Unsupervised multi-modal image translation based on the squeeze-and-excitation mechanism and feature attention module

The unsupervised multi-modal image translation is an emerging domain of computer vision whose goal is to transform an image from the source domain into many diverse styles in the target domain.However,the multi-generator mechanism is employed among the advanced approaches available to model different domain mappings,which results in inefficient training of neural networks and pattern collapse,leading to inefficient generation of image diversity.To address this issue,this paper introduces a multi-modal unsupervised image translation framework that uses a generator to perform multi-modal image translation.Specifically,firstly,the domain code is introduced in this paper to explicitly control the different generation tasks.Secondly,this paper brings in the squeeze-and-excitation(SE)mechanism and feature attention(FA)module.Finally,the model integrates multiple optimization objectives to ensure efficient multi-modal translation.This paper performs qualitative and quantitative experiments on multiple non-paired benchmark image translation datasets while demonstrating the benefits of the proposed method over existing technologies.Overall,experimental results have shown that the proposed method is versatile and scalable.

Traffic Sign Recognition for Autonomous Vehicle Using Optimized YOLOv7 and Convolutional Block Attention Module

The infrastructure and construction of roads are crucial for the economic and social development of a region,but traffic-related challenges like accidents and congestion persist.Artificial Intelligence(AI)and Machine Learning(ML)have been used in road infrastructure and construction,particularly with the Internet of Things(IoT)devices.Object detection in Computer Vision also plays a key role in improving road infrastructure and addressing trafficrelated problems.This study aims to use You Only Look Once version 7(YOLOv7),Convolutional Block Attention Module(CBAM),the most optimized object-detection algorithm,to detect and identify traffic signs,and analyze effective combinations of adaptive optimizers like Adaptive Moment estimation(Adam),Root Mean Squared Propagation(RMSprop)and Stochastic Gradient Descent(SGD)with the YOLOv7.Using a portion of German traffic signs for training,the study investigates the feasibility of adopting smaller datasets while maintaining high accuracy.The model proposed in this study not only improves traffic safety by detecting traffic signs but also has the potential to contribute to the rapid development of autonomous vehicle systems.The study results showed an impressive accuracy of 99.7%when using a batch size of 8 and the Adam optimizer.This high level of accuracy demonstrates the effectiveness of the proposed model for the image classification task of traffic sign recognition.

Simplified Inception Module Based Hadamard Attention Mechanism for Medical Image Classification

Medical image classification has played an important role in the medical field, and the related method based on deep learning has become an important and powerful technique in medical image classification. In this article, we propose a simplified inception module based Hadamard attention (SI + HA) mechanism for medical image classification. Specifically, we propose a new attention mechanism: Hadamard attention mechanism. It improves the accuracy of medical image classification without greatly increasing the complexity of the model. Meanwhile, we adopt a simplified inception module to improve the utilization of parameters. We use two medical image datasets to prove the superiority of our proposed method. In the BreakHis dataset, the AUCs of our method can reach 98.74%, 98.38%, 98.61% and 97.67% under the magnification factors of 40×, 100×, 200× and 400×, respectively. The accuracies can reach 95.67%, 94.17%, 94.53% and 94.12% under the magnification factors of 40×, 100×, 200× and 400×, respectively. In the KIMIA Path 960 dataset, the AUCs and accuracy of our method can reach 99.91% and 99.03%. It is superior to the currently popular methods and can significantly improve the effectiveness of medical image classification.

Two-Layer Attention Feature Pyramid Network for Small Object Detection

Effective small object detection is crucial in various applications including urban intelligent transportation and pedestrian detection.However,small objects are difficult to detect accurately because they contain less information.Many current methods,particularly those based on Feature Pyramid Network(FPN),address this challenge by leveraging multi-scale feature fusion.However,existing FPN-based methods often suffer from inadequate feature fusion due to varying resolutions across different layers,leading to suboptimal small object detection.To address this problem,we propose the Two-layerAttention Feature Pyramid Network(TA-FPN),featuring two key modules:the Two-layer Attention Module(TAM)and the Small Object Detail Enhancement Module(SODEM).TAM uses the attention module to make the network more focused on the semantic information of the object and fuse it to the lower layer,so that each layer contains similar semantic information,to alleviate the problem of small object information being submerged due to semantic gaps between different layers.At the same time,SODEM is introduced to strengthen the local features of the object,suppress background noise,enhance the information details of the small object,and fuse the enhanced features to other feature layers to ensure that each layer is rich in small object information,to improve small object detection accuracy.Our extensive experiments on challenging datasets such as Microsoft Common Objects inContext(MSCOCO)and Pattern Analysis Statistical Modelling and Computational Learning,Visual Object Classes(PASCAL VOC)demonstrate the validity of the proposedmethod.Experimental results show a significant improvement in small object detection accuracy compared to state-of-theart detectors.

Improved multi-scale inverse bottleneck residual network based on triplet parallel attention for apple leaf disease identification

Accurate diagnosis of apple leaf diseases is crucial for improving the quality of apple production and promoting the development of the apple industry. However, apple leaf diseases do not differ significantly from image texture and structural information. The difficulties in disease feature extraction in complex backgrounds slow the related research progress. To address the problems, this paper proposes an improved multi-scale inverse bottleneck residual network model based on a triplet parallel attention mechanism, which is built upon ResNet-50, while improving and combining the inception module and ResNext inverse bottleneck blocks, to recognize seven types of apple leaf(including six diseases of alternaria leaf spot, brown spot, grey spot, mosaic, rust, scab, and one healthy). First, the 3×3 convolutions in some of the residual modules are replaced by multi-scale residual convolutions, the convolution kernels of different sizes contained in each branch of the multi-scale convolution are applied to extract feature maps of different sizes, and the outputs of these branches are multi-scale fused by summing to enrich the output features of the images. Second, the global layer-wise dynamic coordinated inverse bottleneck structure is used to reduce the network feature loss. The inverse bottleneck structure makes the image information less lossy when transforming from different dimensional feature spaces. The fusion of multi-scale and layer-wise dynamic coordinated inverse bottlenecks makes the model effectively balances computational efficiency and feature representation capability, and more robust with a combination of horizontal and vertical features in the fine identification of apple leaf diseases. Finally, after each improved module, a triplet parallel attention module is integrated with cross-dimensional interactions among channels through rotations and residual transformations, which improves the parallel search efficiency of important features and the recognition rate of the network with relatively small computational costs while the dimensional dependencies are improved. To verify the validity of the model in this paper, we uniformly enhance apple leaf disease images screened from the public data sets of Plant Village, Baidu Flying Paddle, and the Internet. The final processed image count is 14,000. The ablation study, pre-processing comparison, and method comparison are conducted on the processed datasets. The experimental results demonstrate that the proposed method reaches 98.73% accuracy on the adopted datasets, which is 1.82% higher than the classical ResNet-50 model, and 0.29% better than the apple leaf disease datasets before preprocessing. It also achieves competitive results in apple leaf disease identification compared to some state-ofthe-art methods.

ANC: Attention Network for COVID-19 Explainable Diagnosis Based on Convolutional Block Attention Module

Aim: To diagnose COVID-19 more efficiently and more correctly, this study proposed a novel attention network forCOVID-19 (ANC). Methods: Two datasets were used in this study. An 18-way data augmentation was proposed toavoid overfitting. Then, convolutional block attention module (CBAM) was integrated to our model, the structureof which is fine-tuned. Finally, Grad-CAM was used to provide an explainable diagnosis. Results: The accuracyof our ANC methods on two datasets are 96.32% ± 1.06%, and 96.00% ± 1.03%, respectively. Conclusions: Thisproposed ANC method is superior to 9 state-of-the-art approaches.

基于改进I-Attention U-Net的锌浮选泡沫图像分割算法

针对泡沫图像的高度复杂性导致其难以被准确分割的难题,本文提出了一种新的I-Attention U-Net网络用于泡沫图像分割.该算法以U-Net网络作为主干网络,使用Inception模块替换第一卷积池化层来提取泡沫图像的多尺度、多层次浅层特征信息;引入金字塔池化模块,通过对不同尺度的特征图求和来提升分割效果;并对自注意力门控单元进行改进,使注意力单元更适合于浮选泡沫图像的分割,强化深层特征的重要性并对不同尺寸的泡沫边界进行强化学习.研究结果表明:本文所提出算法的Jaccard系数为91.73%,Dice系数为95.66%.与同类其他分割算法结果相比,Jaccard系数及Dice系数分别提高了1.59%、0.88%.该模型能够较好地对锌浮选泡沫图像进行分割,解决欠分割与过分割的问题,为后续的泡沫特征提取奠定基础.此外,该方法检测时间和模型参数少,具备可以部署在工业现场计算机的能力,有一定的实际应用价值.

Bilateral U-Net semantic segmentation with spatial attention mechanism

Aiming at the problem that the existing models have a poor segmentation effect on imbalanced data sets with small-scale samples,a bilateral U-Net network model with a spatial attention mechanism is designed.The model uses the lightweight MobileNetV2 as the backbone network for feature hierarchical extraction and proposes an Attentive Pyramid Spatial Attention(APSA)module compared to the Attenuated Spatial Pyramid module,which can increase the receptive field and enhance the information,and finally adds the context fusion prediction branch that fuses high-semantic and low-semantic prediction results,and the model effectively improves the segmentation accuracy of small data sets.The experimental results on the CamVid data set show that compared with some existing semantic segmentation networks,the algorithm has a better segmentation effect and segmentation accuracy,and its mIOU reaches 75.85%.Moreover,to verify the generality of the model and the effectiveness of the APSA module,experiments were conducted on the VOC 2012 data set,and the APSA module improved mIOU by about 12.2%.

Single Image Deraining Using Dual Branch Network Based on Attention Mechanism for IoT

Extracting useful details from images is essential for the Internet of Things project.However,in real life,various external environments,such as badweather conditions,will cause the occlusion of key target information and image distortion,resulting in difficulties and obstacles to the extraction of key information,affecting the judgment of the real situation in the process of the Internet of Things,and causing system decision-making errors and accidents.In this paper,we mainly solve the problem of rain on the image occlusion,remove the rain grain in the image,and get a clear image without rain.Therefore,the single image deraining algorithm is studied,and a dual-branch network structure based on the attention module and convolutional neural network(CNN)module is proposed to accomplish the task of rain removal.In order to complete the rain removal of a single image with high quality,we apply the spatial attention module,channel attention module and CNN module to the network structure,and build the network using the coder-decoder structure.In the experiment,with the structural similarity(SSIM)and the peak signal-to-noise ratio(PSNR)as evaluation indexes,the training and testing results on the rain removal dataset show that the proposed structure has a good effect on the single image deraining task.

基于Inception-LSTM-Attention的冷水机组传感器偏差故障诊断方法

为提升传统的冷水机组传感器偏差故障诊断方法的特征提取效果及故障诊断准确率,提出一种基于Inception模块和融合注意力机制(Attention)的长短时记忆网络(LSTM)相结合(Inception-LSTM-Attention)的冷水机组传感器偏差故障诊断方法。该方法通过Inception模块从冷水机组传感器时序数据中提取多尺度的实时特征,并利用LSTM学习传感器时序数据中存在的时间相关关系;通过在LSTM中融合注意力机制来保证其最终的输出综合了各个时间节点的输出,提升重要信息的影响程度,最大化保留时序数据的全局信息。同时,设计跳跃连接支路缓解网络中存在的梯度消失问题。最后,使用冷水机组实验平台的传感器实测数据对所提方法进行实验验证。研究结果表明:本文方法对于压力类、温度类各传感器的偏差故障诊断平均准确率均在94%以上;对于各传感器中较小偏差故障的故障诊断准确率均在87.6%以上;与主成分分析、卷积神经网络、Inception以及Inception-LSTM这4种方法相比,Inception-LSTM-Attention模型的传感器偏差故障诊断准确率更高。

Gear Pitting Measurement by Multi-Scale Splicing Attention U-Net

The judgment of gear failure is based on the pitting area ratio of gear.Traditional gear pitting calculation method mainly rely on manual visual inspection.This method is greatly affected by human factors,and is greatly affected by the working experience,training degree and fatigue degree of the detection personnel,so the detection results may be biased.The non-contact computer vision measurement can carry out non-destructive testing and monitoring under the working condition of the machine,and has high detection accuracy.To improve the measurement accuracy of gear pitting,a novel multi-scale splicing attention U-Net(MSSA U-Net)is explored in this study.An image splicing module is first proposed for concatenating the output feature maps of multiple convolutional layers into a splicing feature map with more semantic information.Then,an attention module is applied to select the key features of the splicing feature map.Given that MSSA U-Net adequately uses multi-scale semantic features,it has better segmentation performance on irregular small objects than U-Net and attention U-Net.On the basis of the designed visual detection platform and MSSA U-Net,a methodology for measuring the area ratio of gear pitting is proposed.With three datasets,experimental results show that MSSA U-Net is superior to existing typical image segmentation methods and can accurately segment different levels of pitting due to its strong segmentation ability.Therefore,the proposed methodology can be effectively applied in measuring the pitting area ratio and determining the level of gear pitting.

An Efficient Indoor Localization Based on Deep Attention Learning Model

Indoor localization methods can help many sectors,such as healthcare centers,smart homes,museums,warehouses,and retail malls,improve their service areas.As a result,it is crucial to look for low-cost methods that can provide exact localization in indoor locations.In this context,imagebased localization methods can play an important role in estimating both the position and the orientation of cameras regarding an object.Image-based localization faces many issues,such as image scale and rotation variance.Also,image-based localization’s accuracy and speed(latency)are two critical factors.This paper proposes an efficient 6-DoF deep-learning model for image-based localization.This model incorporates the channel attention module and the Scale PyramidModule(SPM).It not only enhances accuracy but also ensures the model’s real-time performance.In complex scenes,a channel attention module is employed to distinguish between the textures of the foregrounds and backgrounds.Our model adapted an SPM,a feature pyramid module for dealing with image scale and rotation variance issues.Furthermore,the proposed model employs two regressions(two fully connected layers),one for position and the other for orientation,which increases outcome accuracy.Experiments on standard indoor and outdoor datasets show that the proposed model has a significantly lower Mean Squared Error(MSE)for both position and orientation.On the indoor 7-Scenes dataset,the MSE for the position is reduced to 0.19 m and 6.25°for the orientation.Furthermore,on the outdoor Cambridge landmarks dataset,the MSE for the position is reduced to 0.63 m and 2.03°for the orientation.According to the findings,the proposed approach is superior and more successful than the baseline methods.

基于STFT和CNN-Attention的配电终端采集模块故障诊断研究

针对复杂工况运行环境下配电终端采集模块故障类型难以识别的问题,提出一种基于短时傅里叶变换(STFT)、卷积神经网络和注意力机制(CNN-Attention)的配电终端采集模块故障诊断方法。首先,分析配电终端采集模块不同故障类型会产生的对应故障数据,建立故障数据集。然后,基于STFT提取故障数据的故障时频特征以形成时频图,采用CNN-Attention模型对时频图进行故障诊断与匹配。算例分析表明,CNN-Attention的故障检测准确率为97.31%,相较于CNN和极限学习机(ELM)模型,故障诊断准确率分别提升了1.22%和4.4%。Attention机制能够有效解决CNN在特征提取时产生的冗余信息导致模型训练慢、难以收敛的问题。该研究实现了配电终端采集模块具体故障类型的准确识别,能为后续配电终端的运维提供参考。

Attention Res-Unet:一种高效阴影检测算法

图像中阴影像素的存在会导致图像内容的不确定性,对计算机视觉任务有害,因此常将阴影检测作为计算机视觉算法的预处理步骤.提出全新的阴影检测网络结构,通过结合输入图像中包含的语义信息和像素之间的关联,提升网络性能.使用预训练后的深层网络ResNeXt101作为特征提取前端,提取图像的语义信息,并结合U-net的设计思路,搭建网络结构,完成特征层的上采样过程.在输出层之前使用非局部操作,为每一个像素提供全局信息,建立像素与像素之间的联系.设计注意力生成模块和注意力融合模块,进一步提高检测准确率.分别在SBU、UCF这2个阴影检测数据集上进行验证,实验结果表明,所提方法的目视效果及客观指标皆优于此前最优方法所得结果,在2个数据集上的平均检测错误率分别降低14.4%和14.9%.

An attention-based prototypical network for forest fire smoke few-shot detection

Existing almost deep learning methods rely on a large amount of annotated data, so they are inappropriate for forest fire smoke detection with limited data. In this paper, a novel hybrid attention-based few-shot learning method, named Attention-Based Prototypical Network, is proposed for forest fire smoke detection. Specifically, feature extraction network, which consists of convolutional block attention module, could extract high-level and discriminative features and further decrease the false alarm rate resulting from suspected smoke areas. Moreover, we design a metalearning module to alleviate the overfitting issue caused by limited smoke images, and the meta-learning network enables achieving effective detection via comparing the distance between the class prototype of support images and the features of query images. A series of experiments on forest fire smoke datasets and miniImageNet dataset testify that the proposed method is superior to state-of-the-art few-shot learning approaches.

Multi-Scale Attention-Based Deep Neural Network for Brain Disease Diagnosis

Whole brain functional connectivity(FC)patterns obtained from resting-state functional magnetic resonance imaging(rs-fMRI)have been widely used in the diagnosis of brain disorders such as autism spectrum disorder(ASD).Recently,an increasing number of studies have focused on employing deep learning techniques to analyze FC patterns for brain disease classification.However,the high dimensionality of the FC features and the interpretation of deep learning results are issues that need to be addressed in the FC-based brain disease classification.In this paper,we proposed a multi-scale attention-based deep neural network(MSA-DNN)model to classify FC patterns for the ASD diagnosis.The model was implemented by adding a flexible multi-scale attention(MSA)module to the auto-encoder based backbone DNN,which can extract multi-scale features of the FC patterns and change the level of attention for different FCs by continuous learning.Our model will reinforce the weights of important FC features while suppress the unimportant FCs to ensure the sparsity of the model weights and enhance the model interpretability.We performed systematic experiments on the large multi-sites ASD dataset with both ten-fold and leaveone-site-out cross-validations.Results showed that our model outperformed classical methods in brain disease classification and revealed robust intersite prediction performance.We also localized important FC features and brain regions associated with ASD classification.Overall,our study further promotes the biomarker detection and computer-aided classification for ASD diagnosis,and the proposed MSA module is flexible and easy to implement in other classification networks.

Image-to-Image Style Transfer Based on the Ghost Module

The technology for image-to-image style transfer(a prevalent image processing task)has developed rapidly.The purpose of style transfer is to extract a texture from the source image domain and transfer it to the target image domain using a deep neural network.However,the existing methods typically have a large computational cost.To achieve efficient style transfer,we introduce a novel Ghost module into the GANILLA architecture to produce more feature maps from cheap operations.Then we utilize an attention mechanism to transform images with various styles.We optimize the original generative adversarial network(GAN)by using more efficient calculation methods for image-to-illustration translation.The experimental results show that our proposed method is similar to human vision and still maintains the quality of the image.Moreover,our proposed method overcomes the high computational cost and high computational resource consumption for style transfer.By comparing the results of subjective and objective evaluation indicators,our proposed method has shown superior performance over existing methods.

Recognition model for coated red clover seeds using YOLOv5s optimized with an attention module

The non-destructive recognition of coated seeds is crucial for advancing studies in coating theory.Currently,the recognition of coated seeds heavily relies on manual visual inspection and machine vision detection.However,these methods pose challenges such as high misclassification rates,low recognition efficiency,and elevated labor intensity.In response to the aforementioned challenges,this study leveraged deep learning techniques to develop a coated seed recognition model named YOLO-Coated Seeds Recognition(YOLO-CSR),aiming to address the challenges posed by coated seed recognition tasks.The experiment of this study mainly includes the following steps:First,a seed coating machine was set up to coat red clover seeds,resulting in three types of coated red clover seeds.Subsequently,by collecting images of the three types of coated seeds,a coated seed image dataset was further constructed.Then,the YOLOv5s was built,incorporating the Convolutional Block Attention Module(CBAM)into the model’s backbone to enhance its ability to learn features of coated seeds.Finally,the training results of YOLO-CSR were compared with those of other classical recognition models.The experimental results showed that YOLO-CSR achieved the best recognition performance on the self-built coated seed image dataset.The average precision(AP)for recognizing the three types of coated seeds reached 98.43%,97.91%,and 97.26%,with a mean average precision@0.5(mAP@0.5)of 97.87%.Compared to YOLOv5,YOLO-CSR showed a 1.18%improvement in mAP@0.5.Additionally,YOLO-CSR has a model size of only 14.9 MB,with an average recognition time(ART)of 10.1 ms and a frame per second(FPS)of 99.Experimental results prove that YOLO-CSR can accurately,efficiently,and rapidly recognize coated red clover seeds.The findings of this study provide technical support for the non-destructive recognition of spherical coated seeds.

基于深度主动学习与CBAM的细粒度菊花表型识别

针对菊花种类繁多,花型差别细微,准确标注比较困难的问题,基于深度主动学习与混合注意力机制模块(Convolutional block attention module,CBAM),提出了一种标号数据不足情况下的菊花表型智能识别方法和框架。首先,通过主动学习策略基于最优标号和次优标号法(Best vs second-best,BvSB)在未标记菊花样本中选取信息量较大的样本进行标记,并将标记后的样本放入训练样本中;其次,使用深度卷积神经网络ResNet50作为本文的主干网络训练标记样本,引入混合注意力机制模块CBAM,使模型能够更为准确地提取细粒度图像中的高层语义信息;最后,用更新后的训练样本继续训练分类模型,直到模型达到迭代次数后停止。实验结果表明,该方法在少量菊花标记样本下,精确率、召回率和F1值分别达到93.66%、93.15%和93.41%。本文方法可为标号数据不足情况下的菊花等花卉智能化识别提供技术支撑。

基于迁移学习的苹果落叶病识别与应用

为解决现有卷积神经网络苹果叶片病害识别模型泛化能力弱,模型体积较大等问题,该研究提出一种基于改进MobileNetV3苹果落叶病识别模型。以健康叶片和常见苹果落叶病为研究对象,包括斑点落叶病、灰斑病、褐斑病、锈病4种,每种病害2级,共9类特征,通过改进网络的注意力模块、全连接层及算子,结合迁移学习的训练方式,构建苹果落叶病识别模型。在扩充前后的数据集上对比不同的学习方式、学习率和注意力模块等对模型的影响,验证模型的识别性能。试验结果表明:采用迁移学习的方式,在训练50轮达曲线收敛,比全新学习的准确率增加6.74~10.79个百分点;使用引入的ET(efficient channel attention-tanh)注意力模块,网络损失曲线更加平滑,模型的参数量更少,模型体积减小了48%,提高了模型的泛化能力;在扩充数据集上,学习率为0.000 1时,结合迁移学习的训练方式,改进MobileNetV3(ET3-MobileNetV3)苹果落叶病识别模型,平均准确率能达到95.62%,模型体积6.29 MB。将模型部署到喷药设备上,可实现基于苹果叶片病害识别的变量喷施,该研究可为苹果叶片病害的检测与果园的现代化管理提供参考。