The existence of absorption and reflection of light underwater leads to problems such as color distortion and blue-green bias in underwater images.In this study,a depthwise separable convolution-based generative adver...The existence of absorption and reflection of light underwater leads to problems such as color distortion and blue-green bias in underwater images.In this study,a depthwise separable convolution-based generative adversarial network(GAN)algorithm was proposed.Taking GAN as the basic framework,it combined a depthwise separable convolution module,attention mechanism,and reconstructed convolution module to realize the enhancement of underwater degraded images.Multi-scale features were captured by the depthwise separable convolution module,and the attention mechanism was utilized to enhance attention to important features.The reconstructed convolution module further extracts and fuses local and global features.Experimental results showed that the algorithm performs well in improving the color bias and blurring of underwater images,with PSNR reaching 27.835,SSIM reaching 0.883,UIQM reaching 3.205,and UCIQE reaching 0.713.The enhanced image outperforms the comparison algorithm in both subjective and objective metrics.展开更多
针对校园网络安全入侵检测中存在的复杂网络流量模式识别困难和实时响应能力不足的问题,提出一种基于双流金字塔增强策略的深度可分离卷积网络(DSCN)模型,以优化校园网络安全入侵检测与响应系统。该模型通过双流结构融合低分辨率与高分...针对校园网络安全入侵检测中存在的复杂网络流量模式识别困难和实时响应能力不足的问题,提出一种基于双流金字塔增强策略的深度可分离卷积网络(DSCN)模型,以优化校园网络安全入侵检测与响应系统。该模型通过双流结构融合低分辨率与高分辨率路径的多尺度信息,显著提升其对复杂网络流量模式的识别能力,同时深度可分离卷积(DSC)通过分解卷积操作,有效降低模型的计算复杂度,进而提升系统的实时响应能力。将该模型在KDD Cup 1999、CICIDS 2017、CICIDS 2021和UNSW-NB15数据集上进行验证,结果表明,该模型在入侵检测率、误报率、响应时间等多项指标上表现优异,为校园网络安全防护提供了一种高效的技术解决方案。展开更多
Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware reso...Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.展开更多
目的近年来,基于深度学习的水印方法得到了广泛研究。现有方法通常对特征图的低频和高频部分同等对待,忽视了不同频率成分之间的重要差异,导致模型在处理多样化攻击时缺乏灵活性,难以同时实现水印的高保真性和强鲁棒性。为此,本文提出...目的近年来,基于深度学习的水印方法得到了广泛研究。现有方法通常对特征图的低频和高频部分同等对待,忽视了不同频率成分之间的重要差异,导致模型在处理多样化攻击时缺乏灵活性,难以同时实现水印的高保真性和强鲁棒性。为此,本文提出一种频率感知驱动的深度鲁棒图像水印技术(deep robust image watermarking driven by frequency awareness,RIWFP)。方法通过差异化机制处理低频和高频成分,提升水印性能。具体而言,低频成分通过小波卷积神经网络进行建模,利用宽感受野卷积在粗粒度层面高效学习全局结构和上下文信息;高频成分则采用深度可分离卷积和注意力机制组成的特征蒸馏块进行精炼,强化图像细节,在细粒度层面高效捕捉高频信息。此外,本文使用多频率小波损失函数,引导模型聚焦于不同频带的特征分布,进一步提升生成图像的质量。结果实验结果表明,提出的频率感知驱动的深度鲁棒图像水印技术在多个数据集上均表现出优越性能。在COCO(common objects in context)数据集上,RIWFP在随机丢弃攻击下的准确率达到91.4%;在椒盐噪声和中值滤波攻击下,RIWFP分别以100%和99.5%的准确率达到了最高水平,展现了其对高频信息的高效学习能力。在Ima⁃geNet数据集上,RIWFP在裁剪攻击下的准确率为93.4%;在JPEG压缩攻击下的准确率为99.6%,均显著优于其他对比方法。综合来看,RIWFP在COCO和ImageNet数据集上的平均准确率分别为96.7%和96.9%,均高于其他对比方法。结论本文所提方法通过频率感知的粗到细处理策略,显著增强了水印的不可见性和鲁棒性,在处理多种攻击时表现出优越性能。展开更多
高光谱图像分类是遥感领域的核心难题,其挑战主要源于高光谱图像的光谱维度高、空间分辨率低,导致分类任务中难以充分提取高光谱图像的空谱特征。针对现有的卷积神经网络(CNN)高光谱图像分类模型参数量大、计算资源消耗高、分类精度不...高光谱图像分类是遥感领域的核心难题,其挑战主要源于高光谱图像的光谱维度高、空间分辨率低,导致分类任务中难以充分提取高光谱图像的空谱特征。针对现有的卷积神经网络(CNN)高光谱图像分类模型参数量大、计算资源消耗高、分类精度不足的问题,提出了一种基于注意力机制的轻量化网络高光谱图像分类模型(AMLW-CNN)。为了增强网络的特征提取能力,将空谱特征提取模块设计为2个多尺度提取模块,并将空间特征提取模块的各卷积层通过残差结构连接,同时引入注意力机制来强化网络对有效特征的提取。另外,为了减少模型参数量,采用非对称卷积来替代三维卷积核、深度可分离卷积来替换二维卷积核。实验结果表明:AMLW-CNN的分类精度优于对比算法,计算复杂度更低,鲁棒性更强。在Indian Pines、Salinas和Pavia U 3个数据集上的总体分类精度分别达到了98.5%、99.8%、99.9%。展开更多
Channel pruning can reduce memory consumption and running time with least performance damage,and is one of the most important techniques in network compression.However,existing channel pruning methods mainly focus on ...Channel pruning can reduce memory consumption and running time with least performance damage,and is one of the most important techniques in network compression.However,existing channel pruning methods mainly focus on the pruning of standard convolutional networks,and they rely intensively on time-consuming fine-tuning to achieve the performance improvement.To this end,we present a novel efficient probability-based channel pruning method for depthwise separable convolutional networks.Our method leverages a new simple yet effective probability-based channel pruning criterion by taking the scaling and shifting factors of batch normalization layers into consideration.A novel shifting factor fusion technique is further developed to improve the performance of the pruned networks without requiring extra time-consuming fine-tuning.We apply the proposed method to five representative deep learning networks,namely MobileNetV1,MobileNetV2,ShuffleNetV1,ShuffleNetV2,and GhostNet,to demonstrate the efficiency of our pruning method.Extensive experimental results and comparisons on publicly available CIFAR10,CIFAR100,and ImageNet datasets validate the feasibility of the proposed method.展开更多
Memristor-based neuromorphic computing shows great potential for high-speed and high-throughput signal processing applications,such as electroencephalogram(EEG)signal processing.Nonetheless,the size of one-transistor ...Memristor-based neuromorphic computing shows great potential for high-speed and high-throughput signal processing applications,such as electroencephalogram(EEG)signal processing.Nonetheless,the size of one-transistor one-resistor(1T1R)memristor arrays is limited by the non-ideality of the devices,which prevents the hardware implementation of large and complex networks.In this work,we propose the depthwise separable convolution and bidirectional gate recurrent unit(DSC-BiGRU)network,a lightweight and highly robust hybrid neural network based on 1T1R arrays that enables efficient processing of EEG signals in the temporal,frequency and spatial domains by hybridizing DSC and BiGRU blocks.The network size is reduced and the network robustness is improved while ensuring the network classification accuracy.In the simulation,the measured non-idealities of the 1T1R array are brought into the network through statistical analysis.Compared with traditional convolutional networks,the network parameters are reduced by 95%and the network classification accuracy is improved by 21%at a 95%array yield rate and 5%tolerable error.This work demonstrates that lightweight and highly robust networks based on memristor arrays hold great promise for applications that rely on low consumption and high efficiency.展开更多
文摘The existence of absorption and reflection of light underwater leads to problems such as color distortion and blue-green bias in underwater images.In this study,a depthwise separable convolution-based generative adversarial network(GAN)algorithm was proposed.Taking GAN as the basic framework,it combined a depthwise separable convolution module,attention mechanism,and reconstructed convolution module to realize the enhancement of underwater degraded images.Multi-scale features were captured by the depthwise separable convolution module,and the attention mechanism was utilized to enhance attention to important features.The reconstructed convolution module further extracts and fuses local and global features.Experimental results showed that the algorithm performs well in improving the color bias and blurring of underwater images,with PSNR reaching 27.835,SSIM reaching 0.883,UIQM reaching 3.205,and UCIQE reaching 0.713.The enhanced image outperforms the comparison algorithm in both subjective and objective metrics.
文摘针对校园网络安全入侵检测中存在的复杂网络流量模式识别困难和实时响应能力不足的问题,提出一种基于双流金字塔增强策略的深度可分离卷积网络(DSCN)模型,以优化校园网络安全入侵检测与响应系统。该模型通过双流结构融合低分辨率与高分辨率路径的多尺度信息,显著提升其对复杂网络流量模式的识别能力,同时深度可分离卷积(DSC)通过分解卷积操作,有效降低模型的计算复杂度,进而提升系统的实时响应能力。将该模型在KDD Cup 1999、CICIDS 2017、CICIDS 2021和UNSW-NB15数据集上进行验证,结果表明,该模型在入侵检测率、误报率、响应时间等多项指标上表现优异,为校园网络安全防护提供了一种高效的技术解决方案。
文摘Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.
文摘目的近年来,基于深度学习的水印方法得到了广泛研究。现有方法通常对特征图的低频和高频部分同等对待,忽视了不同频率成分之间的重要差异,导致模型在处理多样化攻击时缺乏灵活性,难以同时实现水印的高保真性和强鲁棒性。为此,本文提出一种频率感知驱动的深度鲁棒图像水印技术(deep robust image watermarking driven by frequency awareness,RIWFP)。方法通过差异化机制处理低频和高频成分,提升水印性能。具体而言,低频成分通过小波卷积神经网络进行建模,利用宽感受野卷积在粗粒度层面高效学习全局结构和上下文信息;高频成分则采用深度可分离卷积和注意力机制组成的特征蒸馏块进行精炼,强化图像细节,在细粒度层面高效捕捉高频信息。此外,本文使用多频率小波损失函数,引导模型聚焦于不同频带的特征分布,进一步提升生成图像的质量。结果实验结果表明,提出的频率感知驱动的深度鲁棒图像水印技术在多个数据集上均表现出优越性能。在COCO(common objects in context)数据集上,RIWFP在随机丢弃攻击下的准确率达到91.4%;在椒盐噪声和中值滤波攻击下,RIWFP分别以100%和99.5%的准确率达到了最高水平,展现了其对高频信息的高效学习能力。在Ima⁃geNet数据集上,RIWFP在裁剪攻击下的准确率为93.4%;在JPEG压缩攻击下的准确率为99.6%,均显著优于其他对比方法。综合来看,RIWFP在COCO和ImageNet数据集上的平均准确率分别为96.7%和96.9%,均高于其他对比方法。结论本文所提方法通过频率感知的粗到细处理策略,显著增强了水印的不可见性和鲁棒性,在处理多种攻击时表现出优越性能。
文摘高光谱图像分类是遥感领域的核心难题,其挑战主要源于高光谱图像的光谱维度高、空间分辨率低,导致分类任务中难以充分提取高光谱图像的空谱特征。针对现有的卷积神经网络(CNN)高光谱图像分类模型参数量大、计算资源消耗高、分类精度不足的问题,提出了一种基于注意力机制的轻量化网络高光谱图像分类模型(AMLW-CNN)。为了增强网络的特征提取能力,将空谱特征提取模块设计为2个多尺度提取模块,并将空间特征提取模块的各卷积层通过残差结构连接,同时引入注意力机制来强化网络对有效特征的提取。另外,为了减少模型参数量,采用非对称卷积来替代三维卷积核、深度可分离卷积来替换二维卷积核。实验结果表明:AMLW-CNN的分类精度优于对比算法,计算复杂度更低,鲁棒性更强。在Indian Pines、Salinas和Pavia U 3个数据集上的总体分类精度分别达到了98.5%、99.8%、99.9%。
基金the National Natural Science Foundation of China under Grant Nos.62036010 and 62072340the Zhejiang Provincial Natural Science Foundation of China under Grant Nos.LZ21F020001 and LSZ19F020001the Open Project Program of the State Key Laboratory of CAD&CG,Zhejiang University under Grant No.A2220.
文摘Channel pruning can reduce memory consumption and running time with least performance damage,and is one of the most important techniques in network compression.However,existing channel pruning methods mainly focus on the pruning of standard convolutional networks,and they rely intensively on time-consuming fine-tuning to achieve the performance improvement.To this end,we present a novel efficient probability-based channel pruning method for depthwise separable convolutional networks.Our method leverages a new simple yet effective probability-based channel pruning criterion by taking the scaling and shifting factors of batch normalization layers into consideration.A novel shifting factor fusion technique is further developed to improve the performance of the pruned networks without requiring extra time-consuming fine-tuning.We apply the proposed method to five representative deep learning networks,namely MobileNetV1,MobileNetV2,ShuffleNetV1,ShuffleNetV2,and GhostNet,to demonstrate the efficiency of our pruning method.Extensive experimental results and comparisons on publicly available CIFAR10,CIFAR100,and ImageNet datasets validate the feasibility of the proposed method.
基金Project supported by the National Key Research and Development Program of China(Grant No.2019YFB2205102)the National Natural Science Foundation of China(Grant Nos.61974164,62074166,61804181,62004219,62004220,and 62104256).
文摘Memristor-based neuromorphic computing shows great potential for high-speed and high-throughput signal processing applications,such as electroencephalogram(EEG)signal processing.Nonetheless,the size of one-transistor one-resistor(1T1R)memristor arrays is limited by the non-ideality of the devices,which prevents the hardware implementation of large and complex networks.In this work,we propose the depthwise separable convolution and bidirectional gate recurrent unit(DSC-BiGRU)network,a lightweight and highly robust hybrid neural network based on 1T1R arrays that enables efficient processing of EEG signals in the temporal,frequency and spatial domains by hybridizing DSC and BiGRU blocks.The network size is reduced and the network robustness is improved while ensuring the network classification accuracy.In the simulation,the measured non-idealities of the 1T1R array are brought into the network through statistical analysis.Compared with traditional convolutional networks,the network parameters are reduced by 95%and the network classification accuracy is improved by 21%at a 95%array yield rate and 5%tolerable error.This work demonstrates that lightweight and highly robust networks based on memristor arrays hold great promise for applications that rely on low consumption and high efficiency.
