害虫识别是害虫防治的关键基础,由于较难获得足够的害虫种类图像,如何使用少量标记图像构造害虫分类器是一个富有挑战性的问题。现有研究多采用匹配网络框架来解决这个问题,该框架使用元学习避免重新训练深度网络,然而主干网络的特征提...害虫识别是害虫防治的关键基础,由于较难获得足够的害虫种类图像,如何使用少量标记图像构造害虫分类器是一个富有挑战性的问题。现有研究多采用匹配网络框架来解决这个问题,该框架使用元学习避免重新训练深度网络,然而主干网络的特征提取能力有限,元学习算法没有提供较好的权重初始化策略,可能导致网络出现梯度消失或者梯度爆炸的情况。为了解决这一问题,该研究提出一种基于空间注意力增强ResNeSt-101和迁移元学习算法的小样本害虫分类器。首先,通过一个空间注意力模块增强ResNeSt-101以更好地提取害虫图像特征,即在ResNeSt-101的第1阶段的最大池化层之前以及在第2~4阶段的末尾分别附加集成空间注意力模块,并通过数值仿真确定空间注意力增强模块的最佳放置位置为第1阶段的最大池化层之前。随后,通过迁移学习策略初始化网络权重,进而通过元学习进行优化。为了避免网络出现梯度消失或者梯度爆炸的情况,在元学习算法中选择归一化的温度缩放交叉熵损失函数代替三元组损失函数。最后,通过计算查询图像和支持图像深度特征之间的相似度实现害虫分类。所提出方法在自建的害虫图像数据集AD0和MIP50上使用N-类K-例准确率和每张图像处理时间(the time of per image processing,TPIP)进行评估。害虫图像数据集的构建方式如下:首先对公共害虫图像数据集IP102和D0进行清洗,以消除由于英文害虫名称导致的歧义类别;然后移除卵、幼虫和蛹阶段的害虫图像,仅保留成虫阶段的图像。考虑到人工和时间成本,从清理后的IP102害虫数据集中选择50个类别构建MIP50害虫图像数据集。随后,通过害虫的拉丁名称从互联网搜索更多的害虫图像,生成AD0害虫图像数据集。自建的MIP50数据集包括来自IP102的50个类别的16424张成虫图像,AD0包含来自D0的所有40个类别的17112张成虫图像。试验结果表明,当测试集中只有少数未知类别的害虫图像时,本文方法在AD0数据集上的5-类10-例评估准确率达到了96.37%,在MIP50数据集上达到了76.91%。当测试集中同时存在几个未知和已知类别的害虫图像时,所提方法在AD0数据集上的5-类10-例设置下的识别准确率达到了93.73%,在MIP50数据集上达到90.60%。同时,本文方法的TPIP大约为0.44 ms,满足大多数场景下的实时害虫识别要求。此外,消融试验结果表明,基于空间注意力增强ResNeSt-101网络和迁移元学习的小样本害虫分类方法在AD0、MIP50数据集上对未知类别害虫图像的5-类10-例的识别准确率分别提升了5和3个百分点以上,具有良好应用前景。但未来研究中还需进一步研究本方法中存在的问题,如通过采用更好地表征支持集样本与查询集样本之间复杂关系的度量优化本工作中用到的度量以解决增加类别数可能导致分类准确率降低的问题,以及将所提方法应用于现实农业场景进行优化改进以更好提升本文方法的实用性。展开更多
Near infrared-visible(NIR-VIS)face recognition is to match an NIR face image to a VIS image.The main challenges of NIR-VIS face recognition are the gap caused by cross-modality and the lack of sufficient paired NIR-VI...Near infrared-visible(NIR-VIS)face recognition is to match an NIR face image to a VIS image.The main challenges of NIR-VIS face recognition are the gap caused by cross-modality and the lack of sufficient paired NIR-VIS face images to train models.This paper focuses on the generation of paired NIR-VIS face images and proposes a dual variational generator based on ResNeSt(RS-DVG).RS-DVG can generate a large number of paired NIR-VIS face images from noise,and these generated NIR-VIS face images can be used as the training set together with the real NIR-VIS face images.In addition,a triplet loss function is introduced and a novel triplet selection method is proposed specifically for the training of the current face recognition model,which maximizes the inter-class distance and minimizes the intra-class distance in the input face images.The method proposed in this paper was evaluated on the datasets CASIA NIR-VIS 2.0 and BUAA-VisNir,and relatively good results were obtained.展开更多
Existing deep learning Network Intrusion Detection Systems(NIDS)struggle to simultaneously capture fine-grained,multi-scale features and long-range temporal dependencies.To address this gap,this paper introduces Trans...Existing deep learning Network Intrusion Detection Systems(NIDS)struggle to simultaneously capture fine-grained,multi-scale features and long-range temporal dependencies.To address this gap,this paper introduces TransNeSt,a hybrid architecture integrating a ResNeSt block(using split-attention for multi-scale feature representation)with a Transformer encoder(using self-attention for global temporal modeling).This integration of multi-scale and temporal attention was validated on four benchmarks:NSL-KDD,UNSW-NB15,CIC-IDS2017,and CICIOT2023.TransNeSt consistently outperformed its individual components and several state-of-the-artmodels,demonstrating significant quantitative gains.The model achieved high efficacy across all datasets,with F1-Scores of 99.04%(NSL-KDD),91.92%(UNSW-NB15),99.18%(CIC-IDS2017),and 97.85%(CICIOT2023),confirming its robustness.展开更多
基金National Science Foundation of China(62172165)Science and Technology Planning Project of Guangdong Province under Grant(2021B1212040009)+2 种基金Natural Science Foundation of Guangdong Province(2022A1515010325)Guangzhou Basic and Applied Basic Research Project(202201010742)Science and Technology Program of Guangzhou(202206010116,201902010081,107126242281)。
文摘害虫识别是害虫防治的关键基础,由于较难获得足够的害虫种类图像,如何使用少量标记图像构造害虫分类器是一个富有挑战性的问题。现有研究多采用匹配网络框架来解决这个问题,该框架使用元学习避免重新训练深度网络,然而主干网络的特征提取能力有限,元学习算法没有提供较好的权重初始化策略,可能导致网络出现梯度消失或者梯度爆炸的情况。为了解决这一问题,该研究提出一种基于空间注意力增强ResNeSt-101和迁移元学习算法的小样本害虫分类器。首先,通过一个空间注意力模块增强ResNeSt-101以更好地提取害虫图像特征,即在ResNeSt-101的第1阶段的最大池化层之前以及在第2~4阶段的末尾分别附加集成空间注意力模块,并通过数值仿真确定空间注意力增强模块的最佳放置位置为第1阶段的最大池化层之前。随后,通过迁移学习策略初始化网络权重,进而通过元学习进行优化。为了避免网络出现梯度消失或者梯度爆炸的情况,在元学习算法中选择归一化的温度缩放交叉熵损失函数代替三元组损失函数。最后,通过计算查询图像和支持图像深度特征之间的相似度实现害虫分类。所提出方法在自建的害虫图像数据集AD0和MIP50上使用N-类K-例准确率和每张图像处理时间(the time of per image processing,TPIP)进行评估。害虫图像数据集的构建方式如下:首先对公共害虫图像数据集IP102和D0进行清洗,以消除由于英文害虫名称导致的歧义类别;然后移除卵、幼虫和蛹阶段的害虫图像,仅保留成虫阶段的图像。考虑到人工和时间成本,从清理后的IP102害虫数据集中选择50个类别构建MIP50害虫图像数据集。随后,通过害虫的拉丁名称从互联网搜索更多的害虫图像,生成AD0害虫图像数据集。自建的MIP50数据集包括来自IP102的50个类别的16424张成虫图像,AD0包含来自D0的所有40个类别的17112张成虫图像。试验结果表明,当测试集中只有少数未知类别的害虫图像时,本文方法在AD0数据集上的5-类10-例评估准确率达到了96.37%,在MIP50数据集上达到了76.91%。当测试集中同时存在几个未知和已知类别的害虫图像时,所提方法在AD0数据集上的5-类10-例设置下的识别准确率达到了93.73%,在MIP50数据集上达到90.60%。同时,本文方法的TPIP大约为0.44 ms,满足大多数场景下的实时害虫识别要求。此外,消融试验结果表明,基于空间注意力增强ResNeSt-101网络和迁移元学习的小样本害虫分类方法在AD0、MIP50数据集上对未知类别害虫图像的5-类10-例的识别准确率分别提升了5和3个百分点以上,具有良好应用前景。但未来研究中还需进一步研究本方法中存在的问题,如通过采用更好地表征支持集样本与查询集样本之间复杂关系的度量优化本工作中用到的度量以解决增加类别数可能导致分类准确率降低的问题,以及将所提方法应用于现实农业场景进行优化改进以更好提升本文方法的实用性。
基金National Natural Science Foundation of China(No.62006039)National Key Research and Development Program of China(No.2019YFE0190500)。
文摘Near infrared-visible(NIR-VIS)face recognition is to match an NIR face image to a VIS image.The main challenges of NIR-VIS face recognition are the gap caused by cross-modality and the lack of sufficient paired NIR-VIS face images to train models.This paper focuses on the generation of paired NIR-VIS face images and proposes a dual variational generator based on ResNeSt(RS-DVG).RS-DVG can generate a large number of paired NIR-VIS face images from noise,and these generated NIR-VIS face images can be used as the training set together with the real NIR-VIS face images.In addition,a triplet loss function is introduced and a novel triplet selection method is proposed specifically for the training of the current face recognition model,which maximizes the inter-class distance and minimizes the intra-class distance in the input face images.The method proposed in this paper was evaluated on the datasets CASIA NIR-VIS 2.0 and BUAA-VisNir,and relatively good results were obtained.
基金supported by the Opening Foundation of Yunnan Key Laboratory of Smart City in Cyberspace Security(No.202105AG070010).
文摘Existing deep learning Network Intrusion Detection Systems(NIDS)struggle to simultaneously capture fine-grained,multi-scale features and long-range temporal dependencies.To address this gap,this paper introduces TransNeSt,a hybrid architecture integrating a ResNeSt block(using split-attention for multi-scale feature representation)with a Transformer encoder(using self-attention for global temporal modeling).This integration of multi-scale and temporal attention was validated on four benchmarks:NSL-KDD,UNSW-NB15,CIC-IDS2017,and CICIOT2023.TransNeSt consistently outperformed its individual components and several state-of-the-artmodels,demonstrating significant quantitative gains.The model achieved high efficacy across all datasets,with F1-Scores of 99.04%(NSL-KDD),91.92%(UNSW-NB15),99.18%(CIC-IDS2017),and 97.85%(CICIOT2023),confirming its robustness.
