Frequency-based CNN and attention module for acoustic scene classification

Polyphonic sound event detection (SED) is an emerging area with many applications for smart disaster safety, security, life logging, etc. This paper proposes a two-stage polyphonic SED model when strongly labeled data are limited but weakly labeled and unlabeled data are available. The first stage of the proposed SED model is constructed by a residual convolutional recurrent neural network (RCRNN)-based mean teacher model with convolutional block attention module (CBAM)-based attention. Then, the second stage fine-tunes the student model from the first stage by applying the proposed semi-supervised loss function to accommodate the noisy targets of weakly labeled and unlabeled data. The proposed SED model is applied to both Detection and Classification of Acoustic Scenes and Events (DCASE) 2019 Challenge Task 4 and DCASE 2020 Challenge Task 4, and its performance is compared with those of the baseline and top-ranked models from both challenges by measuring the F1-score and polyphonic sound detection score (PSDS). The experiments show that the RCRNN-based first-stage model with CBAM-based attention achieves a higher F1-score and PSDS than the baseline and top-ranked models for both challenges. Furthermore, the proposed two-stage SED model with the semi-supervised loss function improves the F1-score by 6.1% and 4.6% compared to the top-ranked models from DCASE 2019 and 2020, respectively.

Brain-computer interfaces (BCI) allow for the brain to communicate with electronic devices. Concerning the BCI paradigms, motor imagery uses brain signals to decode an imagined movement. However, this is a hard task given the low signal-to-noise ratio. Usually, the main steps in BCI models are pre-processing, feature extraction, and classification. In recent years, Convolutional Neural Networks (CNNs) have been gaining relevance in several areas of science due to their feature extraction, translation invariance, and parameter sharing capa-bilities. Another, more recent way of feature extraction is using attention mechanisms, which are layers of neural networks based on human attention and have the ability to highlight important features. A variation of the attention mechanism is the Convolutional Block Attention Module, which combines the CNN structure with the attention mechanism. In this work, we propose a new model that joins the core architecture of EEGNet, a compact CNN widely used in the literature, with the Convolutional Block Attention Module and residual connections. The residual connections were introduced to lower data degradation throughout the model. The results highlight the residual connection's importance for the performance of the model. The proposed model obtained a kappa result 5.2% better than the EEGNet with a p-value less than 0.01 on BCI Competition IV dataset 2a, which is a well-known dataset for Motor Imagery. Furthermore, the proposal was better than EEGNet for most subjects and had the best-worst case.

In recent years, attention mechanisms have shown great potential in various computer vision tasks. However, most existing methods focus on developing more complex attention modules for better performance, which inevitably increases the complexity of the model. To overcome performance and complexity tradeoffs, this paper proposes efficient spatial and channel attention (ESCA), a symmetric, comprehensive, and efficient attention module. By analyzing squeeze-and-excitation (SE), convolutional block attention module (CBAM), coordinate attention (CA), and efficient channel attention (ECA) modules, we abandon the dimension-reduction operation of SE module, verify the negative impact of global max pooling (GMP) on the model, and apply a local cross-channel interaction strategy without dimension reduction to learn attention. We not only care about the channel features of the image, we also care about the spatial location of the target on the image, and we take into account the effectiveness of channel attention, so we designed the symmetric ESCA module. The ESCA module is effective, as demonstrated by its application in the ResNet-50 classification benchmark. With 26.26 M parameters and 8.545 G FLOPs, it introduces a mere 0.14% increment in FLOPs while achieving over 6.33% improvement in Top-1 accuracy and exceeding 3.25% gain in Top-5 accuracy. We perform image classification and object detection tasks on ResNet, MobileNet, YOLO, and other architectures on popular datasets such as Mini ImageNet, CIFAR-10, and VOC 2007. Experiments show that ESCA can achieve great improvement in model accuracy at a very small cost, and it performs well among similar models.

Acquiring the growth information of tobacco plants quickly and accurately is an important prerequisite for their daily management. In view of the problem that the convolutional neural network is easily affected by background information, this study introduces the Convolutional Block Attention Module into the U <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2-</sup> Net. This method is mainly composed of an encoder and a decoder. Before the feature map is decoded, it will go through the Convolutional Block Attention Module to filter more important features and inhibit the invalid features. In the decoder, in order to fuse the features of objects in different scales, the feature maps in the decoder are up-sampled and then concatenated. This information aggregation solves the problem of difficult object detection at different scales to some extent. The results of this method are compared with four current mainstream deep learning models(U <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> -Net, DeeplabV3+, PSP-Net, U-Net). The result shows: 1) The U <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> -Net network based on the Convolutional Block Attention Module has the highest extraction accuracy for the UAV visible light remote senseing of tobacco(accuracy was 98.53%, recall was 96.00%,F1-Score was 96.38%, kappa coefficient was 0.95, IOU was 93.02%); 2)In the extraction of the three test areas, the extraction value of this method is highly consistent with the label area value(the area ratios of a total of 13 plants in the three regions were kept between 90% and 100%),the effect is much higher than that of DeeplabV3+ and PSP-Net, and slightly higher than the extraction effect of U <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> -Net and U-Net. 3)In the segmentation of three typical canopy closure regions of A, B, and C, most of the accuracy factors in different regions can reach 94% and above. It can be seen that the method in this study has certain advantages in feature extraction and has certain value in practical application.

Modulation recognition of single‐component radar signals has been widely studied in the fewer years. Meanwhile, as the electromagnetic environment in the actual battlefields is ever‐increasingly complicated, the demand for the modulation recognition of multi‐component radar signals is also increasing. However, the modulation recognition of single‐component radar signals performs poorly on multi‐component radar signals. To address this problem, a novel modulation recognition method of multi‐component radar signals based on the deep convolutional neural network with a convolutional block attention module (DCNN‐CBAM) is proposed, and the network architecture mainly includes mobile‐inverted bottleneck convolution (MBConv), Fused‐MBConv, and convolutional block attention module (CBAM). An adaptive optimization algorithm Adam with weighted gradient and dynamic bound is proposed for accelerating convergence speed and generalisation performance. The proposed method can not only recognise dual‐component and three‐component radar signals but also four‐component radar signals. The experimental results demonstrate that the recognition accuracies of all radar signals modulation types are more than 83.57% at −8 dB for dual‐component radar signals, 67.41% for three‐component radar signals, and 57.95% for four‐component radar signals, which verify the better recognition performance of DCNN‐CBAM for multi‐component radar signals. The recognition accuracy of the proposed DCNN‐CBAM for dual‐component radar signals can be up to 95.95% at −6 dB, which shows superior recognition performance over the residual method. This work provides an essential guidance for enhancing the recognition performance of multi‐component radar signals.

In order to realize the quantitative assessment of muscle strength in hand function rehabilitation and then formulate scientific and effective rehabilitation training strategies, this paper constructs a multi-scale convolutional neural network (MSCNN) - convolutional block attention module (CBAM) - bidirectional long short-term memory network (BiLSTM) muscle strength prediction model to fully explore the spatial and temporal features of the data and simultaneously suppress useless features, and finally achieve the improvement of the accuracy of the muscle strength prediction model. To verify the effectiveness of the model proposed in this paper, the model in this paper is compared with traditional models such as support vector machine (SVM), random forest (RF), convolutional neural network (CNN), CNN - squeeze excitation network (SENet), MSCNN-CBAM and MSCNN-BiLSTM, and the effect of muscle strength prediction by each model is investigated when the hand force application changes from 40% of the maximum voluntary contraction force (MVC) to 60% of the MVC. The research results show that as the hand force application increases, the effect of the muscle strength prediction model becomes worse. Then the ablation experiment is used to analyze the influence degree of each module on the muscle strength prediction result, and it is found that the CBAM module plays a key role in the model. Therefore, by using the model in this article, the accuracy of muscle strength prediction can be effectively improved, and the characteristics and laws of hand muscle activities can be deeply understood, providing assistance for further exploring the mechanism of hand functions .

With the continuous development of deep learning technologies, fault prediction techniques based on various neural networks have been evolving. The deep learning modules based on U-Net residual networks have shown significant advantages in both learning efficiency and effectiveness. In this paper, we propose a deep learning model that integrates a 3D U-Net residual architecture, Convolutional Block Attention Module (CBAM), and Multi-scale Enhanced Global Attention (MEGA) module for automatic seismic fault detection and segmentation. This model can effectively handle complex 3D seismic data, fully exploiting both spatial and channel information, significantly improving the prediction accuracy for small faults, while only slightly increasing the computational cost.Firstly, the model uses the 3D U-Net as the backbone framework, where the residual blocks (BasicRes) extract features through multiple convolution layers. The CBAM module is incorporated to apply attention weighting, enhancing the model's ability to focus on critical information. The CBAM module combines channel attention and spatial attention, effectively adjusting the importance of feature maps from different dimensions, enabling the model to identify potential fault features in complex seismic data.Secondly, the MEGA module is introduced into the model, which further improves the model's feature representation ability by fusing multi-scale features and applying a global attention mechanism. By weighting global information, the MEGA module helps the model better capture key seismic fault features during feature fusion. This design allows the model to focus not only on local details but also to fully utilize the global contextual information in 3D data, thereby enhancing the accuracy of fault detection.After validation, the model achieved promising results in seismic fault detection tasks, automatically identifying and segmenting fault structures in seismic data. The accuracy was improved from 80% with the original 3D U-Net residual network to 85%-87%. This provides strong support for applications such as seismic exploration and subsurface imaging.Keywords: Seismic Fault Detection, 3D U-Net, Convolutional Block Attention Module (CBAM), Multi-scale Enhanced Global Attention (MEGA), Deep Learning

In Acoustic Scene Classification (ASC) two major approaches have been followed. While one utilizes engineered features such as mel-frequency-cepstral-coefficients (MFCCs), the other uses learned features that are the outcome of an optimization algorithm. I-vectors are the result of a modeling technique that usually takes engineered features as input. It has been shown that standard MFCCs extracted from monaural audio signals lead to i-vectors that exhibit poor performance, especially on indoor acoustic scenes. At the same time, Convolutional Neural Networks (CNNs) are well known for their ability to learn features by optimizing their filters. They have been applied on ASC and have shown promising results. In this paper, we first propose a novel multi-channel i-vector extraction and scoring scheme for ASC, improving their performance on indoor and outdoor scenes. Second, we propose a CNN architecture that achieves promising ASC results. Further, we show that i-vectors and CNNs capture complementary information from acoustic scenes. Finally, we propose a hybrid system for ASC using multi-channel i-vectors and CNNs by utilizing a score fusion technique. Using our method, we participated in the ASC task of the DCASE-2016 challenge. Our hybrid approach achieved 1st rank among 49 submissions, substantially improving the previous state of the art.

MRI image classification of brain tumors is critical for accurate and early diagnosis. New developments in deep learning have revealed that inserting attention mechanisms into convolutional neural networks can greatly improve classification performance. This work assesses the effectiveness of Squeeze-and-Excitation (SE) and Convolutional Block Attention Module (CBAM) integrated with the ResNet50 model for brain tumor classification. The SE module recalibrates channel-wise feature responses, while CBAM enhances feature representation by combining channel and spatial attention. The SE and CBAM are combined with ResNet50, greatly increasing classification accuracy, precision, and recall when compared to the basic model, according to experimental results on MRI datasets. The suggested model employs attention mechanisms to focus valuable information selectively and suppress irrelevant information. The experiments on publicly available brain tumor datasets display great improvements over basic CNN models, with precision, recall, accuracy, as well as F1 score at 0.9914, 0.9903, 0.9945, and 0.9908, respectively. From these results, the importance of attention mechanisms in deep learning models for medical imaging is highlighted, which suggests that SE and CBAM modules can be available as more dependable and effective instruments for brain tumor classification in clinical settings. Future studies should investigate transformer-based and hybrid attention techniques to enhance automated brain tumor categorization.

Improving convolutional neural network learning based on a hierarchical bezier generative model for stenosis detection in X-ray images

Weeds are one of the primary factors that reduce crop productivity by competing for nutrients and water, causing the plant to lose weight and resulting in reduced grain yield. Traditional agricultural practices often rely on uniform herbicide application, which can contaminate soil and raise costs. On agricultural land, selective weed treatment are an efficient and cost-effective way to control weeds that require a deep learning-based crop and weed segmentation system. Many existing crop and weed segmentation research works focus on achieving precise crop and weed segmentation results, rather than building lightweight models to deploy on edge devices. To attain this, we develop an effective and efficient convolutional neural network, namely the Concatenated Attention U-Net with Convolutional Block Attention Module (CAUC). By integrating Linear Concatenated Blocks (LCB), Attention Gate (AG) connections, and Convolutional Block Attention Module (CBAM), the proposed model efficiently utilizes feature maps among its architectural components to achieve superior performance. Depth-wise convolution layers and 1 × 1 convolution layers in LCBs reduce computational complexity. To enable the proposed model to identify the weed portions in multiple crop fields, we integrated three datasets in this research work, namely the Crop/Weed Field Image Dataset (CWFID), Sugar Beet, and Sunflower datasets. Experimental results on carrot, sugar beet, and sunflower crop datasets demonstrate high Accuracy (99.09%), MIoU (81.02%), and F1-score (99.06%), with a modest model size (5.6 MB) and computational parameters (0.377 million). We developed a lightweight computer vision application (13.7 MB) to demonstrate the model’s efficacy on low-computational devices.

Wind turbine rolling bearings are crucial components for ensuring the reliability and stability of wind power systems. Their failure can lead to significant economic losses and equipment downtime. Therefore, the accurate diagnosis of bearing faults is of great importance. Although existing deep learning fault diagnosis methods have achieved certain results, they still face limitations such as inadequate feature extraction capabilities, insufficient generalization to complex working conditions, and ineffective multi-scale feature capture. To address these issues, this paper proposes an advanced fault diagnosis method named the two-stream feature fusion convolutional neural network (TSFFResNet-Net). Firstly, the proposed method combines the advantages of one-dimensional convolutional neural networks (1D-ResNet) and two-dimensional convolutional neural networks (2D-ResNet). It transforms one-dimensional vibration signals into two-dimensional images through the empirical wavelet transform (EWT) method. Then, parallel convolutional kernels in 1D-ResNet and 2D-ResNet are used to extract multi-scale features, respectively. Next, the Convolutional Block Attention Module (CBAM) is introduced to enhance the network’s ability to capture key features by focusing on important features in specific channels or spatial areas. After feature fusion, CBAM is introduced again to further enhance the effect of feature fusion, ensuring that the features extracted by different network branches can be effectively integrated, ultimately providing more accurate input features for the classification task of the fully connected layer. The experimental results demonstrate that the proposed method outperforms other traditional methods and advanced convolutional neural network models on different datasets. Compared with convolutional neural network models such as LeNet-5, AlexNet, and ResNet, the proposed method achieves a significantly higher accuracy on the test set, with a stable accuracy of over 99%. Compared with other models, it shows better generalization and stability, effectively improving the overall performance of rolling bearing vibration signal fault diagnosis. The method provides an effective solution for the intelligent fault diagnosis of wind turbine rolling bearings.

Acoustic scene classification and related tasks have been dominated by Convolutional Neural Networks (CNNs). Top-performing CNNs use mainly audio spectograms as input and borrow their architectural design primarily from computer vision. A recent study has shown that restricting the receptive field (RF) of CNNs in appropriate ways is crucial for their performance, robustness and generalization in audio tasks. One side effect of restricting the RF of CNNs is that more frequency information is lost. In this paper, we perform a systematic investigation of different RF configuration for various CNN architectures on the DCASE 2019 Task 1.A dataset. Second, we introduce Frequency Aware CNNs to compensate for the lack of frequency information caused by the restricted RF, and experimentally determine if and in what RF ranges they yield additional improvement. The result of these investigations are several well-performing submissions to different tasks in the DCASE 2019 Challenge.

Early detection of breast cancer is crucial for effective treatment, making reliable screening methods essential for identifying initial symptoms.Various imaging techniques, such as mammography, ultrasound, and thermography, are commonly used for screening breast cancer.Traditional diagnostic methods, have significantly contributed to early detection but are often hampered by limitations, including false positives and false negatives, which can lead to unnecessary procedures or delayed treatment.This study proposes a model based on the ResNet50 Convolutional Neural Network (CNN) to detect breast cancer from mammograms, incorporating the Convolutional Block Attention Module (CBAM) as an attention mechanism.The model explores varying numbers of CBAM blocks to determine the optimal configuration while mitigating the risk of overfitting.The performance of the proposed model was evaluated using the dataset obtained from the PatchCamelyon (PCam) dataset by comparing its Area Under the Curve (AUC) scores with those of existing models.The results demonstrate that the ResNet50 model integrated with CBAM blocks shows improved performance with an increased number of blocks, with the model utilizing 8 CBAM blocks achieving the highest performance in terms of AUC, precision, recall, and accuracy.Compared to existing models, the proposed ResNet50 model with 8 CBAM blocks outperformed all others across all performance metrics, confirming the effectiveness of the CBAM integration and the overall model architecture in improving breast cancer detection.

Acoustic scene classification (ASC) is a problem related to the field of machine listening whose objective is to classify/tag an audio clip in a predefined label describing a scene location (e. g. park, airport, etc.). Many state-of-the-art solutions to ASC incorporate data augmentation techniques and model ensembles. However, considerable improvements can also be achieved only by modifying the architecture of convolutional neural networks (CNNs). In this work we propose two novel squeeze-excitation blocks to improve the accuracy of a CNN-based ASC framework based on residual learning. The main idea of squeeze-excitation blocks is to learn spatial and channel-wise feature maps independently instead of jointly as standard CNNs do. This is usually achieved by some global grouping operators, linear operators and a final calibration between the input of the block and its obtained relationships. The behavior of the block that implements such operators and, therefore, the entire neural network, can be modified depending on the input to the block, the established residual configurations and the selected non-linear activations. The analysis has been carried out using the TAU Urban Acoustic Scenes 2019 dataset (https://zenodo.org/record/2589280) presented in the 2019 edition of the DCASE challenge. All configurations discussed in this document exceed the performance of the baseline proposed by the DCASE organization by 13\% percentage points. In turn, the novel configurations proposed in this paper outperform the residual configurations proposed in previous works.