Atrous Spatial Pyramid Pooling) structure to strengthen information interaction and fusion, effectively extract local and global features, and enable the model to focus more on difficult-to-separate areas such as tongue edge, to obtain better segmentation effect. The RTC_TongueNet network model was compared with FCN (Fully Convolutional Networks), UNet, LRASPP (Lite Reduced ASPP), and DeepLabV3 models on two datasets. On the two datasets, the MIOU (Mean Intersection over Union) and MPA (Mean Pixel Accuracy) values of the classic model DeepLabV3 were higher than those of FCN, UNet, and LRASPP models, and the performance was better. Compared with the DeepLabV3 model, the RTC_TongueNet network model increased MIOU value by 0.9% and MPA value by 0.3% on the first dataset; MIOU increased by 1.0% and MPA increased by 1.1% on the second dataset. RTC_TongueNet model performed best on both datasets. In this study, based on DeepLabV3, we apply the improved residual structure and transformer as a backbone to fully extract image features locally and globally. The ECA attention module is combined to enhance channel attention, strengthen useful information, and weaken the interference of useless information. RTC_TongueNet model can effectively segment tongue images. This study has practical application value and reference value for tongue image segmentation.

Learning a Deep Demosaicing Network for Spike Camera With Color Filter Array.

For capturing dynamic scenes with ultra-fast motion, neuromorphic cameras with extremely high temporal resolution have demonstrated their great capability and potential. Different from the event cameras that only record relative changes in light intensity, spike camera fires a stream of spikes according to a full-time accumulation of photons so that it can recover the texture details for both static areas and dynamic areas. Recently, color spike camera has been invented to record color information of dynamic scenes using a color filter array (CFA). However, demosaicing for color spike cameras is an open and challenging problem. In this paper, we develop a demosaicing network, called CSpkNet, to reconstruct dynamic color visual signals from the spike stream captured by the color spike camera. Firstly, we develop a light inference module to convert binary spike streams to intensity estimates. In particular, a feature-based channel attention module is proposed to reduce the noises caused by quantization errors. Secondly, considering both the Bayer configuration and object motion, we propose a motion-guided filtering module to estimate the missing pixels of each color channel, without undesired motion blur. Finally, we design a refinement module to improve the intensity and details, utilizing the color correlation. Experimental results demonstrate that CSpkNet can reconstruct color images from the Bayer-pattern spike stream with promising visual quality.

Biotic Stress Management in Soil-Less Agriculture Systems: A Deep Learning Approach for Identification of Leaf Miner Pest Infestation

Leaf miner pests pose a serious threat to the productivity, profitability, and sustainability of soil-less tomato cultivation systems. Early and accurate identification of leaf miner infestation is crucial for timely pest control measures. This study presents an efficient approach using attention-based convolutional neural networks for timely identification of this pest infestation. The proposed approach uses both spatial and channel attention modules to enhance the feature extraction capability of the convolutional neural network. The custom model developed was trained using an image dataset collected from tomatoes grown in a hydroponic environment. The different hyper parameters were tuned to get the optimal model performance. The experimental results show that the proposed attention-based CNN model achieved an overall accuracy of 97.87%, 97.10% precision, 98.53% recall, and 97.81% F1-score. Additionally, the model performance was compared with other pre-trained models viz., AlexNet, VGG16, and VGG19, and was found to outperform these state-of-the-art CNN models due to its improved feature extraction capability. The efficiency of the model underlines its potential to be deployed as part of automated pest monitoring systems in hydroponic environments. This work contributes to the development of computer vision and deep learning-based solutions for precision agriculture applications.

UTSRMorph: A Unified Transformer and Superresolution Network for Unsupervised Medical Image Registration.

Complicated image registration is a key issue in medical image analysis, and deep learning-based methods have achieved better results than traditional methods. The methods include ConvNet-based and Transformer-based methods. Although ConvNets can effectively utilize local information to reduce redundancy via small neighborhood convolution, the limited receptive field results in the inability to capture global dependencies. Transformers can establish long-distance dependencies via a self-attention mechanism; however, the intense calculation of the relationships among all tokens leads to high redundancy. We propose a novel unsupervised image registration method named the unified Transformer and superresolution (UTSRMorph) network, which can enhance feature representation learning in the encoder and generate detailed displacement fields in the decoder to overcome these problems. We first propose a fusion attention block to integrate the advantages of ConvNets and Transformers, which inserts a ConvNet-based channel attention module into a multihead self-attention module. The overlapping attention block, a novel cross-attention method, uses overlapping windows to obtain abundant correlations with match information of a pair of images. Then, the blocks are flexibly stacked into a new powerful encoder. The decoder generation process of a high-resolution deformation displacement field from low-resolution features is considered as a superresolution process. Specifically, the superresolution module was employed to replace interpolation upsampling, which can overcome feature degradation. UTSRMorph was compared to state-of-the-art registration methods in the 3D brain MR (OASIS, IXI) and MR-CT datasets (abdomen, craniomaxillofacial). The qualitative and quantitative results indicate that UTSRMorph achieves relatively better performance. The code and datasets used are publicly available at https://github.com/Runshi-Zhang/UTSRMorph.

MDRN: Multi-distillation residual network for efficient MR image super-resolution

<p>Super-resolution (SR) of magnetic resonance imaging (MRI) is gaining increasing attention for being able to provide detailed anatomical information. However, current SR methods often use the complex convolutional network for feature extraction, which is difficult to train and not suitable for limited computation resources in the medical scenario. To tackle these bottlenecks, we propose a multi-distillation residual network (MDRN) for more differential feature refinement, which has a superior trade-off between reconstruction accuracy and computation cost. Specifically, a novel feature multi-distillation residual block with a contrast-aware channel attention module was designed to make the residual features more focused on low-vision information, which maximizes the power of MDRN. Comprehensive experiments demonstrate the superiority of our MDRN over state-of-the-art methods in reconstruction quality and efficiency. Our method outperforms other existing methods in peak signal-noise ratio by up to 0.44–1.82 dB in 4× scale when GPU memory and runtime are lower than in other SR methods. The source code will be available at <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://github.com/Jennieyy/MDRN">https://github.com/Jennieyy/MDRN</ext-link>.</p>

Residual self-calibrated network with multi-scale channel attention for accurate EOG-based eye movement classification.

Recently, Electrooculography-based Human-Computer Interaction (EOG-HCI) technology has gained widespread attention in industrial areas, including assistive robots, augmented reality in gaming, etc. However, as the fundamental step of EOG-HCI, accurate eye movement classification (EMC) still faces a significant challenge, where their constraints in extracting discriminative features limit the performance of most existing works. To address this issue, a Residual Self-Calibrated Network with Multi-Scale Channel Attention (RSCA), focusing on efficient feature extraction and enhancement is proposed. The RSCA network first employs three self-calibrated convolution blocks within a hierarchical residual framework to fully extract the discriminative multi-scale features. Then, a multi-scale channel attention module adaptively weights the learned features to screen out the discriminative representation by aggregating the multi-scale context information along the channel dimension, thus further boosting the performance. Comprehensive experiments were performed using 5 public datasets and 7 prevailing methods for comparative validation. The results confirm that the RSCA network outperforms all other methods significantly, establishing a state-of-the-art benchmark for EOG-based EMC. Furthermore, thorough ablation analyses confirm the effectiveness of the employed modules within the RSCA network, providing valuable insights for the design of EOG-based deep models.

DB-EAC and LSTR: DBnet based seal text detection and Lightweight Seal Text Recognition.

Recognition of the key text of the Chinese seal can speed up the approval of documents, and improve the office efficiency of enterprises or government administrative departments. Due to image blurring and occlusion, the accuracy of Chinese seal recognition is low. In addition, the real dataset is very limited. In order to solve these problems, we improve the differentiable binarization detection algorithm (DBnet) to construct a model DB-ECA for text region detection, and propose a model named LSTR (Lightweight Seal Text Recognition) for text recognition. The efficient channel attention module is added to the differentiable binarization network to solve the feature pyramid conflict, and the convolutional layer network structure is improved to delay downsampling for reducing semantic feature loss. LSTR uses a lightweight CNN more suitable for small-sample generalization, and dynamically fuses positional and visual information through a self-attention-based inference layer to predict the label distribution of feature sequences in parallel. The inference layer not only solves the weak discriminative power of CNN in the shallow layer, but also facilitates CTC (Connectionist Temporal Classification) to accurately align the feature region with the target character. Experiments on the homemade dataset in this paper, DB-ECA compared with the other five commonly used detection models, the precision, recall, F-measure are the best effect of 90.29, 85.17, 87.65, respectively. LSTR compared with the other five kinds of recognition models in the last three years, to achieve the highest effect of accuracy 91.29%, and has the advantages of a small number of parameters and fast inference. The experimental results fully prove the innovation and effectiveness of our model.

Autonomous Identification of Bridge Concrete Cracks Using Unmanned Aircraft Images and Improved Lightweight Deep Convolutional Networks

The identification of the development of structural defects is an important part of bridge structure damage diagnosis, and cracks are considered the most typical and highly dangerous structural disease. However, existing deep learning‐based methods are mostly aimed at the scene of concrete cracks, while they rarely focus on designing network architectures to improve the vision‐based model performance from the perspective of unmanned aircraft system (UAS) inspection, which leads to a lack of specificity. Because of this, this study proposes a novel lightweight deep convolutional neural network‐based crack pixel‐level segmentation network for UAS‐based inspection scenes. Firstly, the classical encoder‐decoder architecture UNET is utilized as the base model for bridge structural crack identification, and the hourglass‐shaped depthwise separable convolution is introduced to replace the traditional convolutional operation in the UNET model to reduce model parameters. Then, a kind of lightweight and efficient channel attention module is used to improve model feature fuzzy ability and segmentation accuracy. We conducted a series of experiments on bridge structural crack detection tasks by utilizing a long‐span bridge as the research item. The experimental results show that the constructed method achieves an effective balance between reasoning accuracy and efficiency with the value of 97.62% precision, 97.23% recall, 97.42% accuracy, and 93.25% IOU on the bridge concrete crack datasets, which are significantly higher than those of other state‐of‐the‐art baseline methods. It can be inferred that the application of hourglass‐shaped depth‐separable volumes can actively reduce basic model parameters. Moreover, the lightweight and efficient attention modules can achieve local cross‐channel interaction without dimensionality reduction and improve the network segmentation performance.

Channel-Specific and Spatial Residual Attention Network for Medical Image Denoising.

Medical image quality is crucial for physicians to ensure accurate diagnosis and therapeutic strategies. However, due to the interference of noise, there are often various types of noise and artifacts in medical images. This not only damages the visual clarity of images, but also reduces the accuracy of information extraction. Considering that the edges of medical images are rich in high-frequency information, to enhance the quality of medical images, a dual attention mechanism, the channel-specific and spatial residual attention network (CSRAN) in the U-Net framework is proposed. The CSRAN seamlessly integrates the U-Net architecture with channel-wise and spatial feature attention (CSAR) modules, as well as low-frequency channel attention modules. Combined with the two modules, the ability of medical image processing to extract high-frequency features is improved, thereby significantly improving the edge effects and clarity of reconstructed images. This model can present better performance in capturing high-frequency information and spatial structures in medical image denoising and super-resolution reconstruction tasks. It cannot only enhance the ability to extract high-frequency features and strengthen its nonlinear representation capability, but also endow strong edge detection capabilities of the model. The experimental results further prove the superiority of CSRAN in medical image denoising and super-resolution reconstruction tasks.

Intelligent rolling bearing compound fault diagnosis based on frequency-domain Gramian angular field and convolutional neural networks with imbalanced data

The effective fault feature extraction is the core of rolling bearing fault diagnosis. However, rolling bearings usually operate in normal state and fault duration is very short, which will cause imbalance in fault diagnosis data, thus leading to difficulty in fault feature extraction and low diagnosis accuracy. Meanwhile, mutual interference between multiple fault responses will also lead to poor diagnosis performance. To solve these issues, a novel compound fault diagnosis method with imbalanced data based on frequency-domain Gramian angular field (FGAF) and convolutional neural networks optimized by instance normalization and efficient channel attention (IECNN) is proposed. Firstly, FGAF is adopted to map frequency-domain features of fault signals to the polar coordinate to obtain 2D FGAF feature spectrum. Secondly, an instance normalization module is established to reduce internal covariant shift caused by data distribution discrepancy and improve generalization ability. An efficient channel attention module is constructed to further excavate fault features and improve anti-interference ability. Finally, experiments are conducted under imbalanced dataset and imbalance intensified dataset, and the average accuracy of 99.91% and 99.92% were obtained, respectively, which shows the proposed method has better resistance to data imbalance.